Pilot Summary Report v.3.0 October 2006

Wyckoff/Eagle Harbor Superfund Site
Soil and Groundwater Operable Units
Thermal Remediation
Pilot Study
Summary Report
October 2006
Revision 3.0
Prepared by
US Army Corps of
Engineers
Seattle District
for
US Environmental
Protection Agency
- Region 10
This page intentionally blank
TABLE OF CONTENTS
Section Page No.

EXECUTIVE SUMMARY ...........................................................................................................................1
SECTION 1.0 INTRODUCTION AND PROJECT OBJECTIVES ........................................................5
1.1 STEAM-ENHANCED EXTRACTION TECHNOLOGY DESCRIPTION ......................................................6
1.2 PROJECT OBJECTIVES .....................................................................................................................8
1.3 PILOT STUDY PLANNING DOCUMENTS ...........................................................................................9
1.4 INITIAL CONSTRUCTION ...............................................................................................................11
1.5 PILOT STUDY DOWNSIZING ..........................................................................................................12
SECTION 2.0 SITE HISTORY AND PREVIOUS TREATMENT OPERATIONS............................15
2.1 SITE HISTORY ...............................................................................................................................15
2.2 SITE GEOLOGY AND HYDROGEOLOGY .........................................................................................16
2.3 GROUNDWATER TREATMENT SUMMARY .....................................................................................17
SECTION 3.0 PILOT TEST AREA BASELINE CONDITIONS..........................................................19
3.1 FIELD INVESTIGATION METHODS .................................................................................................20
3.2 PILOT TEST AREA GEOLOGY AND HYDROGEOLOGY ....................................................................21
3.3 CONTAMINANT DISTRIBUTION IN SOIL .........................................................................................22
3.4 MICROBIAL POPULATION EVALUATION .......................................................................................24
3.5 GROUNDWATER RESULTS ............................................................................................................25
SECTION 4.0 DESIGN AND CONSTRUCTION SUMMARY.............................................................29
4.1 DESIGN PROCESS ..........................................................................................................................29
4.2 SYSTEM DESCRIPTIONS ................................................................................................................31
4.3 MONITORING SYSTEM ..................................................................................................................43
SECTION 5.0 OPERATIONS DESCRIPTION ......................................................................................45
5.1 OPERATIONS STRATEGY ...............................................................................................................45
5.2 TEAM COMMUNICATION AND DECISION MAKING ........................................................................51
SECTION 6.0 SUMMARY OF RESULTS ..............................................................................................53
6.1 OPERATIONAL RESULTS ...............................................................................................................53
6.2 SUBSURFACE TEMPERATURES ACHIEVED ....................................................................................59
6.3 COMPARISON OF PROCESS DATA TO SUBSURFACE HEATING OBSERVATIONS ..............................65
6.4 CHEMICAL MONITORING RESULTS ...............................................................................................67
6.5 VAPOR TREATMENT SYSTEM DATA SUMMARY............................................................................71
6.6 WATER TREATMENT SYSTEM DATA SUMMARY ...........................................................................72
6.7 PILOT STUDY MASS REMOVAL ESTIMATES ..................................................................................75
SECTION 7.0 COST SUMMARY ............................................................................................................77
SECTION 8.0 CONCLUSIONS................................................................................................................79
8.1 PERFORMANCE ASSESSMENT .......................................................................................................79
8.2 COMMUNITY AND ENVIRONMENTAL IMPACTS .............................................................................81
8.3 PROCESS OPERATIONS AND DESIGN .............................................................................................82
SECTION 9.0 LESSONS LEARNED.......................................................................................................87
9.1 SITE AND CONTAMINANT COMPOSITION CHARACTERIZATION .....................................................87
9.2 DESIGN .........................................................................................................................................88
9.3 CONSTRUCTION AND EQUIPMENT.................................................................................................94
9.4 OPERATIONS .................................................................................................................................95
9.5 MONITORING PROGRAM ...............................................................................................................96
9.6 TEAM STRUCTURE AND COMMUNICATION ...................................................................................98
9.7 SCHEDULE AND BUDGET ................................................................................................................100
SECTION 10.0 REFERENCES ..............................................................................................................101
Thermal Remediation Pilot Study Summary Report i October 2006, Revision 3.0
Wyckoff/Eagle Harbor Superfund Site Soil and Groundwater Operable Units
TABLES
TABLE 3.2-1 PHYSICAL CHARACTERISTICS AND THERMODYNAMIC PROPERTIES OF SOIL SAMPLES105
TABLE 3.5.1 EXTRACTION WELL BASELINE CHARACTERIZATION SAMPLING RESULTS .................106
TABLE 3.5-2 UPPER AQUIFER BASELINE CHARACTERIZATION GROUNDWATER SAMPLING RESULTS107
TABLE 3.5-3 LOWER AQUIFER BASELINE CHARACTERIZATION GROUNDWATER SAMPLING RESULTS108
TABLE 4.1-1 DESIGN CRITERIA SUMMARY .....................................................................................109
TABLE 4.2-1 LIQUID CONVEYANCE AND TREATMENT SYSTEM ......................................................111
TABLE 4.2-2 STEAM GENERATION AND CONVEYANCE SYSTEM .....................................................114
TABLE 4.2-3 VAPOR TREATMENT SYSTEM .....................................................................................116
TABLE 4.2-4 MASS-AVERAGED COMPONENT PROPERTIES AT 25 C ..............................................120
TABLE 4.3-1 PILOT AREA WELL FIELD AUTOMATED MONITORING SUMMARY ..............................121
TABLE 4.3-2 INJECTION AND EXTRACTION SYSTEM MONITORING SUMMARY ...............................122
TABLE 4.3-3 TREATMENT SYSTEM MONITORING SUMMARY ..........................................................124
TABLE 4.3-4 WORKER SAFETY AND PERIMETER MONITORING SUMMARY ....................................127
TABLE 5.2-1 SUPPORTING GROUPS AND CONTRACTORS ................................................................128
TABLE 5.2-2 PROJECT WEB PAGE CONTENTS .................................................................................129
TABLE 6.1-1 AVERAGE STEAM INJECTION RATES FOR EACH INJECTION WELL DURING THE TWO
MAJOR INJECTION PERIODS ........................................................................................130
TABLE 6.6-1 GROUNDWATER TREATMENT PLANT DATA SUMMARY ..............................................131
FIGURES
Figure 1.0-1 Site Vicinity Map
Figure 2.0-1 Site Layout
Figure 2.2-1 Site Geologic Profile
Figure 3.2-1 Location Map Showing Profiles Through Pilot Study Area
Figure 3.2-2 Profile A-A Through Pilot Study Area
Figure 3.2-3 Profile B-B Through Pilot Study Area
Figure 3.2-4 Profile C-C Through Pilot Study Area
Figure 3.5-1 Monitoring Well Network
Figure 4.0-1 Construction Timeline
Figure 4.2-1 Well Layout
Figure 4.2-2 Injection Well, Extraction Well, And Instrumentation String Construction
Details
Figure 4.2-3 Boiler Plant Layout
Figure 4.2-4 Boiler Plant Layout With Plan Modifications
Figure 4.2-5 Steam And Vapor Process Flow Diagram
Figure 4.2-6 Above Ground Piping Layout
Figure 4.2-7 Vapor Collection Piping Layout
Thermal Remediation Pilot Study Summary Report ii October 2006, Revision 3.0
Figure 4.2-8 Water Treatment Plan Layout
Figure 4.2-9 Water Treatment Plant Process Flow Diagram
Figure 5.1-1 Operations Timeline
Figure 6.1-1 Total Steam Production Rates
Figure 6.1-2 Steam Injection Rates For All 16 Injection Wells Combined
Figure 6.1-3 Steam Injection Rates For The 16 Injection Wells During 2002
Figure 6.1-4 Steam Injection Rates For The 16 Injection Wells During 2003
Figure 6.1-5 Cumulative Steam Injection Volumes For The 16 Injection Wells
Figure 6.1-6 Liquid Extraction Rates From Each Of The Seven Extraction Wells
Figure 6.1-7 Cumulative Water Volumes Extracted From Each Extraction Well
Figure 6.1-8 Total Liquid Extraction Rates From The Seven Extraction Wells
Figure6.1-9 Water Levels Observed In Extraction Wells During Operations
Figure 6.1-10 Water Levels Observed In Monitoring Wells During Operations
Figure 6.1-11 Water Flow Rates For The Pilot Test Area
Figure 6.1-12 Cumulative Water Balance For The Pilot Test Area
Figure 6.1-13 Enthalpy Fluxes For The Pilot Test Area During Operations
Figure 6.1-14 Energy Balance For The Pilot Test Area During Operations
Figure 6.1-15 Cumulative Energy Losses
Figure 6.2-1 Temperatures Of Extracted Liquid From Six Of The Extraction Wells
Figure 6.2-2 Temperature String Layout With Interpolated Temperature Profile Lines
Figure 6.2-3 Comparison Of Temperature Profiles For T-7
Figure 6.2-4 Interpolated Temperature Maps For Slices A-E
Figure 6.2-5 Interpolated Temperature Profiles Along Sections C And F
Figure 6.2-6 Temperature Profiles And Water Levels For Extraction Wells E4 And E6
Figure 6.2-7 Collector Layer Temperatures On December 10 And 12, 2002
Figure 6.2-8 Seatac Airport Precipitation And Collector Layer Temperatures At T-52
Figure 6.2-9 Temperature Sensor Locations In The Vapor Collector Layer
Figure 6.2-10 Temperature Profiles Along The Inside Of The Sheetpile Wall Oct. Dec.
2002
Figure 6.2-11 Temperature Profiles Along The Inside Of The Sheetpile Wall Jan. May
2003
Figure 6.2-12 Temperature Profiles For The E-4 Array For Oct. Dec. 2002
Figure 6.2-13 Temperature Profiles For The E-4 Array For Jan. May 2003
Figure 6.3-1 Calculated Average Temperatures
Figure 6.3-2 Comparison Of Measured And Calculated Temperature Data
Figure 6.3-3 Cumulative Volume Of Water Extracted From Each Extraction Well
Figure 6.3-4 Average Steam Injection Rates From October 10 To December 14, 2002
Figure 6.4-1 Extracted Liquid Pah Concentrations November 2002 To January 2003
Figure 6.4-2 Extracted Liquid Pah Concentrations January To May 2003
Figure 6.4-3 Extracted Liquid Pcp Concentrations November 2002 To January 2003
Figure 6.4-4 Extracted Liquid Pcp Concentrations January To May 2003
Figure 6.4-5 Total Organic Carbon Data Summary
Figure 6.6-1 Total Pah And Pcp In Treatment Plant Influent Prior To Pilot Study
Figure 6.6-2 Total Pah Concentrations Into And Out Of The Daft
Figure 6.6-3 Naphthalene Concentrations Into And Out Of The Daft
Figure 6.6-4 Pcp Concentrations Into And Out Of The Daft
Thermal Remediation Pilot Study Summary Report iii October 2006, Revision 3.0
Figure 6.6-5 Napl Removal Per Month
Figure 6.6-6 Total Pah Concentrations Into And Out Of The Aeration Basin
Figure 6.6-7 Naphthalene Concentrations Into And Out Of The Aeration Basin
Figure 6.6-8 Pcp Concentrations Into And Out Of The Aeration Basin
Figure 6.6-9 Aeration Basin Total Pah And Pcp Removal Effectiveness
Figure 6.6-10 Dissolved Oxygen Concentration In Aeration Basin
Figure 6.6-11 Total Pah Concentration Into And Out Of The Carbon Treatment System
Figure 6.6-12 Pcp Concentration Into And Out Of The Carbon Treatment System
Figure 6.6-13 Carbon Treatment System Total Pah And Pcp Removal Effectiveness
Figure 6.6-14 Total Pah And Pcp Concentrations In Gwtp Effluent
Figure 6.6-15 Ph In Gwtp Effluent
Figure 6.6-16 Temperature And Dissolved Oxygen In Gwtp Effluent
APPENDICES
Appendix A Site Cleanup Levels
Appendix B Pilot Test Area Selection Memorandum
Appendix C Well E4 Pumping Test Results
Appendix D Microcosm Study
Appendix E Instrumentation Design and Construction
Appendix F Operations Logs (CD only)
Appendix G PAH Precipitation and Encrustation Evaluation
Appendix H Calculation Methods for Process Streams
Appendix I Pilot Study Data Summary (CD only)
Appendix J Installation Report for Well 02CD-MW01
Thermal Remediation Pilot Study Summary Report iv October 2006, Revision 3.0
ACRONYMS LIST
AC Alternating current
acfm Actual cubic feet per minute
ADAS Automated data acquisition system
atm Atmospheres
bgs Below ground surface
BTU British thermal units
BTU/gal British thermal units per gallon
C Degrees Celsius
CD Conceptual Design
cfm Cubic feet per minute
CLP Contract Laboratory Program for USEPA
DA Design Analysis
DAA Design Analysis Amendment
DAF Dissolved air flotation, or diffuse air flotation
dBA Decibels
DNAPL Dense Non-Aqueous Phase Liquid
DTS Distributed temperature sensor
EPDM Ethylene propylene diene monomer
F Degrees Fahrenheit
FASP EPA Region 10 Field Analytical Support Program
FPA Former Process Area
FS Feasibility Study
ft Feet
GAC Granulated activated carbon
gal Gallon
GFI Ground fault interrupter
GMS Groundwater Modeling System
gpd Gallons per day
gpm Gallons per minute
GWTP Groundwater treatment plant
HDPE High density polyethylene
hp Horsepower
HX Heat exchanger
ID Inside diameter
in Inches
ISTD In-Situ Thermal Desorption
Thermal Remediation Pilot Study Summary Report v October 2006, Revision 3.0
ITTAP In-situ Thermal Technologies Advisory Panel
kg Kilograms
kg/d Kilograms per day
kg/s Kilograms per second
kPa Kilopascal
lb Pounds
lb/hr Pounds per hour
l.e. Liquid equivalent
LNAPL Light Non-Aqueous Phase Liquid
LRVP Liquid Ring Vacuum Pump
m Meters
mA Milliamps
MarVac Marine Vacuum, Inc.
MBTU Million British Thermal Units
mg/kg Milligram per kilogram
mg/L Milligram per liter
MLLW Mean lower low water
mm Millimeters
MTCA Washington State Model Toxics Control Act
g/L Micrograms per liter
ng/kg nanograms per kilogram
NAPL Non-aqueous phase liquids
NPDES National Discharge Elimination System (permit)
NPL National Priority List
NTU Nephelometric Turbidity Units
OCDD Octachlorodibenzodioxin
OD Outside diameter
O&M Operation and Maintenance
OU Operable Unit
PAH Polynuclear aromatic hydrocarbons
PCP Pentachlorophenol
PETG Polyethylene terephthalate glycol
PLFA Phospholipid fatty acid
ppm Parts per million
psi Pounds per square inch
psig Pounds per square inch (gauge)
PVC Polyvinylchloride
RAMP Remedial Action Management Plan
ROD Record of Decision
Thermal Remediation Pilot Study Summary Report vi October 2006, Revision 3.0
RTU Remote Telemetry Unit
scfm cubic feet per minute at Standard Temperature and Pressure
(298.15 K and 101.33 kPa, unless otherwise noted)
SEE Steam-Enhanced Extraction
sf Square foot
TOC Total Organic Carbon
TPH Total Petroleum Hydrocarbons
USACE United States Army Corps of Engineers
USEPA United States Environmental Protection Agency
WAC Washington Administrative Code
3
yd Cubic yards
2,3,7,8-TCDD 2,3,7,8-tetrachlorodibenzo-p-dioxin (or simply dioxin)
3D Three dimensional
Note regarding units: No attempt was made to provide consistent units throughout the
report. Instruments in different parts of the system may have provided data in
inconsistent units (for example, Fahrenheit for liquid temperature vs. Celsius for
subsurface temperature). Data are reported in the units provided by the instruments. In
other instances of inconsistent units, units may have been determined by the discipline
that was responsible for that section of the report, and those units may not have been
consistent with units used by other disciplines.
Thermal Remediation Pilot Study Summary Report vii October 2006, Revision 3.0
Thermal Remediation Pilot Study Summary Report viii October 2006, Revision 3.0
EXECUTIVE SUMMARY
The U.S. Army Corps of Engineers (USACE) is providing remedial design and
remedial action services to the U.S. Environmental Protection Agency (USEPA) Region
10 for the Wyckoff/Eagle Harbor Superfund Site (the Site), located on Bainbridge Island,
Washington. The USACE designed a pilot study that was intended to determine the
effectiveness of innovative thermal remediation to enhance the recovery of non-aqueous
phase liquids (NAPL) from the site. This was the first time this thermal technology was
attempted in this type of environmental setting: the Pacific Northwest with 50F water on
three sides of the property. This work was performed to meet the requirements of the
Record of Decision (ROD) for the Soil and Groundwater Operable Units (OU) (USEPA
2000). The pilot study was designed to meet objectives divided into two broad
categories:
To assess the likelihood that full-scale thermal remediation will achieve
the cleanup goals for the Site
To provide information for implementation of the potential full-scale
thermal remediation
Initially, the project team was assisted by the In-Situ Thermal Technologies
Advisory Panel (ITTAP) to USEPA. As the project moved forward to construction and
operations, cost-cutting measures resulted in downsizing of the project team and a
gradual loss of input from industry experts.
Bids for a combined construction and operations contract were considerably

higher than the government estimate, due to risk factors added by bidders and to cost
omissions from the estimate. The pilot study was then revised with less-robust extraction
and treatment components, and the existing wastewater treatment plant was incorporated
into the test with very few modifications. Due to cost downsizing, some upgradient
injection and extraction wells were eliminated, reducing the thermal treatment zone to
about half of the originally identified pilot-test area and leaving NAPL-contaminated soil
upgradient of the revised pilot test area.
The funding schedule required the project to be split into three phases:
Infrastructure Construction (earthwork, vapor cap, boiler building and wells),
Mechanical/Process Construction (boiler, pumps, piping, and treatment), and Pilot
Operations, each with a separate contract.
Thermal Remediation Pilot Study Summary Report 1 October 2006, Revision 3.0
The original pilot test area was proposed to cover one acre, with a total volume of
35,300 cubic yards, containing an estimated 40,000 gallons of creosote NAPL. The
actual thermal treatment zone consisted of 16,400 cubic yards. The pilot test area is
separated from the remainder of the site by an open-ended sheet-pile wall.
The pilot study proceeded for six and a half months, from October 2002 to April
2003. Operations were restricted by equipment problems, and the pilot study was
terminated. The average steam injection rate for the operating period was 12 percent of
the design rate, and the average pumping rate was 24 percent of the design rate. The total
time of operation of the vapor-extraction system was about 1 month, operating
continuously no more than 3 days. Because roughly one pore volume (liquid equivalent)
of steam was injected, compared to design and laboratory bench-scale test
recommendations of 2-5 pore volumes, the pilot study should not be viewed as a
definitive application of thermal remediation technology, however, there are some
conclusions that can be made based on the data collected.
Very little steam penetration was achieved due to the low injection and extraction
rates. Steam flow was primarily vertical to the groundwater surface and into the vadose
zone. In March 2003, the average temperature of the vadose zone (excluding the vapor
cap) reached a maximum of 98 C while the saturated zone reached a maximum average
of 72 C.
Repeated technical issues with the pilot study were diverse: issues with the
extraction systems (liquid and vapor), conveyance systems (liquid and vapor), and
treatment plant occurred. The most serious problems were: aspiration of liquid by the
vapor-vacuum pumps, overloading of the biological water-treatment system, deterioration
of gaskets due to materials incompatible with site contaminants, and clogging of pipes
and treatment facilities by precipitating polycyclic aromatic hydrocarbons (PAH)
(especially naphthalene).
There were equipment constraints that limited operations of the system.

Constraints included capacity of the treatment plant, inability to treat the vapor stream
due to equipment failure, installation of only two liquid-ring vacuum pumps instead of
three, the installation of a plate-and-frame heat exchanger for the vapor line instead of a
shell-and-tube heat exchanger, and not enough capacity in the vapor condensate
receivers. These system constraints caused operators to limit steam injection and liquid
extraction rates. The vapor system was offline for months while a different heat
exchanger was designed, manufactured and installed. During this period, operations
consisted of limited injection and extraction that promoted clogging of the liquid
extraction system pipelines and equipment with crystallized PAHs. As a consequence,
the system was shut down.
The equivalent of approximately 2,940 gallons of NAPL was recovered during the
pilot study: 340 gallons as NAPL and 9,800 kg (equivalent to 2,600 gallons) in the
dissolved phase. During the same time period, the equivalent of 1,455 gallons of NAPL
was extracted by the pump-and-treat system in the remaining seven acres of the Former
Process Area (FPA): 1,295 gallons as NAPL and 606 kg (equivalent to 160 gallons) in
the dissolved phase. Prior to the pilot, the average amount of NAPL extracted per month
was approximately 320 gallons with an average of approximately 24 gallons per month in
the dissolved phase. Though the amount of NAPL removed did not show a marked
increase during the pilot study, the amount of contaminants removed in the dissolved
phase increased dramatically.
Available data suggest that significant contamination was removed in the vapor
phase, however some of the planned monitoring instruments were not installed when the
vapor system was in operation, and accurate vapor-flow measurements could not be
made. Additionally, due to the early termination of the study, it was not possible to
assess in-situ degradation via biologic and abiotic processes. The mass removal that
occurred during the pilot study was minor compared with what could have been achieved
with a fully functional steam-enhanced extraction (SEE) and treatment system.
Lessons learned that would apply to any technically complex project include: the
need to inform senior management of issues associated with funding constraints that
impact system capacities, the need to assess risks associated with design features
inconsistent with industry practice, and to have contingency plans in order to manage
these risks effectively.
This page intentionally blank.
SECTION 1.0 INTRODUCTION AND PROJECT
OBJECTIVES
The U.S. Army Corps of Engineers is providing remedial design and remedial
action services to the U.S. Environmental Protection Agency Region 10 for the
Wyckoff/Eagle Harbor Superfund Site, located on Bainbridge Island, Washington (Figure
1.0-1). The USACE designed a pilot study that was intended to determine the
effectiveness of innovative thermal remediation to enhance the recovery of NAPL from
the site. This work was performed to meet the requirements of the Record of Decision
for the Soil and Groundwater Operable Units (OU) (USEPA 2000).
In 1998, USEPA was asked to evaluate in-situ thermal technologies, including

steam injection, as potential remedies for clean up of soil and groundwater contamination
at the Site. In order to do this, a panel of experts on thermal remediation was assembled
by USEPA to provide technical assistance on the evaluation of the application of in-situ
thermal technologies for this site specifically, and to provide input on the design and
operation of the system at sites where USEPA makes a decision to proceed with in-situ
thermal technologies. This panel became known as the In-Situ Thermal Technology
Technical Assistance Panel (ITTAP), and was comprised of the principal engineers and
researchers from both academia and industry who had pioneered the use of thermal
remediation in the United States. USEPA met with the panel several times during 1999
and 2000. Based on the success that had been achieved using thermal remediation at
other sites, the ITTAP panel supported the use of thermal remediation at Wyckoff.
The USEPA identified in-situ thermal technology (steam injection) as a potential

remedy for clean up of soil and groundwater contamination at the Site. The purpose of
the pilot study falls into two broad categories:
1. To assess the likelihood that full-scale thermal remediation will achieve

the cleanup goals for the Site;
2. To provide information for implementation of the potential full-scale

thermal remediation.
The pilot study design was based on meeting these objectives. In the interim, a
sheet pile barrier was constructed to prevent movement of contaminants beyond the Site
boundaries. If the pilot study is not successful in demonstrating the effectiveness of
thermal treatment and/or the full scale system is not implemented for any number of
reasons, a "containment" remedy is now partially in place. The final containment remedy
would include a fully enclosed sheet pile wall (incorporating the existing outer sheet pile
wall) that surrounds contaminated soil and groundwater in the Former Process Area
(FPA), a replacement groundwater pump-and-treat system to maintain the water level
within the sheet pile wall, a soil cap to isolate surface soils in the FPA, and shoreline
improvements.
The purpose of this report is to document the design, construction, and operation
activities for the pilot study performed by the USACE from 2000 through 2003, with
particular attention to shortcomings of the project that contributed to its early termination.
In this regard, the document may provide a roadmap for future projects that will allow
them to avoid pitfalls associated with constraints imposed by complex technologies,
budget, and schedule.
The pilot study began operations in October 2002. Due to complications with
several systems caused by high concentrations of contaminants in the waste stream,
serious limitations to the existing waste-water treatment plant, and multiple equipment
failures, steam injection was stopped in April 2003. The USEPA and USACE are
currently evaluating various options before potentially continuing with the pilot study.
The report is divided into the following sections:

1.0 Introduction
2.0 Site History and Previous Treatment Operations
3.0 Pilot Test Area Baseline Conditions
4.0 Design and Construction Summary
5.0 Operations Description
6.0 Summary of Results
7.0 Cost Summary
8.0 Conclusions
9.0 Lessons Learned
1.1 STEAM-ENHANCED EXTRACTION TECHNOLOGY DESCRIPTION
Steam-enhanced extraction has its roots in the petroleum industry, where steam
injection is used to enhance oil recovery from reservoirs. Beginning in the late 1980s,
steam injection has been adapted for the recovery of volatile and semivolatile
contaminants from unconsolidated media (Hunt et. al, 1988; Udell and Stewart, 1989;
Itamura and Udell, 1993). This research has shown that there are three main mechanisms
in steam injection that contribute to contaminant recovery, as summarized by Udell
(1996) and Davis (1997):
Physical displacement of NAPL as the steam migrates from injection to extraction
wells, assisted by the reduction in NAPL viscosity during heating, leading to
greater mobility.
Increased vaporization and extraction in the vapor phase.
Increased solubilization of contaminants with subsequent removal in the dissolved

state by the groundwater extraction system.
In addition, in-situ destruction due to either chemical or biological mechanisms

may contribute to the remediation process during steam injection (Lief et al. 1998).
Initially, as steam is injected, the steam condenses, transferring the latent heat of
condensation from the steam to the well bore, groundwater, and formation matrix
immediately surrounding the injection zone of the well. As steam injection continues, the
hot water (condensate) moves into the formation, pushing the cold (ambient temperature)
formation water in advance of this front. When the soil at the steam injection point has
absorbed sufficient heat, steam will begin to enter the formation, pushing the cold
formation water and hot condensate water ahead of it. The first fluid to come in contact
with the compounds of concern is the cold water bank, which flushes mobile compounds
in displaced groundwater. The hot water that follows reduces the viscosity of the NAPL,
displacing it by hydrodynamic forces, potentially reducing residual saturation, and
enhancing biodegradation. When the steam front reaches the contaminated zone,
additional contaminant removal occurs through volatilization, evaporation, and steam
distillation of volatile and semi-volatile compounds.
In some cases where the NAPL saturation is high, a bank of NAPL or highly-
concentrated contaminants can be formed in front of the steam zone. A NAPL bank
forms when oil-phase fluids are displaced and driven ahead of the steam zone, or when
compounds volatilized from the steam zone are condensed at the steam front. When the
pollutants represent a mixture of volatile and semi-volatile compounds, as in creosote, the
compounds with lower boiling points will vaporize first, followed by the higher-boiling
point compounds. As a result, there will be a corresponding differentiation of the
constituents in the bank of condensed compounds moving through the formation in
advance of the steam front.
A physical characteristic of steam injection processes is steam override due to

gravitational forces. Steam override is caused by the density difference between steam
and water, and will tend to make the displacement process less effective during steam
injection remediation. The degree of steam override increases as: 1) the difference in
density between the liquid and vapor phases increases, 2) the permeability of the
formation decreases, 3) the viscosity of the liquid phase increases, and 4) the injection
rate decreases. Steam override cannot be eliminated, but by increasing the steam
injection rate, the difference between the vapor and liquid viscous forces is reduced and
thus the amount of override can be decreased (Davis, 1998).
The pressure of the injected steam increases formation pore pressures, which may
also serve to inhibit volatilization. Manipulation of subsurface pressures by reducing or
halting steam injection while continuing aggressive vapor and groundwater extraction
after steam breakthrough to the extraction wells, called pressure cycling, can enhance
the recovery of volatile compounds by creating a thermodynamically unstable condition,
in which vaporization is enhanced by the boiling of pore fluids. For sites contaminated
with semi-volatile and non-volatile mixtures of chemicals such as creosote, pressure
cycling is used both to enhance vaporization of the lighter fractions of contaminants, and
to induce mixing of injected steam and air with the contaminated groundwater. This may
stimulate degradation reactions that take place under aerobic conditions at elevated
temperatures. These reactions may be thermodynamically driven (sometimes called
hydrous pyrolosis/oxidation) or biologically mediated.
1.2 PROJECT OBJECTIVES
Nine primary objectives were developed to meet remedial action objectives in the
2000 ROD for the Soil and Groundwater Operable Units. These nine objectives can be
divided into three broad categories: performance assessment, potential impacts of full-
scale thermal treatment on the environment and surrounding community, and process
monitoring. The specific project objectives are presented below:
Performance Assessment Objectives
Demonstrate that thermal remediation technologies will remove substantially

all mobile NAPL from the pilot test treatment area.
Demonstrate that the post-thermal treatment concentrations of NAPL
constituents dissolved in groundwater that move from the site to Eagle Harbor
and Puget Sound will not exceed marine water quality criteria, surface water
quality, and sediment standards at the mud line.
Demonstrate that surface soil (0 to 15 ft) concentrations within the pilot test
area attain Washington State Model Toxics Control Act (MTCA) Method B
cleanup levels.
Cleanup levels for the site are included as Appendix A.
Community and Environmental Impacts of Full-Scale Thermal Remediation

Objectives
Determine the potential impacts (noise, air emissions, lower aquifer and
odors) of full-scale thermal treatment to the surrounding community.
Evaluate the possible adverse effects that full scale thermal treatment may
have to Eagle Harbor and Puget Sound near shore marine habitats.
Process Objectives
Evaluate operational approaches to thermal remediation that may impact the

removal of NAPL, such as steam movement and recovery of NAPL from the
aquitard.
Evaluate treatment plant performance during the pilot test to allow
optimization of operations and monitoring mass balance of contaminant
removal.
Evaluate microbial populations before and after thermal treatment to assist in
determining long-term contaminant destruction.
Evaluate contaminant oxidation rates during thermal treatment to assist in
mass balance calculations.
Additionally, an objective to evaluate the effectiveness of biological treatment

and break down of primary contaminants in extracted liquids produced during a thermal
remediation system was added to the original list of objectives established in the ROD.
The extent to which each of these objectives was met during this pilot test is discussed in
Section 8.0.
1.3 PILOT STUDY PLANNING DOCUMENTS
Several documents were prepared in support of the pilot test. These documents
contain much of the detail on the technical planning process, design, and operations.
Interim Record of Decision (Interim ROD). Groundwater Operable Unit.

USEPA. September 30, 1994. Documents the selection of containment of
NAPL and contaminated groundwater as the remedy for the site.
Feasibility Study (FS). Soil and Groundwater Operable Units. CH2M

Hill. October 17, 1997. Includes a comparison of 23 alternatives for the
Soil OU. Concludes that regardless of the alternative selected for the Soil
OU, there would be significant remaining contamination associated with
the Groundwater OU.
Focused Feasibility Study for Thermal Remediation Technologies (FFS).

Soil and Groundwater Operable Units. CH2M Hill. June 1998. Includes
a comparison of numerous thermal remediation alternatives to
containment. Concludes that thermal technologies may be preferred over
the containment option.
Conceptual Design (CD). Soil and Groundwater Operable Units.

USACE. September 9, 1999. Describes the full-scale thermal
remediation conceptual design for the soil and groundwater operable units
as well as many of the design considerations made in developing the
conceptual design.
Record of Decision (ROD). Soil and Groundwater Operable Units.

USEPA. February 2000. Documents the selection of thermal remediation
as the remedy for the site to be implemented in two phases: Phase 1 - pilot
test; Phase 2 full scale. As stated in the ROD, implementation of full
scale thermal remediation was dependent on the pilot study reasonably
attaining the performance expectations. Hydraulic containment was
selected as the contingency remedy.
Thermal Remediation Pilot Test Area Selection Memorandum.

Groundwater and Soil Operable Units. USACE. April 21, 2000.
Proposes three areas for consideration for the pilot test and recommends
Area C.
Comprehensive Report. Wyckoff NAPL Field Exploration. Soil and

Groundwater Operable Units. Wyckoff/Eagle Harbor Superfund Site.
USACE. May 2000. Presents results of the soil sampling and analysis
conducted in the upland area of the site.
Thermal Remediation Shoreline Model Report. Groundwater and Soil

Operable Units. USACE. May 22, 2000. Using numerical modeling,
evaluates effects of thermal remediation on temperatures in intertidal
sediments and explores engineering controls.
Final Conceptual (10%) Design (Pilot Study CD). Thermal Remediation
Pilot Study. USACE. August 9, 2000. Documents the initial pilot test
design effort, including consideration of design alternatives.
Final Design Analysis (DA). Thermal Remediation Pilot Study. USACE.

August 31, 2001. Documents the pilot study design effort, including
consideration of design alternatives, provides justification for design
decisions, and presents information on the project schedule.
Steam Injection Treatability Study. Final Report. USEPA. July 11, 2002.
Presents results of the steam treatability study conducted at the Robert S.
Kerr Environmental Research Center in Ada, OK.
Final Design Analysis Amendment (DAA). Thermal Remediation Pilot

Study. USACE. August 6, 2002. Documents changes to the Final Design
Analysis for the Thermal Remediation Pilot Study, dated August 31, 2001.
Preliminary Review Draft Baseline Characterization Report. Thermal

Remediation Pilot Study. USACE. November 1, 2002. Describes the
results from the baseline characterization work performed during
installation of the extraction, injection, and instrument string wells for the
Thermal Remediation Pilot Study.
Draft Remedial Action Management Plan (RAMP). Thermal Remediation

Pilot Study. USACE. November 22, 2002. Management plan for
integrated site operations of the Thermal Remediation Pilot Study and the
groundwater treatment plant and extraction system.
1.4 INITIAL CONSTRUCTION
Following the signing of the ROD in February 2000, a contract was awarded to
Bay West, Inc. (Bay West) in August 2000 for installation of two sheet pile walls. A
1,870-foot-long perimeter vertical barrier wall was installed as part of the thermal
remediation remedy to separate contaminants in the FPA from the marine environment.
Additionally, a 536-foot-long vertical barrier wall was installed in the southern portion of
the upland to isolate the 1-acre pilot test area from the rest of the upland portion of the
site. Sheet pile installations were completed by February 2001.
The final design of the Thermal Remediation Pilot Study was completed in
August 2001 (USACE Design Analysis report). Phase I of this work was performed
under the Site Infrastructure Contract between Marine Vacuum, Inc. (MarVac) and
USACE which was awarded in July 2001. The Phase I work was completed between
August 2001 and March 2002 and primarily consisted of construction of the boiler
building and the subsurface (belowground) portion of the pilot studys elements,
including the installation of the vapor cap, the water supply well, and the steam injection
and extraction wells. Additional contract specifications for the pilot study were
developed by the USACE in the spring of 2001 that included pre-thermal operation of the
existing groundwater treatment plant (GWTP) and extraction system, minor
modifications to the existing GWTP, installation of the thermal components of the
existing boiler building and pilot test area, and a start-up phase for all new equipment.
This work was to be followed by an operation and maintenance (O&M) phase for steam
injection and liquid/vapor extraction and an operational post-thermal period for ongoing
O&M of the existing groundwater treatment plant and extraction system.
The solicitation to contractors was issued in the summer of 2001, however, an

award was not made and the solicitation was cancelled because the proposals received
greatly exceeded the government cost estimate.
1.5 PILOT STUDY DOWNSIZING
Due to the disparity between contractor proposals and the government cost
estimate, USEPA requested the design team redesign the pilot study contract by
downsizing and down scoping the project to meet the project budget while maintaining
the overall goals and objectives of the project and with the government assuming the
majority of the risk. The procurement strategy was also re-evaluated due to schedule and
budget constraints, and it was determined that three contracts, rather than one, would best
serve the needs of the project and help to limit the governments assumption of risk.
Procurement #1 was for interim operations and maintenance of the existing treatment
plant and extraction system and was awarded to SCS Engineers, Inc. (SCS) on January 1,
2002. Procurement #2 was for the construction of the Steam Injection Pilot System and
was awarded to Pease Construction (Pease) on January 23, 2002. Procurement #3 was
for O&M of the steam injection pilot study system, the existing well-field outside of the
pilot test area, and the GWTP and was awarded to SCS on September 20, 2002. The
scope of these changes was presented in the Design Analysis Amendment. Changes to
the design, starting from the Conceptual Design, Design Analysis, Design Analysis
Amendment, and through construction, are discussed in Section 4.0. It should be noted
that because the sheet pile wall and several injection and extraction wells within the pilot
test area had already been installed prior to the request to downsize the overall pilot
study, the technical teams ability to select appropriate design elements to adjust was
limited. However, a more aggressive approach to the downsizing may have allowed a
better fit to the reduced budget.
SECTION 2.0 SITE HISTORY AND
PREVIOUS TREATMENT OPERATIONS
The Wyckoff Soil and Groundwater OUs occupy a relatively flat lowland and
intertidal area bounded by a densely vegetated bluff on the south. The lowland area has
an average elevation of approximately 16 feet MLLW while the hillside area rises to
elevations above 200 feet. The north and west portions of the site are bounded by Eagle
Harbor, and Puget Sound abuts the eastern margin of the site (Figure 2.0-1). The entire
Wyckoff property occupies approximately 57 acres (about 18 of which encompass the
Soil OU), including a spit that is approximately 8 acres in size with about 0.8 miles of
shoreline, which extends northward into Eagle Harbor. The spit was extended and filled
at least twice prior to the 1950s, and was the location of the FPA where wood treatment
activities that caused the current soil and groundwater contamination were carried out.
The focus of the pilot study was the pilot test area located in the FPA within the
Soil and Groundwater OUs. The pilot test area comprises approximately 12% of the
surface area of the FPA.
2.1 SITE HISTORY
Prior to 1904, the Wyckoff property was owned by a sand mining operation and a
brickyard. From 1904 through 1988, the site was used for the treatment of wood
products (e.g., railroad ties and trestles, telephone poles, pilings, docks and piers) by a
succession of owners and companies. Chemicals used at the site include creosote,
pentachlorophenol, solvents, gasoline, antifreeze, fuel, waste oil and lubricants. These
chemicals were stored in above-ground storage tanks, conveyed through above- and
below-ground piping, disposed in sumps, spilled and buried on site.
USEPA began an investigation of the property in 1971, and the site was
subsequently placed on the National Priority List (NPL) in 1987. In 1988, the Wyckoff
Company ceased all operations on the property. In 1989 a groundwater pump and treat
system was installed to treat contaminants in the FPA. In 1993, USEPA assumed
management of the Soil and Groundwater OUs, and in 1994 the assets of the former
Wyckoff Company (now Pacific Sound Resources) were placed into an environmental
trust.
2.2 SITE GEOLOGY AND HYDROGEOLOGY
The Wyckoff site straddles the boundary between (1) a glacial drift plain
deposited 13,000 to 15,000 years ago as part of the Vashon Stade of the Fraser Glaciation
and (2) marine and fluvial deposits of the Seattle Basin. The subsurface can be divided
into the following general hydrogeologic units:
Vadose zone
Unconfined upper aquifer
Low-permeability aquitard
Semi-confined lower aquifer
Figure 2.2-1 displays a southwest-northeast profile through the site, which shows
typical relationships between soil types and occurrence of NAPL.
The vadose zone is approximately 10 ft thick and consists of fill and native
materials composed of discontinuous silt and fine sand layers. NAPL has been observed
in the vadose zone, but it is not pervasive.
The upper aquifer consists of approximately 5-10 ft of fill, silt, and fine sand
(similar to vadose zone soils), overlying a sequence of marine sand and minor
interbedded gravel, silt, and clay. The marine sands range in thickness from 5-70 ft, with
thickness increasing to the northeast. The upper aquifer in many parts of the site is
grossly contaminated with NAPL and dissolved-phase constituents. Prior to installation
of the outer sheet-pile wall, the upper aquifer was in direct communication with Eagle
Harbor and Puget Sound. Tidal effects, transmitted from the lower aquifer through the
aquitard, are still evidenced in some wells. The natural groundwater gradient is from the
south and radially towards Eagle Harbor and Puget Sound.
Separating the upper aquifer from the lower semi-confined aquifer is a relatively
impermeable layer comprised of marine silt and glacial till. The top of the aquitard
extends from near ground surface in the southern part of the site to approximately 75 ft
bgs in the north. At the time of design, the aquitard was thought to be continuous
throughout the site, with thickness varying from as little as 5 ft to 40 ft. More recent
interpretations indicate that the aquitard is absent in the southeast corner of the site and
may be laced with interconnected sand lenses elsewhere. NAPL has been observed in
several borings that penetrate the aquitard, primarily associated with sand and gravel
interbeds. More recent interpretations indicate that the aquitard is absent in the southeast
corner of the site and may be laced with interconnected sand lenses elsewhere.
The lower aquifer consists primarily of sand, with lesser amounts of silt, clay and
gravel. The base of this aquifer has not been determined with certainty, although the few
boring logs that penetrate deep enough suggest that the aquifer extends to approximately
200 ft bgs. Tidal effects are ubiquitous in the lower aquifer, presumably due to tidal
loading on the aquitard. Horizontal groundwater gradients are towards Eagle Harbor and
Puget Sound. Vertical gradients are generally upward (from lower to upper aquifer), but
short-term reversals may occur at low tide. Low levels of PAH contamination are present
in some monitoring wells and NAPL was observed in CW-15 during the baseline
groundwater characterization sampling in November 2002.
2.3 GROUNDWATER TREATMENT SUMMARY
In 1988 the Wyckoff Company was ordered by USEPA to install groundwater

extraction wells and a GWTP to extract and treat contaminated groundwater, minimize
further releases of contaminants to the surrounding surface water, and to recover as much
NAPL as possible. The system includes both biological and physical/chemical unit
processes, including activated carbon. The GWTP remained in operation under Pacific
Sound Resources until November 1993 when the USEPA took over the site under
Superfund.
The GWTP accommodates a maximum practical flow rate of 80 gpm. On

average, the treatment plant treats approximately 2.5 million gallons of groundwater per
month, for a total from November 1993 to September 2002 of approximately 250 million
gallons of groundwater. Separated from the extracted groundwater was approximately
59,000 gallons of NAPL that was shipped off-site for destruction.
Until the pilot study was conducted, the treatment system remained relatively
static as far as processes and upgrades. However, during the conceptual design phase, the
GWTP was recognized to be near the end of its useful life. Therefore, prior to the start of
the pilot study, minor upgrades to replace the already failing depurator and for new
piping to sustain the higher temperature waste stream anticipated from the pilot test area
were designed in order to handle the anticipated increase in contaminant loading during
steam injection operations. Details on the design changes and implementation are
included in Section 4.
SECTION 3.0 PILOT TEST AREA
BASELINE CONDITIONS
The pilot test area was selected in April 2000 (Appendix B) because information
gathered during previous investigations indicated that this area contained all of the
geologic units present at the site with the exception of the marine silt unit, and that
relative volumes of the geologic units are representative of the overall project area. The
down-sized project area includes approximately 16,400 cubic yards of soil in the vadose
zone and the contaminated aquifer. Also, it was believed that there were at least five feet
of LNAPL in the upper portion of the aquifer, as well as several feet of DNAPL directly
overlying the aquitard in this area, and that significant NAPL contamination was present
in the non-marine clay and glacial unit. NAPL was believed to occur as deep as 11 feet
into the aquitard. As a result of observations during drilling conducted for baseline
characterization, estimates of the amount of LNAPL and DNAPL in the area were
approximately 10,000 and 30,000 gallons, respectively.
During design of the original pilot test area injection and extraction well arrays, it
was known that residual NAPL existed up-gradient of the area, but it was believed that
this contamination could easily be removed by excavation to prevent interference with
the pilot study. Downsizing of the pilot test area wellfield resulted in exclusion of some
areas with mobile NAPL upgradient of the treatment zone, in addition to the areas with
residual NAPL. Neither the mobile nor the residual NAPL was excavated prior to the
start of pilot study operations.
During the installation of the injection and extraction wells and instrument strings,
baseline characterization of the pilot test area was carried out. The purpose of the
baseline characterization was to determine the location of NAPL in the pilot test area and
to measure the concentrations of the contaminants of concern in soils within the pilot test
area for comparison with post thermal treatment conditions. Site-specific hydraulic
conductivity data were determined through pumping tests conducted at extraction well E-
4 (Appendix C). These data assisted in selecting the optimum steam injection well screen
length, and to confirm the original design assumptions used for estimating flow rates. In
addition, soil samples were collected from the pilot test area to evaluate microbial
populations in the vadose and saturated zones. Three soil samples were collected and
analyzed for physical characteristics and thermodynamic properties.
3.1 FIELD INVESTIGATION METHODS
Field work was performed by the USACE beginning in May 2001 and completed
in March 2002. Soil borings for sampling and analysis were completed using either a
truck-mounted or a track-mounted geoprobe, depending on the requirements for diameter
of the sampling device. Samples were collected from locations pre-identified for
installation of injection wells, extraction wells, or instrument strings. Cores were
collected in either a 2.2 inch or 3 inch outside diameter by four foot long steel sample
barrel fitted with a clear polyethylene terephthalate glycol (PETG) or polyvinyl chloride
(PVC) sample sleeve. The larger diameter borings, accomplished with the track-mounted
geoprobe, were necessary to accommodate the downhole instruments used in the holes
completed as instrument strings. Continuous cores were collected until the aquitard was
encountered. Each core retrieved was logged for stratigraphy and evaluated for the
presence of NAPL by a USACE geologist. The following conservative assumptions were
used to translate observed NAPL occurrence into potential in situ cumulative thickness:
1. If mobile NAPL was noted, the entire soil interval in which the NAPL was found
was assumed to contain NAPL;
2. If mobile NAPL stringers were noted, and soil grains were coated throughout the
soil interval, then the entire soil interval was assumed to contain NAPL;
3. If mobile NAPL stringers were noted, and a heavy sheen was present, then half of
the soil interval was assumed to contain NAPL;
4. If NAPL stringers were noted, then each stringer was assumed to represent 0.1
feet of NAPL; and
5. If soil grains were coated or a sheen was present, no NAPL thickness was
assumed.
Two samples for chemical analysis were collected from each four foot long core,
one representing the top two feet and a second representing the bottom two feet. Samples
from 0 to 6 foot depth were considered vadose zone samples, and were analyzed for
semi-volatile organic compounds by an EPA Contract Laboratory Program (CLP)
laboratory using EPA SW-846 Method 8270B. The samples that had low PAH
concentrations were reanalyzed using Selected Ion Monitoring to allow for reporting
limits that were less than the MTCA Method B soil cleanup levels. Each sample from the
saturated zone was analyzed for PAH, diesel and oil range petroleum hydrocarbons, and
pentachlorophenol (PCP) by the EPA Region 10 Field Analytical Support Programs
(FASP) mobile laboratory. The on-site FASP laboratory used gas chromatography/flame
ionization detector methods to identify PAH, C12 to C24 (diesel range) and C24 to C35
(motor-oil range) concentrations. PCP concentrations were measured using electron
capture detector methods. Approximately 50 samples from the saturated zone were also
analyzed by the CLP method to obtain a broader target analyte list than was possible with
the FASP analytical protocols. Seventeen vadose zone samples were shipped to an off-
site laboratory for dioxin-furan analysis. The results of this investigation are presented in
the Preliminary Review Draft Baseline Characterization Report.
3.2 PILOT TEST AREA GEOLOGY AND HYDROGEOLOGY
The depth to the top of the aquitard in the pilot test area ranges from
approximately 25 feet at the southern part of the treatment area to 35 feet at the northern
portion. Four of the geologic units present in the FPA were found in the pilot test area:
fill; marine sand and gravel; non-marine clay; and glacial clay, silt and sand. A marine
silt that approaches thicknesses of 37 feet in other parts of the site is absent in the pilot
test area, as is a fluvial sand that occurs in relatively restricted areas. Figure 3.2-1 shows
a location map for three profiles through the pilot test area. The profiles are displayed in
Figures 3.2-2, 3.2-3, and 3.2-4.
Approximately the top five to seven feet of the pilot test area appears to be sand
or silt fill, although in places the fill is up to 18 feet deep. The fill material was imported
from nearby sources, and consists of silt and fine-grained sand similar in physical
characteristics to the underlying marine sand and gravel unit, making it difficult to
differentiate between the two units. Locally the fill consists of fine brown sand and
contains anthropogenic material such as bricks, broken glass, and metal fragments.
The central and southwestern portion of the pilot test area is covered by a
relatively thin (less than five feet thick) layer of non-marine clay. This clay is believed to
have been excavated from the upland bluff to the south of the site. It consists of fine to
medium brown sand, overlain by very soft to medium gray clay with occasional plant
fibers, wood fragments, and roots.
The marine sand and gravel unit comprises the majority of the treatment zone,
with a thickness ranging from a few feet at the southern end of the pilot test area to
approximately 25 feet at the northern end. It is a nearshore marine/beach deposit, which
is present over nearly all of the Wyckoff facility. The unit is generally continuous to the
top of the glacial deposits. It consists of loose to dense, poorly graded, gray to dark gray,
fine to medium sand with shell fragments throughout. Gravel lies in contact with the
aquitard in the central part of the pilot test area, extending from the northern point to the
southern extent of the treatment area.
The glacial clay, silt, and sand lies stratigraphically beneath the marine sand and
gravel unit, and forms an aquitard separating the upper aquifer from the lower aquifer in
the FPA. The aquitard has a northeastward slope that averages 7 degrees from horizontal.
The elevation of the aquitard was well predicted over most of the pilot test area, with the
exception of the northeast corner where a distinct depression exists. The glacial deposits
consist mainly of gray to brown, stiff to hard clay and silt, with some sand and gravel,
and no visible organic matter. Sand intervals within the glacial aquitard have been
observed in the central portion of the FPA, close to the pilot test area. The glacial deposit
was not sampled during the baseline characterization.
The water table in the pilot test area is generally six to ten feet below ground
surface. Aquifer testing conducted after installation of the pilot test area extraction wells
indicated that horizontal conductivity within the pilot test area is in the range of 15 to 30
feet/day (average 26 feet/day), with an average vertical anisotropy of 4.7. Thus the actual
hydraulic conductivity is less than half the value of 54 feet/day that was used in the
thermal remediation model, and the vertical anisotropy is approximately of the value of
20 used originally. The piezometric level in the lower aquifer is one to two feet higher
than the static water level in the upper aquifer, indicating that vertical flow is upward
through the aquitard under average conditions. Evidence for tidally-influenced
fluctuations in pilot test area extraction wells E-5 and E-6 during aquifer testing implies
that there is fairly good communication between the upper and lower aquifers in part of
the pilot test area. Hydraulic communication may be enhanced by interconnected sand
layers within the aquitard.
Samples taken from three instrument string locations (T-1, T-22, and T-30) were
analyzed for physical characteristics and thermodynamic properties. The samples were
chosen to represent the range of soil types observed in the pilot test area. Results of the
analysis are presented in Table 3.2-1.
3.3 CONTAMINANT DISTRIBUTION IN SOIL
Both LNAPL and DNAPL were found during the baseline characterization, with
significant accumulations in both the northern and western portions of the pilot test area.
In the western portion, there is considerable LNAPL in the non-marine clay and marine
sand and gravel unit, generally at depths of 12 to 16 feet bgs. In this LNAPL zone the
highest concentrations of PAH and total petroleum hydrocarbons (TPH) were detected.
Generally the total PAH concentrations are in the range of 5,000 to 10,000 mg/kg at this
depth, but concentrations as high as 105,000 mg/kg were measured. TPH concentrations
are generally about three times the concentration of the PAH, but the ratio is lower when
the concentrations of PAH are greater. Total TPH concentrations as high as 125,000
mg/kg were measured, and the highest concentrations of TPH are co-located with the
highest PAH concentrations. Naphthalene makes up 90 percent and more of the PAH
found in the LNAPL in the western portion of the pilot area. Thus, the LNAPL in the
pilot test area appears to be primarily composed of naphthalene and petroleum
hydrocarbons. Naphthalene, with a melting point of 80 C, is normally a solid at ambient
temperatures. It is likely that the petroleum hydrocarbons in this area act as solvents to
maintain the naphthalene in a dissolved state. Crystalline PAHs also occur within this
area, usually at or near the base of the non-marine clay.
In DNAPL zones along the western portion of the pilot area, soil samples show
total PAH concentrations in the range of 1,000 to 3,000 mg/kg. Naphthalene was also the
most common PAH found in the DNAPL, but in some individual samples, phenanthrene
was more abundant than naphthalene.
In the southeastern portion of the pilot area, particularly around extraction well E-
5, contaminant concentrations in many of the soil samples were less than detection limits.
Only a few soil samples have been collected from the northeastern portion of the pilot
test area, and the data do not allow conclusions to be drawn about the level of
contamination in that area. At the east and central area next to the sheet pile wall are two
instrument string borings (T-52 and T-53) that again show very high naphthalene and
TPH concentrations in the 8 to 10 foot depth interval, similar to that found in the western
portion of the site.
The samples submitted to the CLP laboratory in order to obtain a broader target
analyte list showed that PAHs were the most abundant constituents detected by this
method. Other constituents detected less frequently and at generally smaller
concentrations include 1-methylnaphthalene, 2-methynaphthalene, 1,1-biphenyl, 4-
methylphenol, 2,4-dimethylphenol, and acetophenone.
PCP was typically detected in soil samples collected from the pilot test area only
at low concentrations (usually less than 1 mg/kg). Most of the detections were in the
vadose zone soil samples. However, a few locations along the northeastern part of the
sheet pile wall, such as T-49, E-2, and E-6A, had higher concentrations over significant
depths. PCP concentrations in the range of 1 to 20 mg/kg were detected in the 8 to 18
foot depth range of these boreholes.
A total of 17 soil samples were analyzed for dioxins and furans. These samples
were from either the 0 to 4 inch or 4 to 4.33 ft depth interval. Octachlorodibenzodioxin
(OCDD) was detected in the highest concentration of all the congeners with
concentrations generally in the range of 10,000 to 40,000 ng/kg. The highly-toxic
2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD or simply dioxin) was detected only
at low concentrations, the highest detection being 0.9 ng/kg. The surface soil samples
had toxicity equivalent concentrations in the range of 10 to 130 ng/kg. The
concentrations in the deeper samples were generally lower, ranging from 0.05 to 38
ng/kg.
3.4 MICROBIAL POPULATION EVALUATION
In order to evaluate microbial populations before steam injection, soil samples

were obtained during the installation of the pilot system to determine PAH degradation
and microbial community structure. Microcosms were constructed using vadose and
saturated zone samples from various areas of the pilot test area, then incubated for 8 or 9
weeks, before sacrificing to determine PAH concentrations. Control microcosms were
also constructed using sodium azide to eliminate microbial activity, however, control
microcosms exhibited a loss of PAHs that could have been due to microbial activity or to
vaporization of contaminants during the incubation period. Vadose zone soil microcosms
showed relatively little change in PAH concentrations during the incubation period.
Most of the saturated zone microcosms exhibited a loss of PAHs, however, without good
data from the control microcosms and without demonstrating oxidation activity by
oxygen uptake or byproduct formation, microbial degradation cannot be documented.
In order to determine microbial community structure, which can be defined as the

relative abundance and diversity of microorganisms present, vadose and saturated zone
samples were submitted to Microbial Insights, Inc, for phospholipid fatty acid (PLFA)
analysis. Results from this analysis show that the total biomass values suggest similar
cell populations in all samples, with estimates ranging from 1.27 to 8.7 x 106 cells/gram
dry weight. In general, the viable biomass and microbial type distributions were similar
for a given sample before and after the degradation study. All samples were dominated
(>45%) by Gram-negative bacteria. Two indicators of metabolic activity and stress
suggested that the saturated zone is a less hospitable environment for bacteria than the
vadose zone. Results of this study are included in Appendix D.
Because the pilot study was not completed, confirmation soil samples needed to
complete the microcosm study were not collected and the study has not been completed.
3.5 GROUNDWATER RESULTS
Groundwater samples were obtained from seven extraction wells (E-1 through E-
7), five lower aquifer monitoring wells (99CD-MW02, 99CD-MW04, CW-05, CW-09,
and CW-15), and three upper aquifer monitoring wells (MW-17, MW-18, and MW-19) in
early November 2002. Well locations are shown in Figure 3.5-1. Data summary tables
are included in Tables 3.5-1 through 3.5-3 and in Appendix I. At this time, some amount
of steam had been injected into the pilot test area, and temperatures in the upper half of
the upper aquifer around the injection wells had already reached temperatures as high as
approximately 5060 C. Temperatures around the extraction wells remained close to
ambient temperatures. Thus, these are not truly background samples, but they provide
the best information available on groundwater conditions at the start of the pilot study.
3.5.1 Pilot Test Area Extraction Wells
Pilot test area well monitoring was implemented to test remedy effectiveness.
Specifically, trends in dissolved-phase constituents would be tracked to assist in post-
treatment predictions of attainment of performance objectives. Two extraction wells
were sampled with the upper and lower aquifer samples on November 5, 2002 (E-4 and
E-6). All seven extraction wells in the pilot test area were sampled on November 7,
2002. Extraction wells E-1, E-2, E-3, E-6 and E-7 have similar high concentrations of
PAH. Extraction well E-4 had PAH concentrations that were an order of magnitude
lower, which might be expected since soils immediately to the east of this well were
relatively clean. Extraction well E-5 was surrounded by mostly clean soils, thus it
contained only very low concentrations of PAH early in the pilot study. It appears that
the sample from extraction well E-1 was over diluted by the laboratory, as the reporting
limits are high, and even the constituent with the highest concentration (naphthalene) is
qualified as estimated (J qualified). This was a consistent issue with the high
concentration extraction well water samples sent to the USEPA CLP laboratory.
Additionally, the two samples collected from E-6 showed high variability in results, most
likely due to the use of the sampling port for sample collection. Low-flow sampling
methods were not used to collect samples from the extraction wells.
Among PAHs, naphthalene is present at the highest concentration, as would be

expected due to its abundance in the soil and relatively high solubility. The high
molecular weight PAHs were generally not detected in this first round of groundwater
samples. PCP was only detected in extraction well E-6 effluent in this initial round of
sampling, and the concentration was relatively low (22 g/L) compared to maximum
concentrations in other parts of the site. Other constituents that were detected include
1,1-biphenyl, 2,4-dimethylphenol, 2-methylnaphthalene, 2-methylphenol, and 4-
methylphenol, the same constituents that were found in the soil samples.
3.5.2 Upper Aquifer Monitoring Wells
Upper aquifer monitoring was initiated to document conditions (for completeness)

in the upper aquifer outside of the pilot test area. Three monitoring wells in the upper
aquifer, MW-17 which is to the west of the pilot test area, MW-18 which is just outside
of the pilot test area to the northeast, and MW-19 which is to the southwest, were also
sampled in November 2002. MW-17 and MW-18 are in highly contaminated areas, and
thus, show PAH contamination in excess of the cleanup levels (see Appendix A for
specific cleanup levels). MW-19 is in a cleaner area of the site, and contains less
contamination. However, the concentrations of benzo(a)pyrene, benzo[b]fluoranthene,
benzo[k]fluoranthene, chrysene, and indeno(1,2,3-cd)pyrene in this well exceed the
cleanup levels.
3.5.3 Lower Aquifer Monitoring Wells
Sampling of lower aquifer monitoring wells was intended to document any

adverse effects of thermal remediation in the pilot test area on groundwater quality in the
lower aquifer. Five lower aquifer wells were sampled in November 2002: CW-05, CW-
09, CW-15, 99CD-MW02, and 99CD-MW04. The lower aquifer monitoring wells
closest to the pilot test area, 99CD-MW02 and 99CD-MW04, both showed low levels of
PAH contamination at this time. PAH detected in these samples included acenaphthene,
anthracene, phenanthrene, fluoranthene, and pyrene. None of these concentrations were
greater than the cleanup levels. The other lower aquifer monitoring wells that were
sampled are all closer to the shoreline than the pilot test area, and all have historically
shown contamination. During the November 2002 sampling round, all three wells
contained PAH concentrations that slightly exceeded the cleanup levels. A small amount
of NAPL was purged from CW-15 during this sampling round. Compounds that exceed
the cleanup levels included benzo(a)anthracene, chrysene, benzo(a)pyrene, benzo(b)
fluoranthene, benzo(k) fluoranthene, and pentachlorophenol.
3.5.4 Summary of November 2002 Groundwater Results
Groundwater samples were collected from upper aquifer extraction and

monitoring wells, as well as from lower aquifer monitoring wells in November 2002 just
after the start of steam injection. Extraction wells E-1, E-2, E-3, E-6 and E-7 have
similar high concentrations of PAH. Extraction well E-4 had PAH concentrations that
were an order of magnitude lower, which might be expected since soils immediately to
the east of this well were relatively clean. Extraction well E-5 was surrounded by mostly
clean soils, thus it contained only very low concentrations of PAH early in the pilot
study. Upper aquifer monitoring wells MW-17 and MW-18 are in highly contaminated
areas, and thus showed PAH contamination in excess of the cleanup levels. Upper
aquifer monitoring well MW-19 is in a cleaner area of the site, and contains less
contamination. The lower aquifer monitoring wells closest to the pilot test area, 99CD-
MW02 and 99CD-MW04, both showed low levels of PAH contamination at this time.
Lower aquifer monitoring wells CW-05, CW-09, and CW-15 are all closer to the
shoreline than the pilot test area, and all have historically shown contamination.
SECTION 4.0 DESIGN AND CONSTRUCTION
SUMMARY
This section describes the design and construction of the pilot study systems. The
treatment duration, treatment area, above-ground conveyance systems, and fluid-cooling
and treatment systems were progressively downsized throughout the redesign and
construction process. Since the downsizing had significant impacts on the project
outcome, the evolution of each major system, including revisions based on revised
calculations and modeling and changes made in the field, are described in this section.
Detailed original design criteria and assumptions are documented in the pilot
study planning documents described in Section 1, specifically the Pilot Study Conceptual
Design and the Design Analysis. Elements of the system that were redesigned are
described in the Design Analysis Amendment. Additional changes were made during
construction and pilot study operations. Construction at the site began in November 2000
and continued through the operational phase of the pilot study. A timeline showing when
each system was installed is included in Figure 4.0-1. The layout of the systems at the
Site during the operational phase of the pilot study is shown in Figure 2.0-1.
Three construction contractors were responsible for installation of the major

systems at the Site: Bay West, Inc., Marine Vacuum, Inc. and Pease Construction, Inc.
USACE-Seattle District personnel designed, procured, and installed all subsurface
instruments and the automated data acquisition system (ADAS). Construction oversight
was provided by USACE personnel.
4.1 DESIGN PROCESS
The design process was initiated through a series of four meetings between
USEPA, USACE, and the ITTAP, from April 1999 through October 1999. During this
time, several planning documents were prepared, with review by ITTAP and other
technical reviewers within USEPA and USACE:
o Conceptual Design. Soil and Groundwater Operable Units. 1999.
o Thermal Remediation Shoreline Model Report. Groundwater and Soil Operable
Units. 2000.
o Thermal Remediation Pilot Test Area Selection Memorandum and Amendment.

Groundwater and Soil Operable Units. USACE. 2000.
After the ITTAP meetings were concluded, some of the panel members were
retained to review the pilot study design documents, including:
o Thermal Modeling Protocol. 2000.
o Thermal Modeling Proposal. 2000.
o Comprehensive Report. Wyckoff NAPL Field Exploration. Soil and

Groundwater Operable Units. 2000.
o Final Conceptual (10%) Design. Thermal Remediation Pilot Study. 2000.
o Final Design Analysis. Thermal Remediation Pilot Study. 2001.
The last of the design documents, listed below, received internal (USEPA and
USACE) project team review only:
o Final Design Analysis Amendment (DAA). Thermal Remediation Pilot Study.

2002.
A three dimensional (3D) thermal model was used to provide design criteria such
as optimum well spacing and screen lengths, steam injection rates, vapor and liquid
extraction rates, heating and steam-breakthrough times, and surface temperatures.
Modeling code requirements and a comparison of available codes were presented in the
Thermal Modeling Proposal and the Thermal Modeling Protocol. The 3D numerical,
thermal, multiphase, multicomponent flow and transport code M2NOTS (Adenekan,
1992) was selected because of its previous use for solving problems involving steam
injection into NAPL-contaminated soils, and also because of the long service record of
the underlying flow code TOUGH (Pruess, 1987). Code support was provided by the
University of California at Berkeley. During preparation of the DAA, the design model
was converted to NUFT (Nitao, 1998), a similar code that was more user-friendly and
could be used with the Department of Defense Groundwater Modeling System (GMS)
graphical-user interface.
During the redesign as described in the DAA, calculations for non-condensable

gas flow and numerical modeling (NUFT) of the estimated liquid flow rate, vapor flow
rate, heating time, steam injection break-though, and pressure cycling parameters were
re-run. These new values were to be used by the operations team during and after active
steaming operations to evaluate the performance of the system.
Fluid conveyance and treatment system designs incorporated design criteria

developed by field testing, the above subsurface and process-design modeling, and
consultation with ITTAP members. The site groundwater model was completed while
system design was in progress, providing calibrated hydraulic parameters that were used
to update the thermal model and well field design.
Criteria generated during the course of the original design process, as well as the
revised criteria developed during the redesign, are summarized in Table 4.1-1.
4.2 SYSTEM DESCRIPTIONS
The design and construction of each of the major systems of the pilot study are
described in this section. The original design for each system is presented and then
compared to the revised design as well as the system that was actually constructed at the
site. Changes were made to the design, both prior to and during construction, as well as
during operation of the pilot study, in order to meet project schedule and budget
constraints. In general, the changes in the project design required changes to be made to
the operational strategies for the individual systems. The original design and subsequent
changes made to the major systems are summarized in Tables 4.2-1 through 4.2-3.
4.2.1 Sheet Pile Wall
From November 2000 to February 2001 a 536-foot long sheet pile wall
surrounding the steam injection pilot test area was constructed by Bay West. The pilot
test area sheet pile wall was designed to mimic the shape of the entire FPA to evaluate
NAPL removal collected against the wall and the impacts of heating on the sheet pile
wall. The driving depths of the sheet pile varied from 15 to 45 feet. The toe of the wall
was keyed a minimum of 4 feet into the upper surface of the underlying glacial till. The
original design and location of the sheet pile wall was not changed during the course of
the project.
4.2.2 Well Field
This section describes the design and construction of the injection and extraction
wells located in the pilot test area. The original design and subsequent changes made to
these systems are summarized in Table 4.2-1.
Instrumentation strings for subsurface water level and heat monitoring were also
installed in the well field. Subsurface instrumentation included vibrating wire pressure
transducers, thermocouples, and a fiber optic distributed temperature sensor (DTS)
system. Details on instruments and the ADAS are included in Appendix E.
Well locations and screen lengths shown in the original design (DA) were based
on model simulations of a 7-spot well array, using the M2NOTS code. Model runs were
performed with variable vertical and horizontal hydraulic conductivities to provide a well
field design that could accommodate a range of field conditions. Wells were designed to
be located in the field according to the following criteria:
The maximum well-spacing (distance between injection and extraction wells) is

1.6 times the aquifer depth, based on model results.
Arrays would be 7-spot if possible (6 injection wells surrounding a central

extraction well).
Arrays should cover the NAPL-contaminated portion of the pilot test area, with
the southern-most injection wells outside the NAPL-contaminated area.
Injection and extraction wells should alternate along the sheet pile wall.
These design criteria were derived from site data, heat- and fluid-flow
calculations and models, assumptions based on standard industry practice, and input from
ITTAP.
The resulting well field design consisted of two 7-spot arrays, one 6-spot array,
and 4 partial arrays bisected by the sheet pile wall, incorporating 25 injection wells and
14 extraction wells. However, due to the need to reduce project costs, only 16 injection
wells and seven extraction wells were included in the redesign (DAA) and installed at the
site. At the time the decision was made to reduce the number of injection and extraction
wells, several wells had already been installed. Locations of the injection and extraction
wells are shown in Figure 4.2-1. Well construction details are shown in Figure 4.2-2.
Specifications for the extraction well pumps are shown in Table 4.2-1.
The southern-most arrays shown in the original design were eliminated during the
redesign, and it was understood at the time that the relatively small volume of NAPL in
those arrays would not be treated. It was further believed that contaminants could
potentially be displaced in the upgradient direction by the expanding steam zone, with
some potential for recontamination of the treated area once steam injection was halted.
However, the project team felt that hydraulic control, established by the southern
extraction wells (E-3, E-5, and E-7), would limit recontamination to the southernmost
portion of the pilot test area. It should be noted that it was intended that this area would
be re-treated during full scale operations.
The original design required injection wells to be 4-inch and 6-inch inside
diameter (ID) with a service port on the side of the riser. The larger wells could be used
for extraction if necessary. Each injection well head would include a pressure-relief
valve and an air-injection port. Of the 16 injection wells installed at the site, ten are 6-
inch diameter wells and six are 4-inch diameter wells. An air-injection port was installed
at each injection wellhead. The pressure-relief valves were not installed at each well-
head because the pressure relief system was installed at the boiler itself.
The original design required extraction wells to be 6-inch and 10-inch ID. The
wells were to contain pneumatic piston pumps (Blackhawk or QED Hammerhead) to
remove groundwater and NAPL. The larger extraction wells (up to five) would be fitted
with dual pumps at two different elevations to allow combined removal of dissolved
phase and NAPL.
Of the seven extraction wells installed at the site, six are 10-inch diameter wells
and one is a 6-inch diameter well. Before bids were received for the construction
contract, Blackhawk pumps were deleted and replaced with QED pumps due to field
problems observed at a different SEE site. The QED pumps were believed to be more
compatible with elevated temperatures and chemically reactive environments, and also
were not likely to generate liquid emulsions when pumping water and NAPL together.
During negotiations for the construction contract, the upper pumps were eliminated to
save costs. QED pumps were supposed to be installed in all extraction wells one foot
above the bottom of the screen. During thermal operations it was discovered that some of
the pumps were placed several feet higher than intended because of miscommunication
between the contractor and USACE oversight personnel regarding reference points. The
pumps that were positioned incorrectly were repositioned during operations.
Model simulations performed for the original design utilized 10-foot injection
screens located at the base of the aquifer, with extraction wells screened from the base of
the aquifer to 8 feet below ground surface. Aquifer testing during well installations
indicated average horizontal and vertical hydraulic conductivities of 25 and 5 ft/day,
respectively. Since these values were near the high end of the range used in model trials,
injection screen lengths were reduced to 5 feet. Extraction well screens were extended to
4 feet below ground surface to improve near-surface contaminant removal.
The original design required all injection and extraction wells to be constructed
entirely of type 304 stainless steel, with 0.02-inch-slot wire-wound screens set at the base
of the aquifer, and 5-foot sumps placed into the aquitard. A filter pack of #10-20 sand
was selected, retaining at least 70% of the formation. The design required the wells to be
grouted with 40% silica-cement grout. Wells were installed according to these design
parameters.
The injection and extraction well installations penetrated the geomembrane in the
vapor cap (Section 4.2.4). The original design required high-density polyethylene
(HDPE) boots to be installed where wells penetrated the vapor cap. HDPE boots were
installed at the extraction wells to seal penetrations in the geomembrane, however, boots
were not installed for the injection wells or the instrument string penetrations because it
was thought at the time that the boots would be melted by the high temperature of the
injected steam. Instead, these penetrations were grouted through the cap to the surface.
Integrity of the geomembrane was necessary to optimize performance of the vapor
collectors and limit fugitive emissions.
4.2.3 Steam Generation and Conveyance
This section describes the design and construction of the steam generation and
conveyance system. The steam generation and conveyance system consists of a feed
water supply and treatment system, a deaerator, a boiler, a blow down tank, and piping
from the boiler to the injection wells. The original design and subsequent changes made
to these systems are summarized in Table 4.2-2.
The original design called for a boiler feed-water supply rate of 50 gpm. Feed
water was to be supplied from an on-site well. The feed-water requirement was
decreased to 40 gpm during the redesign due to treatment plant capacity constraints.
Water was to be pumped from the well, using a submersible pump, to a 1,500-gallon
water storage tank located near the boiler building via an underground 4-inch diameter
cast iron pipe. Three feed water pumps were to be used to convey the stored water to a
water softening system consisting of two water softeners that are regenerated with a brine
solution from a brine storage tank. Treated feed water was then to be fed to the boiler
through four plate and frame heat exchangers (HX-1, HX-2, HX-3, and HX-4) designed
to preheat the boiler feed water (upstream of a deaerator) and recover heat from the
extracted liquid and vapors. The feed-water heater was expected to increase the overall
thermal efficiency of the process.
The required on-site well was installed in January 2002. The pump size was
reduced during construction from the designed 400 gpm to 225 gpm. The formation in
which this well was drilled is capable of producing greater than 200 gpm. Well
construction information is found in the On-Site Water Supply Well Report (USACE,
2002). The subsurface feed water piping, pump, water storage tank, deaerator, and water
softening system were installed from June through September 2002. Changes made to
these systems are detailed in Table 4.2-2.
An 800-horsepower fire-tube boiler, capable of delivering 25,000 lb/hr (3.47

kg/s), with an efficiency of 86.8%, was called for in the original design. The costs of
leasing and purchasing were evaluated during design, and purchasing was judged to be
cost-effective. The boiler was installed in a pre-engineered metal building with a
concrete foundation. The layout of the boiler plant is shown in Figures 4.2-3 and 4.2-4.
The boiler installed at the site followed the design requirements, except the boiler
capacity was increased to 27,600 lb/hr because it was from a different manufacturer than
the one originally specified. A blow down tank, needed to eliminate the build up of
mineral scale from the boiler, was also called for in the design. The redesign required the
boiler blow down to be discharged to the product tank located in the boiler building
where it was to be combined with the extracted liquids on the way to the treatment plant.
Since the boiler size was increased, the boiler supplier recommended increasing the
designed capacity of the system for handling blow down. The tank size necessary to
handle blow down was increased and a tank was installed outside the boiler building
under a contract modification. This tank is allowed to discharge liquid into the swale on
the north side of the boiler building that discharges onto the contaminated soil in the FPA
instead of going to the treatment plant for processing. Boiler blow down water is not
considered contaminated.
Maximum fuel consumption of the boiler was expected to be 5,400 gpd of low-
sulfur #2 fuel oil. The original design required two storage tanks: a 6,000-gallon day
tank, and a 20,000-gallon reserve tank. The fuel oil storage and supply system was
sized to match the expected boiler operations. Total fuel usage was expected to be
250,000 to 300,000 gallons, the amount required to heat the pilot test area and deliver 2
pore volumes l.e. of steam to the subsurface. These two fuel storage tanks were installed
as designed.
Modeling for the original design predicted an average maximum steam injection
rate of 2 gpm l.e. per well (1,000 lb/hr). The predicted total steam requirement was 44
gpm (22,000 lb/hr), assuming that flow rates for injection wells adjacent to the sheet pile
wall would be reduced by 50%. The predicted steam flow rate for each individual
injection well was unchanged after aquifer testing and the redesign effort, due to the
compensating effects of higher hydraulic conductivity, lower extraction vacuum, and
reduced injection screen length. The reduction in the number of injection wells decreased
the total steam requirement, however, to 32 gpm l.e. (16,000 lb/hr instead of 22,000
lb/hr).
The original design required steam to be delivered to wells by insulated piping,

with a 6-inch ID main distribution pipe. The steam lines were to be sloped continuously
downward toward an anchor point, then lifted up 2 feet, and then sloped back downward
toward another anchor point. Steam flow to the subsurface was to be controlled at each
well with a manually-operated pressure-regulating valve and flow rate was to be
monitoring using flow meters.
Installation of the steam injection piping to the well field was completed in July
2002. Steam is distributed to the injection wells through ASTM Grade A-53 Schedule 40
black steel piping. The pipes were insulated to prevent unnecessary heat loss. An
aboveground 6-inch diameter main distribution pipe, in a U-shaped layout, provides
steam to the injection wells with branches from the main. Piping was installed on above
ground pipe supports. The slope of the steam line was continuously downward toward
the boiler plant rather than sloping the line toward an anchor point as called for in the
design as a contractor-designed request. Drip legs were installed near four concrete
anchor points. Condensation is piped back to the nearest injection well. Pressure
regulating valves and flow meters were installed as designed. Flow was determined by
pressure differential across a calibrated flow control valve. The steam generation, vapor
and liquid process flow diagram is shown in Figure 4.2-5. The aboveground piping
layout for steam, vapor and liquid conveyance is shown in Figure 4.2-6.
4.2.4 Contaminated Vapor/Liquid Conveyance and Treatment
This section describes the design and construction of the contaminated

vapor/liquid conveyance and treatment system. The vapor system consists of seven
vapor collectors in the subsurface of the well field, piping from the well field, a heat
exchanger, condensate pumps and tanks, liquid ring vacuum pumps, a sour gas vapor line
to the boiler, and a thermal oxidizer. The liquid system consists of piping from the
extraction wells in the well field, heat exchangers, a product tank, and the groundwater
treatment plant. The original design and subsequent changes made to these systems are
summarized in Tables 4.2-1 and 4.2-3.
4.2.4.1 Product Composition
Concepts regarding the composition of recovered product were based on site NAPL-
testing results as reported by USEPA in the Steam Injection Treatability Study, known
thermodynamic relationships, and average chemical properties (Table 4.2-4).
Although NAPL component concentrations were known to vary over the project site,
it was understood that the most abundant compounds in extracted groundwater would be
LPAH. The liquid solvent fraction in the creosote mixture was believed to consist of
diesel-range aliphatic compounds, which were known to have low solubility. Due to
their high vapor pressures, diesel-range aliphatic compounds would be transported
primarily in the vapor phase rather than the aqueous phase, and their presence in the
recovered product would strongly affect its characteristics. The pilot study was expected
to employ aggressive vapor extraction, which would draw the aliphatic fraction toward
the extraction wells and the surrounding formation; thus the compounds comprising the
liquid solvent portion of the creosote mixture were expected to be present in both
extracted liquid and vapor, and the recovered product was expected to be primarily in
liquid form.
4.2.4.2 Vapor Conveyance and Treatment System
The vapor collector system in the well field was designed to be installed on the
original ground surface of the pilot test area, and was to consist of an 8-inch thick gravel
layer containing 4-inch diameter collector laterals. The laterals were 30 to 80-foot-long
slotted steel pipes; one collector pipe was to be located above each well array. The
original design assumed a vacuum of 1 psi (6900 Pa), however this was reduced to 0.2 psi
(1500 Pa) during the redesign, and the design of the lateral pipe diameter was reduced to
3 inches. A six-inch soil layer was to be laid over the gravel to help protect the vapor cap
geomembrane. The cap was to consist of a 60-mil HDPE membrane, overlain by a 12-
inch protective soil layer, and a 6-inch gravel layer to allow for vehicle traffic. Four 10-
foot-square panels of different polypropylene geotextiles were to be installed below the
geomembrane to test their ability to withstand thermal and chemical conditions during
steam injection. Soil partition cells were to be installed throughout the cap to allow for
experimentation with vapor extraction rates.
The vapor cap and collector system installation was completed by MarVac in
September 2002. The installed vapor collection system and vapor cap followed the
original design. Vapor cap soil partitions were not installed due to budget constraints.
The collector laterals used were 4-inch diameter in lieu of the 3-inch diameter in the
redesign in order to accommodate the required number and length of slotted openings in
the pipe. The vapor cap collection piping and branch detail are shown in Figure 4.2-7.
The laterals are 37.5 to 88.5 feet long slotted steel pipes with one collector pipe. The
laterals were connected to vapor-extraction piping from extraction wells and conveyed to
the heat exchanger and treatment system. The vapor cap was graded with a crown at the
center of cap and sloped at a minimum of 1% toward the sheet pile wall and the swale at
the southern end of the pilot test area. Weep holes were installed in the sheet pile walls at
the elevation of the top of the gravel surface.
Two 10-foot x 10-foot test cells of geotextile fabric (one a polyester fabric, one a
polypropylene fabric) were installed beneath the vapor cap in the pilot test area. The test
cell locations are shown on Figure 4.2-1. The vapor cap system was installed prior to the
installation of any wells, the subsurface monitoring instrumentation system, and
conveyance piping.
The original design showed vapors being extracted from the wells and the surface
vapor collectors through insulated 8-inch maximum diameter black steel pipe. Three
liquid-ring vacuum pumps would be used to pull the vapor through the pipe, each rated at
450 cfm at 40 deg C and 33.6 kPa absolute pressure. Vacuum at the well heads was
intended to be 0.5 atm. For the redesign, the vapor extraction piping was downsized from
8-inch maximum to 6-inch, and the insulation was deleted. Additionally, the number of
vacuum pumps was reduced to two (one operating and one standby), each with a total
maximum flow rate of 450 acfm, or 140 scfm, (56% of the original design criterion) to
achieve a vacuum pressure at the well heads of 0.25 atm.
According to the original design, extracted vapor was to be cooled in the boiler
plant by two heat exchangers (HX-3 and HX-4), allowing water vapor and other
condensable gases extracted from the vapor cap and extraction wells to be condensed.
Any condensate that formed was to be collected in a condensate tank associated with HX-
3 and then added to the liquid treatment stream. Cooled non-condensable vapor was to
be used to supplement the boiler combustion air during boiler operation. The second heat
exchanger (HX-4) was deleted in the redesign. The original design included the use of a
steam-regenerated carbon-adsorption system when the boiler was not in operation. For
the redesign, this unit was replaced with a thermal oxidizer.
Non-insulated, six-inch diameter pipe of standard weight (Schedule 40) black

steel was installed at the site to convey extracted vapor to the boiler plant, as per the
redesign. The original design showed the vapor line sloping towards a low point at the
four anchor points to collect condensate, where a pump would convey condensate to the
nearest extraction well. The installed vapor line was sloped toward the boiler plant and
an extraction main drip leg, consisting of an 8-inch diameter pipe section, was installed
outside the boiler plant to collect condensate. The condensate was then pumped via a
condensate pump (PDP-1) to the liquid product tank in the boiler plant. The pump was
housed in a vinyl storage shed in the pilot test area.
Vapor from the well field and surface vapor collectors was cooled in one plate
and frame heat exchanger (HX-3). Condensables were collected in a condensate receiver
and pipeline drip leg (8 inch diameter by 24 inches long) installed downstream of HX-3
and upstream of the vacuum pumps to prevent liquids from damaging the pumps. The
vacuum pumps had filters that prevented particulate matter from entering the pump. A
bypass line for the non-condensable vapor was installed to allow for commissioning of
the liquid ring vacuum pumps (LRVP). Shortly after steam injection was started,
however, a slug of liquid inundated the collection system and entered one of the vacuum
pumps. Gasket and seal materials of LVRP-1 failed. At the same time, the contaminant-
containing liquid in HX-3 caused the rubber seals between the plates to swell, come out
of their frames, and leak water onto the boiler building floor. The vapor system was shut
down and it was never operated again.
The cooled non-condensable vapor was then conveyed through one of the two
installed liquid ring vacuum pumps (LRVP-1 and LRVP-2). During operations, the non-
condensable vapor was thermally treated in the boiler via the combustion air line. A
thermal oxidizer fueled by propane was installed for thermal treatment of the non-
condensable sour gas when the boiler was shut down. During commissioning of the
thermal oxidizer, however, it was found that the burner was too small for the vapor
system and a larger burner (and fan motor) was installed after operations were terminated
in April 2003. However, the vapor recovery system was never restarted after the vacuum
pump and heat exchanger failed and the thermal oxidizer was never used.
4.2.4.3 Liquid Conveyance and Treatment System
Thermal modeling predicted an average liquid extraction rate of 5 gpm per

extraction well, for a total of 45 gpm from the pilot test area. According to the model, a
maximum total extraction rate of 50 gpm would be required to keep water levels from
rising during the initial expansion of the steam zone. The average individual well
pumping rates did not change substantially for the redesign, however, the total average
pumping rate was decreased to 25 gpm due to the reduced number of extraction wells.
Extraction from the pilot test area was partially constrained by the 80 gpm maximum
capacity of the treatment plant. The treatment plant needed to be able to handle both
extracted liquid from the pilot test area as well as groundwater pumped from the FPA
necessary to maintain hydraulic control of the entire site. According to the site
groundwater model, the year-round average pumping requirement outside the pilot test
area would be 20 to 40 gpm after the outer sheet pile wall was installed.
Liquid from the pilot test area extraction wells was to be conveyed from the well
field to the boiler plant through a 3-inch pipe and collected in a product tank prior to
being cooled in the heat exchangers. The total organic carbon (TOC) content of the
liquid was to be measured using an in-line TOC analyzer. As per the design, the
extracted liquid was conveyed from the well field to the boiler plant through a 3-inch
diameter pipe of standard weight (Schedule 40) black steel. Extracted liquid then flowed
into the product tank. Influent of the product tank was analyzed for TOC using an
automated, in-line TOC analyzer with data collected through the ADAS.
Extracted fluids from the well field, including condensable gases, were to be
cooled in two heat exchangers. The heat exchangers included in the original design
required up to 150 gpm of cooling water at 15 C. During steam production, the boiler
feed water would be used to cool extracted liquids in two liquid-to-liquid heat exchangers
(HX-1 and HX-2) and vapors in two vapor-to-liquid heat exchangers (HX-3 and HX-4)
(see discussion above on vapor conveyance and treatment). Since the boiler requirement
was to be 57 gpm, the excess 93 gpm was to be routed through a cooling tower, and the
addition of the recycled water to the cooling stream would reduce the well-water
requirement to 70 gpm. It was understood in the original design that the available
cooling capacity might occasionally be insufficient during peak vapor extraction periods,
and vapor extraction rates might have to be reduced.
Design changes assumed a reduced cooling water demand of 40 gpm, and

eliminated the cooling tower. The cooling capacity of the boiler-feed water was given as
2.1 Mbtu/hr, roughly equivalent to 10% of the anticipated enthalpy of the injected steam.
As in the original design, it was understood that the cooling capacity could be insufficient
for predicted peak vapor extraction rates (at peak rates, the extracted vapor was expected
to be mostly steam). The cooling capacity of the system was further constrained by the
capacity of the groundwater treatment plant to discharge untreated non-contact cooling
water. The non-contact cooling water discharge pipe was connected to the main
treatment plant effluent pipe after tank 303. The size and configuration of the discharge
outfall into Puget Sound limited the amount of treated water that could be discharged,
which was reduced by the amount of non-contact cooling water.
Two plate and frame heat exchangers (HX-1 and HX-2) were installed in the
boiler plant to cool extracted liquids. New 2-inch HDPE piping was installed to convey
cooled liquids from the heat exchangers to the existing treatment plant. Extracted liquid
from the pilot test area was combined with extracted liquids from the former process area
at the treatment plant header. A new 2-inch pipeline was installed to convey non-contact
cooling water from the heat exchangers to the treatment plant sumps. This inadvertently
limited the amount of cooling water the treatment system could handle and therefore
limited the amount of steam that could be injected into the pilot test area.
Liquid treatment design was intended to employ the existing treatment plant as
much as possible, however, the Conceptual Design Report stated that the existing plant
had reached the end of its useful life. It was understood that the bioreactor could become
inoperative due to changes in temperature and contaminant concentrations resulting from
thermal treatment (ITTAP, 1999). The Conceptual Design Report recommended
additional granular activated carbon (GAC) polishing capacity to provide backup for
potential loss of the bioreactor due to microbial upset or structural failure. A new
treatment plant had been designed at this time at EPA, however, it was not built to
support the pilot study due to the need to reduce overall project costs.
Liquids would exit the heat exchangers in the boiler plant at 40 C, and be
conveyed to an existing 58,160 gal equalization tank (T-401) at the treatment plant. An
emulsion-breaking chemical would be added upstream of the equalization tank, and the
tank would be fitted with a skimmer to remove floating product. Coagulant and polymer
would be added, and more product would be separated by a dissolved air flotation tank
(DAFT). Skimmed product would go to an existing froth tank (SEP-108) and then would
be routed to a product storage tank (T-105). Remaining water would be recycled back
through the DAFT, mixed with a neutralizing chemical, routed to another equalization
tank (T-402), and then to the existing aeration basin (T-203). Vapors from all tanks prior
to the aeration basin would be extracted and conveyed to the vapor treatment system.
Corroded air spargers and piping in the existing aeration basin would be replaced,
and additional spargers would be installed in the existing clarifier to augment the capacity
of the aeration basin, basically creating a second aeration basin. Effluent from the two
aeration basins would be routed to an additional DAFT, and separated solids from the
DAFT would be recycled back to the aeration basins. Clarified effluent would be routed
to two existing multimedia filters and a third new one, then through the existing GAC
system, to a new 25,000 gal effluent storage tank. The existing effluent tank would be
replaced due to damage from the earthquake that occurred on February 28, 2001.
Recovered NAPL would be pumped from the froth tank to existing 10,150 gal and
7,100 gal product storage tanks, removed by tanker truck, and disposed off-site. Spent
GAC, multimedia filter material, and sludge from the bioreactor would also be disposed
off-site. Treated effluent would continue to be discharged to Puget Sound via the
existing effluent pipe and diffuser structure, in compliance with regulatory requirements.
During the redesign, the skimmer at tank T-401 and vapor recovery from all tanks were
eliminated. Additionally, the second DAFT, the third multimedia filter, and the
conversion of the clarifier to an aeration basin were eliminated from the design. The
effluent tank that was severely corroded and damaged by the earthquake was to be
repaired instead of being replaced. This repair was completed by Marine Vacuum, Inc.
when the build-up of tank bottom sludges was removed prior to the start-up of the pilot
study.
The liquid treatment plant layout and process flow diagram are shown in Figures
4.2-8 and 4.2-9, respectively. The liquid treatment used the existing treatment plant and
disposal practices, with the following exceptions:
A chemical tank containing caustic for emulsion breaking was installed

that included a chemical feed pump. A static mixer installed within the
feed line to T-401 enabled mixing of the caustic into the liquid stream
which was originally designed as an automated system. During the re-
design, it was converted to a manually operated system due to cost
constraints. However, this chemical feed system was not used during the
pilot study.
Coagulant and polymer were introduced to the liquid stream via a static
mixer upstream of the DAFT. A small polymer tank was installed at the
DAFT effluent as a wet well for the DAFT pumps. Coagulant and
polymer addition was not implemented until March 2003.
The froth tank, T-108, was elevated and piping modified in order to
gravity feed the recycled liquid back to the DAFT.
A chemical tank containing acid was installed to neutralize the liquid

stream from the DAFT. A static mixer and chemical feed pump were
installed to introduce the acid into the liquid stream, which was originally
designed as an automated system. During the re-design, it was converted
to a manually operated system due to cost constraints. However, the acid
feed system was not used during the pilot study.
4.3 MONITORING SYSTEM
The monitoring system was continuously revised from conceptual design to

construction. Extensive data collection was performed throughout the duration of the
pilot study in order to evaluate the systems effectiveness in meeting the nine project
objectives presented in Section 1.1. Data were to be collected from the following site
areas or systems:
Pilot test area. Temperature and pressure were to be monitored in the subsurface
to track the movement of heat and fluids throughout the treatment area. Specific
media, locations, parameters measured, testing frequency (proposed and/or
actual), and whether the monitoring was implemented are summarized in Table
4.3-1. See Appendix E for more detail concerning pilot test area monitoring.
Steam generation system. Flow, temperature, and pressure of steam production

from the boiler into the wells were to be monitored to support heat flux
calculations. Specific media, locations, parameters measured, testing frequency
(proposed and/or actual), and whether the monitoring was implemented are
summarized in Table 4.3-2. Due to cost constraints, some originally planned
instruments were not installed or were installed after active steam injection
started. Additionally, some instruments may not have performed as intended and
data were lost or unreliable.
Liquid and vapor extraction systems. Flow, temperature and chemical

characteristics of extracted liquids, as well as flow and temperature of extracted
vapors, were to be monitored to support operational decisions and determine
remedy effectiveness. Specific media, locations, parameters measured, testing
frequency (proposed and/or actual), and whether the monitoring was implemented
are summarized in Table 4.3-2. Due to cost constraints, some originally planned
instruments were not installed or were installed after active steam injection
started. Additionally, some instruments may not have performed as intended and
data were lost or unreliable.
Groundwater treatment plant. Water samples were collected from selected

sampling points within the GWTP to monitor performance and optimize
operations. Effluent discharge samples were collected from the GWTP to
demonstrate compliance with substantive requirements of the NPDES permit and
1988 Consent Decree. Specific media, locations, parameters measured, testing
are summarized in Table 4.3-3.
Upper and lower aquifer groundwater. Groundwater samples were to be collected

from groundwater wells and extraction wells to evaluate the effectiveness of the
thermal treatment system and to evaluate potential for off-site migration of NAPL
or contaminants of concern. Specific media, locations, parameters measured,
testing frequency (proposed and/or actual), and whether the monitoring was
implemented are summarized in Table 4.3-3.
Site perimeter. Noise, air quality, intertidal conditions, boiler air emissions and
sheet pile wall integrity were to be monitored to evaluate the potential impacts of
full-scale thermal treatment to the surrounding community and the near shore
marine habitats. Specific media, locations, parameters measured, testing
are summarized in Table 4.3-4.
Initial noise monitoring (baseline) was eliminated during negotiations with the
O&M contractor. It was agreed during negotiations that a previously conducted
noise survey would suffice as the baseline survey. Air quality and boiler air
emissions for the community were to be provided by a separate contractor (URS)
at the request of USACE. Air quality PM-10 monitors were set-up and calibrated
but never used. The boiler stack emissions were to be monitored by URS when
the boiler operations had stabilized and the sour gas to be burned in the boiler had
reached the standard operating conditions. Since this condition was never reached
before the vapor extraction system failed, stack emissions were not monitored.
Sheet pile wall leakage monitoring was performed approximately 3 months after
the sheet pile walls were installed. The next round of sampling that was
contracted out to URS was to be performed after the pilot scale operations were
completed. This sampling event was never initiated.
SECTION 5.0 OPERATIONS DESCRIPTION
This section discusses the proposed operations strategy and details of actual
operations. Additionally, the nature of the project team, communication methods,
decision-making process, and tools used to aid in communications and decision-making
are described. Operations log sheets that were posted daily to the project web page are
included as Appendix F.
5.1 OPERATIONS STRATEGY
As conceived, the operations strategy represented the consensus of experts from

industry, academia, and USACE. The fundamental principles and procedures were
developed from knowledge of other sites and from engineering analyses that included
fluid and heat flow calculations and numerical modeling. However, due to the innovative
nature of thermal treatment technology and the heterogeneous subsurface conditions at
the project site, procedures were to be continuously re-evaluated during the pilot study.
Revisions to the intended treatment strategy were necessary because of conditions
encountered during system construction, commissioning, equipment limitations,
monitoring results during operations, system performance during operations, and cost
constraints.
5.1.1 Intended Operations Phases
The thermal treatment process was to be performed in three phases, as detailed in

the RAMP and summarized below. A fourth phase related to site-wide management of
groundwater was part of the operations plan, but it is not specifically a pilot study
concern and is not discussed here.
5.1.1.1 Initial Heating
Objective: Heat subsurface to steam temperature, thereby enhancing mobility of

NAPL; continue until recovery of liquid NAPL has peaked and begins to diminish.
Implementation: Initiate steam injection while maintaining hydraulic control

through liquid extraction at the extraction wells and pneumatic control through vapor
extraction at the extraction wells and in the vapor collector layer. Target injection rates
were 400-800 lbs/hr per well or an overall rate of between 6,500 to 13,000 lbs/hr of
steam. Liquid extraction rates were targeted at 30-50 gpm for the entire pilot test area,
which translates to an average of 4-7 gpm per well. Vapor extraction was to be
accomplished through establishing a vacuum (up to 0.25 atm).
Duration: Breakthrough of steam to the extraction wells (and nearly complete

site heat-up) was anticipated to take 2-4 weeks. Period to reach diminishing liquid NAPL
recovery was uncertain.
5.1.1.2 Pressure Cycling
Objective: Optimize vapor and liquid-phase recovery at low NAPL saturations

by maintaining an economical mixture of groundwater and steam in the aquifer. Pressure
cycling is believed to enhance vaporization of lighter fractions of contaminants and
increase dissolution rates of NAPL. Pressure gradients established during pressure
cycling may promote flow of NAPL.
Implementation: Steam flow to the injection wells is reduced, while liquid and
vapor extractions continue. When contaminant recovery decreases, steam injection is re-
initiated. On breakthrough of steam to extraction wells, steam flow is reduced, and a new
cycle commences.
Duration: A few days to a few weeks per cycle. This phase was to continue
until contaminant recovery diminished. Total duration of the pressure cycling phase was
uncertain, but contract limitations restricted duration of the initial heating and pressure
cycling phases to six months, with options for an additional two months.
5.1.1.3 Fluid Extraction and Monitoring
Objective: Confirm continued reduction of contaminants.
Implementation: Continue extracting liquids and vapors, while monitoring

contaminant concentrations. Vapor extraction would diminish and eventually cease
during this phase.
Duration: Six months after cessation of steam injection, with an additional two
months optional.
5.1.2 Actual Operations Phases
Of the three intended operational phases, the pilot study never went beyond the
initial heating phase. Failure of the vapor extraction system and the treatment plant
aeration basin led to cessation of operations on December 15, 2002, and prevented
achievement of target temperatures in the subsurface. When operations resumed on
January 13, 2003, the vapor extraction system was inoperable, so the guiding principle
became to maintain subsurface temperatures (using low steam injection rates), while
minimizing steam breakthrough to the extraction wells and the surface. It was anticipated
that aggressive operations (injection and extraction at design rates) would commence
once system problems had been resolved. A detailed timeline is presented in Figure 5.1-
1.
The actual implementation phases were as follows:
Phase ICommissioning and Startup. This phase started October 1, 2002 and
continued until October 29, 2002. It should be noted, though, that not all systems (or
components of systems) were completely installed or commissioned by October 29.
During this phase, sporadic steam injection started (during daylight hours only), and
liquid extraction commenced.
Phase IIContinuous Steam Injection. This phase started on October 30, with more
aggressive steam injection during daytime, while boiler operators became familiar with
equipment. Twenty-four hour operations started on November 7. Phase II continued
until the vapor system and aeration basin failed on December 15. The following points
highlight operating conditions:
Out of 39 potential days of continuous operation, the pilot system was

successfully operated for 20 days, with six days completely shut down and 13
days of partial operations.
The longest period of continuous operation was November 18-23.
Multiple problems were encountered throughout Phase II. The most serious were
caused by material incompatibility with the effluent, resulting in failure of seals
and gaskets in the liquid and vapor conveyance lines. Additionally, waxy and
crystalline deposits of PAH clogged lines from the wellfield to the treatment
plant. Analysis of one sample of crystalline deposit revealed that it consisted of
75% naphthalene and 25% other PAHs. An evaluation of the extracted liquid
composition and the physical/chemical processes that may have occurred to
create this situation is included in Appendix G.
For days when the boiler was operating, steam injection averaged 5,180 lbs/hr
(32% of maximum design rate), with a peak rate of 11,263 lbs/hr on November
9. Injection rates were, in part, limited by uncertainty about whether the effluent
discharge line into Puget Sound would be able to handle the increased non-
contact cooling water flow. Final connections for non-contact cooling water
were completed by December 12, after an engineering analysis determined the
effluent line could handle the flow into the treatment plant.
Liquid extraction averaged 16 gpm (32% of maximum design rate). Most

extraction was from extraction wells E-2, E-3, E-4, E-5, and E-6. Performance
of extraction wells E-1 and E-7 decreased, partly due to crystalline PAH
formation in the pump exhaust ports. Precipitation in exhaust lines was
exacerbated by well head plumbing, which routed the narrow diameter lines
outside the wellhead and allowed rapid cooling on contact with ambient air. In
the case of extraction well E-1, significant precipitation of crystalline PAH was
also occurring inside the well. Lower permeability soils in the western part of
the pilot test area (E-1 and E-7 arrays) may have contributed to problems
encountered in those wells. Low permeability affected both the ability to extract
liquids and to inject steam. Limited steam injection in turn made it difficult to
raise temperatures to the point where precipitation of PAH would be avoided.
Based on discussions with field personnel (during Phase III engineering

evaluations), it is believed that the vapor extraction system was not being
operated as intended. During Phase I, the vapor collectors and extraction wells
were tested with valves in the open position, and there is no evidence from field
data sheets that valve positions were ever altered. Modeling during design
suggests that subsurface vacuum would be insignificant when vapor collector
valves were operated in this manner due to high permeability in the vapor cap.
Optimal valve positioning was to be determined by observation of the vapor
system performance during the pilot study. Since the vapor system was in
operation for a very short time period, these observations were not
accomplished.
Failure of the aeration basin coincided with a dramatic drop in dissolved oxygen
in the basin, which suggests that the bacteria had utilized all of the available
oxygen during consumption of high concentrations of contaminants. Air
diffusers in the basin were unable to provide rapid enough replenishment of
oxygen, causing bacteria to die or go dormant. See Section 6.6.3 for a more
detailed account of the aeration basin failure.
The condensing vapor-to-liquid heat exchanger (HX-3) failed on December 15
after the very narrow passages between the plates plugged with solid
naphthalene crystals and the seals between the plates disintegrated due to
chemical non-compatibility with the waste stream.
The vapor collection system failed on December 15 when one of the liquid ring
vacuum pumps was inundated with liquid. The source of this liquid has not
been confirmed; however, liquid in the vapor line may have come from non-
contact cooling water from the failed heat exchanger, vapor condensate, or rain
water from the vapor collector lines in the well field.
Based on water level collapse observed from December 14 to December 16 (from

an average of 6.1 to 2.1 ft MLLW), the volume of water displaced by steam was
equivalent to 25% of the saturated volume on December 14. Although the
volume calculation was based on simplifying assumptions regarding aquifer
dimensions, temperature sensor data corroborate the conclusion of the analysis.
A value of 24.6% was obtained for the proportion of temperature sensors
registering greater than 100 oC below 6.1 ft MLLW. The calculated
displacement equates to a cylinder (half cylinder for wells along sheet-pile wall)
with a radius of 11 ft around each injection well. Based on interpolated
temperature profiles (see Section 6.2.2), the actual steam distribution around
each well probably approximated an inverted cone, rather than a cylinder.
Average site temperature increased from 19 oC to 71 oC.
Phase IIIEngineering Evaluations. This phase consisted of preliminary evaluations

of the condition of the pilot system and development of potential engineering solutions to
the problems encountered. It lasted from December 16, 2002 to January 12, 2003. Steam
injection was not conducted during this period. To regain hydraulic control, liquid
extraction from the pilot test area started again on January 4, 2003. Average site
temperature decreased to 56 oC. Engineering evaluations and system repairs continued
during Phase IV.
Phase IVLow-Level Steam Injection. This phase was initiated to maintain

subsurface temperatures in anticipation of resolution of engineering problems and
resumption of aggressive thermal operations. Phase IV lasted from January 13 to March
22 and consisted of continuous steam injection at relatively low rates, with liquid
extraction to maintain hydraulic control. Vapor was not extracted during this phase.
Phase IV ended when problems developed with the boiler. Highlights of operations are
as follows:
Out of 69 potential days of continuous operation, the system was successfully
operated for 65 days, with only 4 days of partial operations (including the first
day of this phase).
The primary problems encountered during this phase were a continuation of the
clogging problems in liquid lines and the treatment plant caused by wax and
crystal buildup.
For days when the boiler was operating, steam injection averaged 2,283 lbs/hr
(14% of design rate), with a peak rate of 3,072 lbs/hr on January 13.
Liquid extraction averaged 20 gpm (40% of design rate). Most extraction was
from extraction wells E-2, E-4, E-5, and E-6. Extraction well E-1 was not put
back on-line, because the inside of the well was fouled with crystallized PAH.
Exhaust ports for extraction wells E-3 and E-7 repeatedly fouled with crystalline
PAH, limiting pump performance. When functioning, even low extraction rates
at extraction well E-7 resulted in significant drawdown, indicating limited
recharge to the well.
Average site temperature increased to a maximum of 78 oC on March 5 and then

decreased slightly to 76 oC at the end of the phase.
Phase V Sporadic Low-Level Steam Injection. This phase lasted from March 23 to
April 15, at which time it was decided to discontinue all pilot operations. Problems with
the boiler burner and with maintaining a proper air-fuel mixture in the boiler prevented
continuous operations. Panting in the boiler, believed to be due to improper air-fuel
mixtures, eventually caused welds to crack. Operationally, key points are as follows:
Out of 24 potential days of operation, the boiler was partially operational on seven
days, although steam injection was recorded for just two of those days.
For days when the boiler was operating and steam was being injected, injection
rates averaged 1,834 lbs/hr.
Liquid extraction averaged 14 gpm out of extraction wells E-2, E-3, E-4, E-5, and
E-6. Extraction wells E-1 and E-7 were not turned on. Extraction well E-3
continued to experience exhaust port crystallization problems.
Average site temperature decreased to 68 oC at the end of the phase.
Phase VILiquid Extraction. This phase started April 16 and continues to the present.
Pilot test area wells are operated to the minimal extent necessary to maintain hydraulic
control.
5.2 TEAM COMMUNICATION AND DECISION MAKING
An accelerated approach to sampling, analysis and operational-decision making

was required for this project. Automated data reporting systems as well as fast
turnaround sample analysis and reporting were used to minimize the operations teams
response time needed to adjust system operating parameters. A description of the project
teams roles and responsibilities, as well as the communication tools used, are included in
this section.
5.2.1 Project Team
The project team consisted of representatives from USEPA Region 10, the
USACE Seattle District Office, the USACE Hazardous, Toxic, and Radioactive Waste
Center of Expertise in Omaha, Nebraska, and numerous contractors (Table 5.2-1). The
project team provided the overall framework for the construction, operations,
maintenance and data collection and management activities. Within the project team was
a core technical team (Operations Team) made up of individuals who had developed site-
specific expertise in geologic, hydrologic, chemical analytical methods and operations.
This team was to provide a continual, integrated, and multidisciplinary presence
throughout the project, and was the primary team responsible for the daily decision-
making, optimization of field activities and interactions with USEPA.
The Operations Team was led by a team Coordinator who facilitated the teams
decision-making process. The Operations Team consisted of the following roles:
USEPA Remedial Project Manager
USACE Project Manager
USACE Site Manager Operations Team Coordinator and Health and Safety
Officer
USACE Hydrogeologist
USACE Monitoring Coordinator and QA/QC Officer
O&M Contractor Representative
SteamTech Thermal Expert and other Expert Consultants (on an as-needed basis)
The Operations Team members reviewed project data and convened mid-morning
each workday to review monitoring and process data. The SteamTech expert and other
consultants were asked to participate on a less frequent basis but were expected to keep
current with data on the website. The Team Coordinator provided a summary of the data
and reported system status at the beginning of each daily Operations Team meeting. The
Operations Team then decided on operational objectives for the next 24-hour period.
Once the operational goals for the project were decided, the Operations Team
Coordinator directed the O&M Contractor (SCS Engineers) to implement the decisions.
The decisions and directions provided to the Contractor were documented in a daily
Operations Team Meeting Summary posted to the project website.
Additional support was provided to the Operations Team by project team

technical staff within the USACE-Seattle District and the USACE HTRW Center of
Expertise when needed to resolve reoccurring or consistent issues. For example, when
treatment plant breakdowns impacted operations, a Process Engineer was added to the
Operations Team to help resolve technical problems, provide advice, and assist in making
operational decisions. Technical Support Team members were expected to review
project data on the website on a regular basis and stay current regarding on-site
developments and progress.
Several communication methods and tools were used to ensure that all team
members were aware of current site activities and could participate effectively. Methods
and tools included daily conference calls, meetings, email, and a webpage for posting
data, meeting minutes, and notes from technical discussions. Contents of the project web
page are listed in Table 5.2-2.
SECTION 6.0 SUMMARY OF RESULTS
This section presents the pilot study technical data from the pilot study operations.
In section 6.1, the operational results are presented from a water and energy balance
standpoint. Section 6.2 presents the subsurface temperature monitoring results. In
Section 6.3, the heating observed is compared to heating estimates derived from the
energy balance, including estimates for heat losses. The results of the chemical
monitoring program and the performance of the vapor and water treatment systems are
presented in Sections 6.4 through 6.6. The available estimates for chemical mass
removed from the subsurface are listed in Section 6.7. Calculation methods used during
the data interpretation and graphing process are included as Appendix H. Data collected
during the Pilot Study are included as Appendix I.
6.1 OPERATIONAL RESULTS
Water and energy balance results are presented in this section.
6.1.1 Steam Injection Rates and Totals
Three different lines of data were used to document the actual steam injection
rates from early October 2002 to May 20, 2003:
1. Steam injection rates were measured at each of the 16 injection wells.
2. The diesel usage was converted to an equivalent steam production rate.
3. The water used by the boiler was converted to equivalent steam injection rates.
The estimates are shown on Figure 6.1-1. Generally, the numbers from the steam
rate measurements and the diesel usage agree well. Comparatively, the water usage
seems to underestimate the steam production rate for times where good and frequent data
exists from the well field. The feed water orifice plate was apparently never calibrated
properly, so water usage data have been deemed unreliable. Consequently, where the
data was available, the diesel usage data was used to fill in steam production data for
times where well-field data was missing.
The resulting, probable steam injection rates over the duration of the project are
shown in Figure 6.1-2. Steam injection rates typically varied between 2,000 and 8,000
lbs/hr during operations. The highest rates of around 12,000 lbs/hr were achieved in early
November, during the stage where rapid heating was targeted. After restarting steam
injection in January of 2003, injection rates were between 2,000 and 3,000 lbs/hr, or
approximately half of the rates achieved during heat-up in November-December, 2002.
Figures 6.1-3 and 6.1-4 show the injection rates for the injection periods of 2002
and 2003, respectively. Please note that the scales are different, since the steam injection
rates were higher in 2002 than in 2003. Typical injection rates were in the 100 to 500
lbs/hr range per well during November and December of 2002, until injection was ceased
on December 15, 2002. Injection rates were limited by the ability of the formation to
accept steam. The data show large differences in the injection rates among the wells,
with some wells injecting less than 100 lbs/hr, and others consistently injecting more than
300 lbs/hr.
During 2003, typical injection rates were in the 50 to 300 lbs/hr range per well,
until injection was ceased on April 12, 2003. As for Figure 6.1-3, the data show large
differences in the injection rates among the wells.
Table 6.1-1 lists the typical injection rates for each well during the two major
injection periods. As discussed later, there is a clear trend showing higher injection rates
in the wells located in the eastern half of the pilot test area.
Figure 6.1-5 shows the cumulative amount of steam injected for each of the 16
injection wells, based on the well head steam flow measurements. It should be noted that
there is substantial variation in the injection rates among the wells. Generally, the eastern
wells (I-2, I-7, I-8, I-9, I-10, I-14, and I-15) injected more steam than the western wells
(I-1, I-3, I-4, I-5, I-11, I-12, I-13, and I-16).
6.1.2 Pore Volumes Injected
A total of 993,000 gallons of water was injected in the form of steam. The pilot
test area volume was estimated at approximately 16,400 cubic yards. Using an average
porosity of 28%, this yields a pore volume of 927,000 gallons. The resulting number of
steam pore volumes is:
pvsteam = Vsteam / pvpilot test = 993,000 / 927,000 = 1.1
This is a very modest number of pore volumes, and not a sufficient amount of
steam to heat the entire pilot volume to steam temperature, given the energy removal
taking place by the extraction system, and the heat losses. Goals for other, primarily non-
creosote, sites have targeted between two and five pore volumes of steam. In conclusion,
early termination of this pilot test meant that the volume of steam injected was
approximately half of the amount that would be a typical minimum for this site volume.
6.1.3 Extraction Rates and Totals
The liquid extraction rates based on pump stroke counters are shown in Figure
6.1-6 for each of the seven extraction wells. Because an ideal discharge volume was
assumed for each stroke for all pumps (actual values were different for each well and less
than the ideal), the extraction rates are considered approximate. Extraction wells E-1 and
E-7 had low pumping rates most of the time, and neither pumped liquid after February 4,
2003.
The cumulative liquid extraction for all the seven extraction wells combined is
shown in Figure 6.1-7. This figure reveals the low volumes extracted from extraction
wells E-1, E-3, and E-7, the western-most extraction wells. In addition, extraction wells
E-1 and E-7 are located adjacent to the sheet-pile wall.
The total extraction rate from the seven pilot test extraction wells combined is
shown on Figure 6.1-8. Before the site was heated, extraction rates typically ranged
between 3 and 5 gpm. Then, during the aggressive heating period in November-
December, the extraction rate varied between 10 and 20 gpm, with a typical average rate
of 14 gpm. The pumping rate was maintained in this range, with a few excursions,
during the remainder of operations. In the first half of April 2003, the rate was lowered to
between 5 and 9 gpm.
6.1.4 Observed Water Levels During Operation
The measured water levels in the operational extraction wells are shown in Figure
6.1-9. Figure 6.1-10 shows the water levels measured in instrument string monitoring
locations.
Initial water levels in the pilot test area were around 8 ft MLLW. During the first
aggressive heating period, the water level was generally lowered by between three and 15
ft in the extraction wells (Figure 6.1-9). At the instrument string monitoring locations,
water level drops of between two and eight ft were observed (Figure 6.1-10).
During the cessation of operations between December 15, 2002 and January 3,
2003, a substantial water level rebound was observed in both the extraction wells (where
pumping was ceased) and the monitoring locations between the wells. Rises to above 10
ft MLLW (two ft above the starting water level) were observed in both the extraction
wells and the monitoring locations. The average pilot test area water level rose 6.8 ft
during this period, which corresponds to an average recharge rate of approximately 12
gpm. Sources contributing to recharge of the aquifer are groundwater flow from
upgradient, through the sheet-pile wall, through the aquitard from the lower aquifer, and
infiltrating precipitation through the cap.
A large increase in the water levels occurred in the first half of November 2002.
This water level rise corresponds to a period where the injection rate (in the form of
steam) exceeded the liquid extraction rate. At extraction well E-2, in particular, the water
level rose to 12.9 ft MLLW, which is 0.7 ft above the top of the screen. This rise in water
level coincided with a slug of liquid that hit the vacuum pumps on November 10, 2002.
Although a vacuum of 0.25 atm is insufficient to draw water to the level of the
aboveground conveyance lines, water may have been entrained along with air being
removed by the vapor extraction system (in a slurping effect).
During 2003, water levels were lowered again, but to different levels than in
2002. Less draw-down was observed.
Three extraction wells, E-3, E-4, and E-5, stand out by having much more draw-
down than the other wells and the monitoring locations. These three wells are the only
extraction wells completely surrounded by injection wells, and the only extraction wells
not placed adjacent to the sheet-pile wall. Drawdown is a function of well efficiency,
formation permeability, and injection and extraction rates in each array. Because the
pressure transducers are located in the filter pack, a change in well efficiency is not
believed to be the cause of unusual drawdown. Formation relative permeability with
respect to water may have been affected by development of an oil-phase condensation
bank, thereby impeding flow of groundwater to the extraction wells. Exceeding the
supply of water, by over-extraction, could also have resulted in excessive drawdown.
Although there were clearly times when extraction exceeded injection, it is not certain
whether that was the only factor responsible for drawdown observed in extraction wells
E-3, E-4, and E-5. E-3, in particular, displayed significant drawdown even at very low
extraction rates. [Extraction well E-7 also showed considerable drawdown, but this was
primarily due to lower permeability along the western side of the pilot test area.]
6.1.5 Water Mass Balance
Water fluxes in and out of the pilot test area are shown in Figure 6.1-11. Prior to
the injection of steam, a net extraction rate of between 0 and 5 gpm was maintained.
Then, during the first two weeks of steam injection (between October 29, 2002 and
November 10, 2002), more water was injected as steam than was extracted as water.
Toward the end of this two-week period, water levels were observed to increase
dramatically at nearly all monitoring locations (Figure 6.1-9 and 6.1-10).
Apart from a single day on December 12, 2002, the water balance was maintained
during the rest of the pilot test such that net injection did not occur. Figure 6.1-11 shows
that a typical net extraction rate of between five and 15 gpm was maintained.
The cumulative volume of water injected and extracted is shown in Figure 6.1-12.
Note that the rate and volume of condensate produced (represented by the yellow lines in
Figures 6.1-11 and 6.1-12) were estimated due to the lack of data. Equipment to measure
the flow of condensate was not installed and therefore condensate flow data are not
available.
The net liquid extraction (black line) shows a steady increase during operations,
except during the period from late October to mid November discussed above. A net total
of 2.6 million gallons of water was extracted from the pilot test area (3.6 million gallons
total extracted, 1.0 million gallons injected as steam).
The estimated amount of water that entered the pilot test volume from the outside
was calculated based on the net water extraction and the observed water level changes in
the pilot test area (based on average water levels as presented in Figures 6.1-9 and 6.1-
10).
During the aggressive heating period from October 22, 2002, to November 28,
2002, the water balance indicates that water was leaving the pilot test area. However, the
balance reversed in late November, and during the rest of the pilot test water was entering
at substantial rates from outside the pilot test area.
6.1.6 Energy Balance
Enthalpy fluxes for injection, extracted fluids (steam and heated vapor and water),
estimated heat losses through the surface, bottom, and sheet-piles, and net enthalpy
exchange for the site are shown in Figure 6.1-13.
Injection rates were in the 2 to 8 million BTU/hr range during November-

December 2002, and decreased to between 2 and 3.5 million BTU/hr in 2003. The
enthalpy balance clearly shows that energy is added, and the site is heated, any time
steam was injected. During steam generator shut-down periods, such as the two periods
in late November, the continued extraction leads to the pilot test volume cooling, as the
net enthalpy flux becomes negative.
It is noteworthy that the estimated heat loss through thermal conduction through
the surface, bottom, and sides of the pilot test area are in the 0.6 to 0.9 million BTU/hr
range after the site heated substantially. This is a relatively high enthalpy flux, equal to
an enthalpy flux of between 600 and 900 lbs/hr of steam. After cessation of steam
injection in mid-April, the enthalpy flux became negative, and the pilot test area cooled
accordingly with a net energy loss of about 0.5 to 0.6 million BTU/hr.
Cumulative energy amounts for the same streams are shown in Figure 6.1-14.
This figure also contains an estimated stored energy amount calculated based on the
average temperature measured in the many temperature sensors and an estimate for the
pilot test volume and heat capacity. A total of 9,400 million BTU of steam energy was
injected, with about half of this (4,800 million BTU) injected in 2002. By early-mid
November, increased temperatures were observed in subsurface temperature sensors (as
well as in extracted liquids), and energy removal started becoming significant.
During the period without steam injection from December 15, 2002 to January 13,
2003, a total of 800 million BTU was removed from the pilot test volume. This was
reflected in a similar decrease in the average subsurface temperatures (represented by the
blue line in Figure 6.1-14).
The heat losses through the vapor cap, through the bottom of the site, and through
the sheet-pile wall were estimated based on the observed temperatures near those
boundaries. Figure 6.1-15 shows the estimated cumulative energy losses, broken down in
the three categories. The energy loss through the surface cap is significant, since
relatively high temperature gradients (in the range of 20 C per meter) were observed in
the soils right under the cap. The energy loss through the bottom was lower, as indicated
by a lower temperature gradient between the two deepest temperature sensors (generally
below 8 C per meter). For the sheet-pile wall, the average temperatures of sensors
adjacent to the wall, but inside, was used. The estimate is uncertain, since a temperature
drop-off distance of five ft (1.5 m) was assumed. The data available from instrument
strings outside the sheet-pile wall are not of proper resolution to measure the expected
temperature increases, as the closest instrument strings were roughly 10 ft from the wall.
Overall, an estimated 3,400 million BTU was lost to the surrounding areas by
thermal conduction. This is approximately 36 % of the injected steam energy. In a
following section, a comparison of the heating calculated based on the energy balance to
that observed by the temperature monitoring sensors will be made.
6.2 SUBSURFACE TEMPERATURES ACHIEVED
Subsurface temperature monitoring results are presented in this section.
6.2.1 Extracted Water Temperatures
The temperature of the extracted liquids from each of the seven extraction wells is
shown on Figure 6.2-1. Increased extracted liquid temperatures were apparent around
11/10/02, less than one week after the onset of continuous steam injection. The heating
occurred fastest in extraction wells E-1, E-2, and E-6, until the water recovery rate from
extraction well E-1 diminished a week later.
Generally, the wells that had consistent water flow through December of 2002
heated to temperatures between 90 and 160 oF (E-2 and E-3). The average weighted
temperature of the extracted water rose from the background temperature (57 oF) to 130
o
F, before steam injection was halted on December 15, 2002.
When extraction and injection was resumed in early January of 2003, the
temperatures quickly rose again, and generally kept rising until the end of steam injection
on April 12, 2003. The average temperature of the extracted water had then increased to
160 oF.
After cessation of steam injection, gradual temperature decreases in the extracted

water were observed, and the average temperature fell to around 125 oF in a period of
five weeks, as extraction continued. Extraction well E-6 experienced the sharpest drop in
temperature (>40 oF from late March to late April, 2003). This preferential cooling in the
vicinity of E-6 is consistent with aquifer testing results that indicate potential for
upwelling from the lower aquifer in this area. See Section 6.2.2.1 for additional
discussions about temperature fluctuations at extraction well E-6.
6.2.2 Subsurface Temperature Sensor Data
Subsurface temperature data were collected using thermocouples and the DTS
system. Generally, data were processed once per day. The data were presented as
temperature profiles at individual instrument strings and as vertical sections and pseudo-
horizontal slices through the pilot test area. The slices and sections were obtained from
3-D interpolation of temperature data using the kriging method in the GMS software.
The lower four slices roughly paralleled the northeasterly dip of the aquitard and were
evenly spread (vertically) throughout the aquifer. From bottom to top (from aquitard
through the aquifer), the slices were labeled A through D. A fifth, horizontal slice (Slice
E) was created for the collector layer. Nine vertical sections were taken at various
orientations through the pilot test area (Figure 6.2-2).
To enable direct comparison of data from DTS and thermocouples, an instrument

string containing both systems was installed at T-7. Although the temperature readings
were taken at slightly different vertical spacings1 m (3.28 ft) for DTS, 5 ft (1.52 m)
for thermocouplesthere is excellent overall agreement between profiles (Figure 6.2-3).
However, DTS readings were consistently 3 oC higher than those from thermocouples.
Manual measurements at E-5 verified that the DTS readings were high by 3 oC. The
difference is readily apparent on October 2, 2002, and it persists through the highest
temperatures reached in this instrument string on December 14. No attempt was made to
correct for the difference between the two systems for data calculations or graphical
presentation.
The effect of sporadic operations during the month of Octoberwhile equipment

was being commissioned (Phase I)can be seen as pockets of higher temperature in the
pilot test area (Figure 6.2-4). Because steam injection was limited to relatively low rates
during daylight hours only, most heating was restricted to the area slightly above and
adjacent to the well screens, from mid-aquifer (Slice C) down to the aquitard (Slice A).
With the start of continuous, 24-hr operations in early November 2002, the isolated heat
blooms centered on injection wells started to coalesce. At this point it is apparent that
steam over-ride was causing greater transfer of energy to the vadose zone (Slices C and
D) than lower levels in the aquifer. This observation is borne out by Sections C and F
(Figure 6.2-5), as well as by temperature profiles through individual instrument strings.
During early December, hot spots were developing in the vapor collector layer (Slice E),
and steam was noted coming out of a sheet-pile wall interlock near injection well I-1.
During Phase II, the average pilot test area temperature reached a peak of approximately
71 oC on December 15th, although the base of the aquifer averaged only 52 oC. These
average temperatures are not strictly comparable to those shown on Figure 6.3-1, because
the figure uses all subsurface temperature data (including data from instrument strings
outside the sheet-pile wall and from upgradient), whereas this section limits the
discussion to areas within the sheet-pile wall where steam injection was occurring.
Phase III was marked by slow cooling, as well as by some redistribution of heat
throughout lower parts of the aquifer (Figure 6.2-4). During this period average site
temperature decreased to 56 oC.
Upon resumption of low-level steam injection operations in Phase IV, hot spots
developed within a week or two (Figure 6.2-4). As was observed during Phase II, the hot
spots were concentrated around injection wells, and most of the heat was being
transferred to the upper part of the aquifer and the vadose zone (Slices C and D).
Although sporadic operations continued into April (Phase V), very little steam
was injected. Some cooling can be seen in all five slices for the period of March 24 to
April 7, 2003.
6.2.2.1 Evidence for Aquitard Leakage in the Vicinity of E-6
An evaluation of pumping test data (Appendix C) suggested that there was

communication between the upper and lower aquifers in the vicinity of extraction well E-
6. Temperature data provide support for this hypothesis. On three occasions after abrupt
cessation of steam injection, dramatic cooling was observed in the lower half of
extraction well E-6 (Figure 6.2-6). It is believed that cold water upwelling from the
lower aquifer caused temperatures to drop as much as 50 oC within 24-48 hours. The
change in temperature was not accompanied by a significant change in water level. The
temperature response at extraction well E-6 contrasts with that of extraction well E-4,
which saw little immediate change when steam injection stopped.
6.2.2.2 Vapor Collector Layer Temperatures
Contoured temperature data for the collector layer during the week of December
8, 2002 point to uneven temperature distribution that is possibly related to uneven
distribution of vacuum across the pilot test area. On the west side of the pilot test area, a
cold temperature trough corresponds with the position and alignment of one of the
subsurface horizontal vapor collectors (Figure 6.2-7). In contrast, hot spots along the
eastern and northern side of the site appear to be associated with other horizontal
collectors. Permeability variations across the site may have influenced migration of
vapor into the collector layer. In particular, subsurface soils are recognized as being
more permeable in the east and north than in the west. However, it is unlikely that
permeability is the sole factor responsible for temperature patterns observed, because the
subsurface beneath the cold trough is not different from other areas on the western side of
the site in terms of permeability. There are at least two possible explanations for the
observed patterns.
Scenario One
Vacuum was drawing preferentially from the western side in the collector where
the cold trough is located. Preferential application of vacuum may have been
caused by fouling of horizontal collectors (PAH precipitation) or incorrect
collector valve settings;
Cool air was drawn into the collector on the west, either through holes in the
impermeable membrane or along the join between the membrane and the sheet-
pile wall;
The collectors in the north and east were under little or no vacuum;
The hot spots in the east and north are due to steam override promoted by low
injection rates; and
Association of the hot spots with collectors in the north and east is coincidental.
Scenario Two
Vacuum was drawing preferentially from the eastern and northern side;
Vacuum enhanced vertical movement into and then across the relatively high
permeability materials in the collector layer; and
Little or no vacuum in the west resulted in a stagnant, cool zone in the collector
layer.
When steam injection stopped on December 10, 2002, the effect on temperatures
in the collector layer was pronounced (Figure 6.2-7). The hot spots in the north and east
cooled dramatically, with ten sensors in the collector layer along the eastern side of the
pilot test area registering decreases of 25-73 oC (average 51 oC). Although steam
injection resumed on December 11, 2002, effects were not readily apparent in the
collector layer the following day (when the maximum temperature differential was
registered). The most likely explanation for the drop in temperature is that precipitation
ponded along the eastern side of the pilot test area and then drained into the collector
layer, either through holes in the impermeable membrane or along the join between the
membrane and the sheet-pile wall.
After the final cessation of steam injection on December 15, 2002, marking the
end of Phase II of operations, collector layer temperatures in the east and north again
declined sharply (Figure 6.2-4). Figure 6.2-8 shows the response of one collector layer
thermocouple (T-52) along with precipitation recorded at SeaTac airport. The preceding
six days had experienced measurable precipitation, and heavy rains occurred on
December 14, 2002. With multiple variables potentially influencing collector layer
temperatures (namely, steam injection, vacuum in the vapor collector, and precipitation),
it is difficult to establish cause-and-effect for December observations. However, support
for the second scenario is seen in temperature and precipitation data from March 2003.
The month of March was selected because significant rainfall events occurred, steam
injection rates were fairly constant, and the vapor system was not running. On March 12,
2003, the top sensor in instrument string T-52 experienced a 26 oC drop (Figure 6.2-8).
The temperature drop coincided with significant rainfall on the sixth consecutive day of
rain. Similarly, on March 21 the same sensor registered a further 15 oC drop, also
corresponding to a period of significant rainfall. Given consistency of operations during
these two periods, it appears that temperature variations in some areas of the cap were
affected by infiltration of water. The degree of temperature drop seems most consistent
with contact with cool water and not with conduction of cooler temperatures through the
cap.
Generally speaking, the drop in temperature during December is a phenomenon

restricted to the eastern and northern side of the pilot test area (Figure 6.2-9). In March
2003, however, the collector layer sensors exhibiting the greatest (>20 oC) temperature
change were scattered across the pilot test area. The discrepancy in sensor response
between December 2002 and March 2003 may indicate different causal relationships or
may indicate that the cap became progressively more permeable. It is worth noting that
steam emissions through the cap were not reported, which suggests that they were
successfully drawn off by the vacuum system (when functioning) or that leaks were small
enough that the steam condensed immediately after breaking through the membrane.
6.2.2.3 Heat Distribution Along the Inside of the Sheet-Pile Wall
Although it is impossible to predict how thorough NAPL removal would have

been along the inside of the sheet-pile wall, temperature data show that that area was
heated as effectively as the central part of the pilot test area (Figures 6.2-10 to -13). In
general, temperature profiles are similar for instrument strings along the sheet pile wall
and for the extraction well E-4 array in the center of the site.
6.2.2.4 Heat Transfer through the Sheet-Pile Wall
Conductive transfer of heat through the sheet-pile wall becomes noticeable by the
end of the first week of December, and maximum temperatures of slightly greater than 25
o
C (average 21 oC) are achieved by the end of March or early April at T-74, -76, and -81.
These instrument strings are located approximately 10 ft outside the sheet-pile wall.
Prior to commencement of operations, the average temperature at T-74, -76, and -81 was
14 oC.
6.2.2.5 Subsurface Heterogeneities
Heterogeneities in subsurface soil types are a feature of the Wyckoff site. The
pilot test area was selected in part to establish the ability of thermal remediation to clean
up all soil types affected by creosote contamination. There are some indications that
heterogeneity influenced heat propagation. At instrument string T-27, hot water and
steam were able to move above and below a silt and silty sand layer from 8-12.5 ft
MLLW (Figure 6.2-12). A temperature differential of 40 oC between the low
permeability layer and the zones immediately above and below persisted through May
2003 (Figure 6.2-13).
A clay layer from approximately 1-5 ft MLLW at instrument string T-56 clearly
influenced temperature distribution (Figure 6.2-10). Unlike T-27, though, temperature
was evenly distributed through that layer by February 2003, and the entire interval
achieved a temperature of 100 oC in March 2003 (Figure 6.2-11). Evidence from T-56
suggests that it is possible to thoroughly heat even low permeability zones.
Instrument string T-29 appears to show similar geologic control (at 2.5 ft
MLLW) on temperature distribution to that of T-27 (Figures 6.2-12 and -13). However,
the boring log indicates that the entire interval in question consists of poorly sorted sand
with gravel. Evidence for a low permeability layer does not exist, although less-than-
complete recovery during drilling may have obscured real soil heterogeneities.
The different responses over time at T-27 and -56 may reflect overall permeability
differences in the western and central parts of the site. Conductive heat transfer may
have been more important in the lower permeability western area, which allowed slower,
more even heating with time. In contrast, higher permeability areas in other parts of the
pilot test area experienced fairly rapid temperature changes, with lower permeability
interbeds lagging significantly behind.
6.2.2.6 Effectiveness of Low-Rate Steam Injection
During the period from January to March 2003, which was characterized by low-
level steam injection, there was only minor improvement in the ability to bring hot
temperatures down to the aquitard surface. As a consequence of the low injection and
extraction rates, most of the injected steam condensed rapidly, and the majority of the
affected zones were swept by hot water (not steam). The low injection and extraction
rates employed were not effective at improving temperature distribution through the
aquifer. This point is illustrated by a comparison of temperature profiles in Sections C
and F for December 10 and March 17 (Figure 6.2-5). These two dates represent the times
closest to maximum heating during Phase II (continuous steam injection) and Phase IV
(low-level steam injection). While there is noticeably better distribution of temperature
at the base of the aquifer on March 17 than December 10 (Figure 6.2-4), some of that
evening out of temperatures is the result of conductive heat transfer throughout the
aquifer during Phase III. It is clear that low-level steam injection exacerbated the
problem of steam over-ride, as exhibited in Slice D. By contrast, injection at design rates
would more likely have created greater pressure differentials, thereby resulting in better
penetration of steam into pore spaces and, presumably, better sweep across the aquitard.
6.3 COMPARISON OF PROCESS DATA TO SUBSURFACE HEATING

OBSERVATIONS
In this section, the heating observed in the subsurface is compared to heating

estimates derived from the energy balance, including estimates for heat losses.
6.3.1 Average Temperatures Used for Energy Balance and Heat Loss
Calculations
Based on subsurface sensors, the average temperatures were calculated for the
following selections:
All pilot test area sensors.
The shallowest sensors in each monitoring location, excluding the sensors in

injection wells (these are relatively deep and do not represent the temperature near
the vapor cap).
The deepest sensors located near or in the top of the aquitard (monitoring
locations, extraction wells, and injection wells).
The sensors located next to the sheet-pile wall (17 instrument strings total:
injection wells I-1, I-2, I-3, I-6, I-16, extraction wells E-1, E-2, E-6, and E-7, and
temperature monitoring locations T1, T4, T7, T10, T49, T52, T53, and T56).
The average temperatures were calculated for each day of operation (Figure 6.3-
1). The measured temperature averages were consistently less than 75 oC, which shows
that the operational target of boiling temperatures (100 oC at the water table and higher
below) was never reached. The data also clearly shows that the bottom sensors stayed
cooler than average, with the average temperature peaking around 52 oC in March of
2003.
The temperatures next to the sheet-pile wall were generally above the average
temperatures, indicating that heating along the sheet-pile was not hindered compared to
the overall average heating. However, this data is partially skewed towards higher
temperatures, since five out of the 17 temperature monitoring locations along the wall
were injection wells. This is a larger fraction of injection wells than the average (16
injection wells, 97 total inside monitoring locations). Since each injection well has only
three sensors, this effect is minor, and it can be concluded that the sheet-pile wall did not
hinder steam flow or heating next to, and inside the wall.
6.3.2 Comparison of Observed Heating to Energy Balance Calculations
Based on the energy balance, average obtained temperatures for the pilot test
volume were calculated and compared to the measured temperatures (Figure 6.3-2).
The measured and calculated temperatures match relatively well. However, it

should be kept in mind that the energy removal from steam being extracted, and the
energy losses by thermal conduction, were estimated based on relatively sparse data, and
based on several assumptions. Therefore, the good match of the data could be somewhat
coincidental. However, overall it may be concluded that the energy balance calculations
and the observed subsurface temperature changes are in agreement, and that no apparent
contradictions exist in the data set.
It is interesting to note that the injection and extraction rates varied systematically
across the site, with much higher flow rates in the eastern half of the site. The cumulative
amounts of water extracted for each well is shown in Figure 6.3-3. Figure 6.3-4 shows the
average steam injection rate from October 30, 2002, to December 14, 2002, the period
where steam was injected at rates as high as practical. Both figures are consistent with
soil borings indicating that the formation was tighter in the western section of the pilot
test area.
6.4 CHEMICAL MONITORING RESULTS
Chemical monitoring results for extracted liquid, non-condensable gases, and

groundwater monitoring wells are discussed in this section.
6.4.1 Extracted Liquid Composition
Extraction well PAH data are summarized in Figures 6.4-1 and 6.4-2. The work
plan called for up to twice daily sampling of the extraction wells. Samples were collected
daily during steam injection from all extraction wells that were operational. Naphthalene
is by far the most abundant PAH in the effluent samples.
Considerable difficulties were encountered with naphthalene crystals plugging the

pump exhausts, thus requiring frequent maintenance. This appears to be particularly true
for extraction wells E-1 and E-7 (Figure 3.5-1), both of which are located in the western
portion of the pilot test area and within the area where extremely high naphthalene
concentrations were found. These wells had low pumping rates most of the time.
Extraction well E-1 did not pump after November 21, 2002, and extraction well E-7 was
not used after February 4, 2003. Later in the pilot operations, it appears that naphthalene
crystallization problems were also encountered in extraction well E-3. This extraction
well is right in the center of the area with high naphthalene content. Thus, some of the
lack of effluent concentration data is because the pumps were undergoing maintenance
and were out of service. At other times, a well was operating and a sample was obtained
but was not submitted for analysis due to the presence of a visible NAPL layer in the
sample. However, it appears that samples containing NAPL were analyzed on several
occasions (see data for E-1 on 11/15/02, E-3 on 2/4/03 and 2/17/03, and E-7 on 1/16/03).
Although soil sampling in the immediate vicinity of extraction well E-5 had
shown that the area around this well contained small amounts of contamination,
significant amounts of NAPL were present within the E-5 array to the east and northeast.
Initial effluent samples from this well showed little or no PAHs for the first month of
operations. After that time, contaminant extraction rates from that well were significant,
and they remained similar to the other extraction wells for the rest of the pilot study.
Presumably, contaminants along the eastern part of the array were mobilized and driven
inward (away from the injection wells) and extracted at E-5.
At about the time that the steam zones appear to extend from the injection wells to
the extraction wells over some depth intervals in the northern portion of the site, which
occurred around December 10, the concentrations in the effluent samples increased. For
E-2, the increase in concentration was approximately an order of magnitude, while for
wells E-3, E-4, and E-6, the increase in concentrations was closer to two orders of
magnitude. E-3 appears to have consistently the highest PAH concentrations in the
effluent, however, the extraction rate from this well was always low. E-5 exhibited an
even greater relative increase in concentrations because its initial concentration was so
low. These increases in concentration are expected, as the solubility of PAHs generally
increases exponentially with temperature (Bamford et. al, 1998).
PCP concentrations in the effluent are shown in Figures 6.4-3 and 6.4-4. These
figures show that most of the PCP extracted came from well E-2, where concentrations
went as high as 1,900 g/L. E-2 is at the northern-most portion of the pilot test area,
where there were not much soils data available from the baseline sampling. Extraction
well E-6 on the eastern portion of the pilot test area also showed significant PCP
concentrations during the period of low-level steam injection (Phase IV).
6.4.2 Extracted Liquid Total Organic Carbon Results
The TOC content of the liquid effluent stream from the pilot test area was
measured automatically with a frequency of every 20 minutes for most of the period of
operations. Frequency changed to every hour in March 2003. Figure 6.4-5 displays the
average daily TOC reading and cumulative TOC removed for the period from November
7, 2002 to April 10, 2003. The TOC analyzer was not operational until November 9,
2002, so data are not available for the period of limited injection and extraction during
Phase I and the beginning of Phase II.
With the start of 24-hour injection and extraction on November 7 (Phase II), the
TOC concentration of the effluent started gradually increasing. Within 3 to 4 weeks,
TOC had doubled to approximately 500 mg/L. By December 15, the average daily TOC
concentration was approximately 1,900 mg/L. Spikes of greater than 1,000 mg/L became
common on November 10, followed by spikes of greater than 3,000 mg/L on December
3. The highest spike recorded was 10,000 mg/L (out of range) on December 14. During
Phase II, the cumulative amount of carbon removed was estimated as 2,600 kg, which
equates to an average of 67 kg/d. Assuming a simple 1:1 relationship between mass of
carbon and mass of creosote, with no corrections for density or other elements that
contribute to contaminant mass, the creosote equivalents of 2,600 kg and 67 kg/d are 680
gal and 18 gpd, respectively. Given the simplifying assumptions, these figures probably
underestimate the amount of creosote removed in the dissolved phase.
Phase III saw very little extraction, during which TOC was estimated based on
values for December 16 and January 20. Total extracted during Phase III was estimated
as 170 kg (44 gal creosote).
Phase IV was characterized by less fluctuation in TOC than was observed during
Phase II. Average daily TOC values ranged between approximately 500 and 1,400 mg/L,
with spikes over 2,000 mg/L occurring relatively infrequently (roughly 1% of all TOC
readings during this phase). Greater consistency in TOC concentrations is presumably a
reflection of fairly constant subsurface temperatures. The total carbon removed during
this phase was estimated as 6,300 kg (1,700 gal creosote equivalent) or approximately 91
kg/d (24 gpd creosote). Although contaminants were apparently being removed at higher
average rates than during Phase II, it is worth noting that peak removal during Phase II
was equivalent to roughly 250 kg/d (66 gpd creosote).
During Phase V, when steam injection was sporadic and limited, TOC
concentrations averaged approximately 570 mg/L (45 kg/d or 12 gpd creosote). Spikes
over 1,000 mg/L were rare, and none were recorded over 2,000 mg/L. Total removed
during this phase was estimated as 850 kg (220 gal creosote).
For the period from November 7, 2002 to April 10, 2003, approximately 9,300 kg
of carbon were removed from the pilot test area in the dissolved phase. This mass
removed is roughly equivalent to 2,600 gal of creosote.
6.4.3 Extracted Vapor Composition
Non-condensable gas monitoring was planned as part of the pilot study. Vapor
sampling equipment was not installed between the heat exchanger and the vapor
treatment system as planned for in the design; therefore, chemical data for the vapor
phase are not available for analysis.
6.4.4 Monitoring Well Sampling Results
Groundwater monitoring was conducted of both upper and lower aquifer

monitoring wells to evaluate potential impacts from the pilot study. Samples were
collected from both aquifers in November 2002. Additional samples were collected from
the lower aquifer in December 2002 and January 2003. Locations of the monitoring
wells are shown in Figure 3.5-1.
6.4.4.1 Upper Aquifer Monitoring Results
During November 2002, groundwater samples were obtained from the upper
aquifer wells MW-17, MW-18, and MW-19, and two extraction wells within the pilot test
area, E-04, and E-06. PAH were detected in all of these wells. Concentrations of PAH
were greatest in MW-17 and MW-18 located outside the pilot test area. Concentrations
of several compounds, including naphthalene, exceeded groundwater cleanup levels.
Concentrations of PAH in MW-19, which is located southwest (upgradient) of the pilot
test area, were significantly less; however, the concentrations of carcinogenic PAH
exceeded groundwater cleanup levels.
6.4.4.2 Lower Aquifer Monitoring Results
During December 2002 and January 2003, groundwater samples were obtained
from the lower aquifer monitoring wells 99CD-MW02, 99CD-MW04, CW-05, CW-09,
and CW-15 (Figure 3.5-1). The December 2002 sampling round showed very few
detections of PAHs in any of these monitoring wells. Only well CW-15 had contaminant
concentrations that were greater than the cleanup levels. In the January sampling round,
the concentrations in all the wells except for CW-09 returned to approximately the level
they had been in the November 2002 sampling round, and some constituents went to
higher concentrations. CW-09, 99CD-MW02, and 99CD-MW-04 still were less than
cleanup levels for all constituents, however, CW-15 and CW-5 had concentrations that
were greater than cleanup levels.
In January 2003, monitoring well 02CD-MW01 was also sampled. This well was
installed for the purposes of monitoring the lower aquifer just outside the pilot test area to
the northeast, in an area where there was concern about the integrity of the aquitard
(Appendix J). This was the only time that this well was sampled. It contained small
amounts of various PAHs, including chrysene and benzo(b)fluoranthene that were greater
than the cleanup levels.
Data from these monitoring wells during the first three months of the pilot would
suggest that groundwater quality in the lower aquifer was not adversely affected by the
pilot operation.
6.4.5 Chemical Data Quality
A review of laboratory data for extracted liquid or groundwater samples included

a review of laboratory performance criteria and sample-specific criteria as specified in the
RAMP. Results were evaluated to determine if the measurement quality objectives for
the project were met. The data were determined to be acceptable for project uses as
qualified. No data were rejected.
Field measurements were not subject to a formal data quality review. However,
dissolved oxygen and carbon dioxide results from the daily sampling of the extraction
wells were rejected due to the following issues:
The sampling ports used to collect samples from the extraction wells did
not allow for low flow sampling methods to be used, resulting in non-
representative samples.
The pumps installed in the extraction wells (Hammerhead) allowed air

from the compressor to mix with the extracted liquid prior to and during
sampling, resulting in non-representative samples.
The sampling ports allowed mixing of ambient air with groundwater

samples, potentially resulting in non-representative samples.
The light to dark brown color of the extracted liquid interfered with the
color interpretation of the CHEMetrics colorimetric field test kit.
6.5 VAPOR TREATMENT SYSTEM DATA SUMMARY
Non-condensable gases were treated by running them through the burners of the
gas-fired boiler. For regulatory compliance, boiler emissions monitoring was planned.
The pilot monitoring plan contained provisions for emissions stack testing when boiler
operations had stabilized and the sour gas to be burned in the boiler had reached the
standard operating conditions, but the contract and equipment were not in place prior to
start of operations. In addition, the steady-state operational conditions that were to be
reached prior to stack sampling never occurred. Therefore, chemical data for stack
emissions are not available for analysis. Boiler emissions testing to determine maximum
potential emissions was predicated on operation of the boiler full time at capacity.
Because the boiler never achieved full time operations at capacity, testing would not have
been relevant to address regulatory compliance.
6.6 WATER TREATMENT SYSTEM DATA SUMMARY
The water treatment system consists of a series of processes designed to remove

contaminants from the combined liquid waste streams from the former process area and
pilot area (Figure 4.2-9). The main components include the DAFT, the aeration basin,
and three activated carbon vessels. Treated water is discharged to Puget Sound. The
effectiveness of each system in removing contaminants from the waste stream is
discussed in this section. Treatment plant data are summarized in Table 6.6-1.
6.6.1 Extracted Liquid Waste Stream
The flow of combined extracted liquid from the former process area and the pilot
area stayed fairly constant during the pilot study at around 40 to 50 gpm (approximately
30 to 35 gpm from the FPA and 10 to 15 gpm from the pilot test area). Prior to the start
of the pilot study (January through September 2002), the average total PAH and PCP
concentrations in water from the FPA that went into the treatment plant were 15,500 and
860 g/L, respectively. Total PAH and PCP concentrations in treatment plant influent
from January through December 2002 are shown in Figure 6.6-1. Concentrations of
contaminants going into the treatment plant increased dramatically beginning in
November 2002 and stayed relatively high through April 2003 when the pilot study
operations ceased. Concentrations of total PAH, naphthalene, and PCP in treatment plant
influent (sampling location SP-0) are shown in Figures 6.6-2 through 6.6-4. Influent
concentrations peaked during the weeks between December 2 and December 15, 2002,
overloading the capacity of the treatment system and contributing to the decision to stop
aggressive steam injection operations.
6.6.2 Dissolved Air Floatation Tank
The DAFT is designed to remove contaminants in the non-aqueous or oil phase

by floating the separate phase to the surface with air bubbles and removing it to a
separate holding tank (tank 108). Only a small volume of separate phase NAPL was
removed by the DAFT during the pilot study. Crystals of naphthalene began clogging the
DAFT in January 2003 and additions of a combination of chemical flocculent and
polymer were used to reduce this problem beginning routinely in March 2003. A sample
of this material was analyzed for PAH by the Region 10 laboratory. The sample was 75
percent naphthalene, with the remaining 25 percent identified as various other PAH. The
effectiveness of the DAFT in removing total PAH, naphthalene, and PCP are shown in
Figures 6.6-2 through 6.6-4.
Samples of the DAFT influent and effluent were collected once a week during
pilot operations and results are at times inconsistent and highly variable. Total PAH and
PCP concentrations in DAFT effluent are at times greater than concentrations of the
DAFT influent. Concentrations also fluctuate from week to week. This may be due to
the heterogeneous nature of the samples (SP-0 is collected from a pipeline and SP-4 is
collected from the well-mixed tank effluent line), allowing either NAPL sheen or
droplets, or naphthalene crystals, to impact the samples.
The amount of NAPL collected from the site from December 1998 through June
of 2003 is shown in Figure 6.6-5. The bulk of the NAPL removed from the site has been
from accumulations in sumps in extraction wells in the former process area. Much less
has traditionally been separated from the liquid waste stream running through the
treatment plant and separated by the DAFT (formerly by the depurator). This trend
continued during the pilot study. A detailed analysis of the mass of contaminants
removed during the pilot study is included in Section 6.7.
6.6.3 Aeration Basin
The aeration basin is the primary treatment method for the liquid waste stream. In
the aeration basin, site-specific microorganisms biodegrade site contaminants.
Concentrations of total PAH, naphthalene, and PCP going into and out of the aeration
basin are shown in Figures 6.6-6 through 6.6-9. Concentrations are typically reduced by
up to 99 percent within the aeration basin. During the weeks from December 2 through
December 15, 2002, the concentrations of contaminants coming into the aeration basin
peaked and overloaded the capacity of the biological system. The dissolved oxygen and
total PAH concentrations in the aeration basin are shown in Figure 6.6-10.
During the week of December 2, the aeration basin showed effects of higher
concentrations of PAHs and the clarifier visibility decreased. On December 8, a polymer
ball formed in the static mixing chamber between the aeration basin and the clarifier.
The ball plugged the mixing chamber and caused the aeration basin to overflow with
liquid. The liquid flowed into the aerobic digester, causing solids to mix into the liquid.
The liquid then flowed back into the aeration basin, loading the basin with digester solids.
Basin monitoring revealed very high concentrations of PAHs and low dissolved oxygen
values.
After the mixer plugging problem occurred, examination of the aeration basin
microorganism life revealed very little activity. The aeration basin was taken off line
(from December 8 to December 11), put in a closed loop operational mode, and
inoculated with backwash tank solids to rehabilitate it. Solids from the mixing chamber
plug problem also migrated to the multimedia filters and carbon vessels. The filters and
vessels were backwashed several times to remove the solids. Effluent was not allowed to
discharge to Puget Sound until all systems had been backwashed and were operational.
On December 15, a nearly complete die-off of aeration basin microorganisms

occurred due to high concentrations of PAH in influent from the pilot test area.
Dissolved oxygen concentrations dropped to approximately 1 mg/L and the treatment
plant was put into recycle mode with no discharge to Puget Sound. The basin was again
re-inoculated with backwash tank solids to rehabilitate it before it was put back on line on
December 18.
Following this series of events, dissolved oxygen concentrations were maintained

above 2 mg/L by switching on the additional air blower to restore or increase values.
Increased monitoring of dissolved oxygen concentrations (once every four hours) began
in early January 2003 and continued until additional air diffusers were added later that
month.
The ability of the aeration basin to handle peak contaminant loading was
considered suspect throughout the design process. From a conceptual standpoint, failure
of the aeration basin was to be dealt with by diverting flow through the carbon vessels.
In terms of implementation, though, funding was limited for additional carbon vessels,
and the existing setup made changing carbon laborious. Prior to early December, the
aeration basin handled increased contaminant loading exceedingly well. Following the
installation of additional air diffusers, dissolved oxygen levels within the basin became
much more stable and the system treatment capacity was increased. Ultimately, though,
this limit of treatment capacity by the aeration basin was one of the major determining
factors in ending the pilot study.
6.6.4 Clarifiers, Multimedia Filter, and Carbon Treatment System
From the aeration basin, water flows through a clarifier and is pumped in parallel
through two multimedia filters simultaneously with split flows before being transferred to
two 8,000 lb granulated activated carbon filters, arranged in series. (A third carbon filter
tank was put on line in May 2003.) The effectiveness of the carbon system in removing
total PAH and PCP is shown in Figures 6.6-11 through 6.6-13. The carbon system is
highly effective at reducing contaminant concentrations to less than the surface water
discharge limits for the site.
6.6.5 Compliance Monitoring for Treatment Plant Effluent
Treatment plant effluent must meet specific limits set in the ROD prior to
discharge to Puget Sound. Effluent is monitored for PAH, PCP, pH, temperature,
dissolved oxygen, and solids (total and dissolved). Discharge limits for total PAH and
PCP were not exceeded during the pilot study (Figure 6.6-14). Discharge limits for pH
and temperature were also not exceeded during the pilot study (Figures 6.6-15 and 6.6-
16). Dissolved oxygen concentrations did not meet the discharge limit set in the ROD
(>6 mg/L) (Figure 6.6-16). Total and dissolved solids were also measured in treatment
plant effluent, however these results, measured in mg/L, could not be compared to
discharge limits set in the ROD for turbidity, measured in NTU.
6.7 PILOT STUDY MASS REMOVAL ESTIMATES
The mass of contaminants removed from the subsurface could be measured as the
following four quantities:
1. NAPL recovery. The rate of NAPL removal from the pilot area was not measured,
since the effluents were combined with the liquids from the other site extraction
wells located in the Former Process Area. The data on overall NAPL removal
over the project period was used to determine if more than usual was recovered in
the period of the pilot study. These data are shown on Figure 6.6-5. This simple
analysis indicates that compared to the trend before steam injection was begun, an
approximate volume of 340 gal of NAPL was recovered over the six-month
operational period due to the thermal enhancements.
It should be noted that NAPL removal was tracked through measurement of the
level of liquid collected in tank 108. This methodology was inaccurate, because
tank 108 collected NAPL from the combined waste streams from the FPA and the
pilot test area; therefore, an estimate of the amount removed due to thermal
treatment of the pilot test area alone could not be made.
2. Removal of dissolved phase contaminants. The TOC data, when combined with
the liquid extraction rate data, provides a trend and allows for an estimate of the
mass removed in this phase. This estimate yields a total volume equivalent of
2,600 gal (~9,300 kg) of organics.
3. Removal in non-condensable vapor: No analytical data were recorded for the

vapor stream. A mass estimate would have to rely on assumed flow rates and
concentrations. Since the naphthalene concentrations were very high, leading to
both crystallization in the well and pipes, and formation of waxes in the process
equipment, it is likely that a substantial mass was unaccounted for in this stream.
However, without chemical sampling of the stream, calculation of amount
removed is not possible.
4. In-situ destruction/degradation. This could be due to oxidation, hydrolysis,

pyrolysis, and be either biological or chemical reactions. The importance of such
mass removal could not be evaluated due to the complexity of such
measurements.
In conclusion, the volumes and masses recovered from the pilot test area over the
6 months of operations can be estimated as follows:
Recovered as NAPL: 340 gal (~1,300 kg).
Recovered as dissolved contaminants: 2,600 gal (~9,300 kg).
Recovered in vapors: Unknown.
Degraded in-situ: Unknown.
Total: 2,900 gal (~11,000 kg) plus an unknown amount.
From other thermal remediation sites, such as the Visalia Pole Yard, it is indicated that
the removal in the NAPL and vapor state are substantial contributors to the mass
removal. Thus, it is likely that the true mass removal is substantially higher than the
quantity listed above.
SECTION 7.0 COST SUMMARY
This section includes the costs for USACE design, installation of sheet pile walls,
installation of subsurface wells and vapor cap, installation of thermal remediation and
subsurface monitoring equipment, actual operation and maintenance expenses for
operation of the thermal processes, USACE construction oversight, and USACE
oversight of the operations and maintenance. The following table represents actual
expenditures through April 15, 2003.
Contractor Description of Work Costs

BayWest/JD Installation of inner sheet pile wall $1,400,000
Fields/Hurlen
Construction
Boiler building construction, installation of $1,830,000
Marine Vacuum, Inc.
vapor cap and vapor collection piping,
installation of the water supply well,
extraction wells, and injection wells
Pease Construction, Treatment plant upgrades, installation of $2,414,000
Inc. the boiler and above ground piping system,
startup and commissioning of the thermal
treatment system
Operations and maintenance of site during $1,193,000
SCS Engineers, Inc.
steaming operations
$140,000
Geomation Installation of thermocouples and
automated data system
Installation of distributed temperature $145,000
Sensor Highway
sensor (DTS) system well field
(Sensa)
Water quality and air monitoring $183,000
URS Corporation
$16,000
Shannon and Wilson Installation of subsurface instruments (nine
replacement borings)
Fuel supplier $59,000
Williams Heating Oil
Contractor SubTotal $7,380,000
Contractor Description of Work Costs
Pre-award of Contracts Design and $1,500,000
USACE
Planning
Construction Management and Oversight $870,000
USACE
USACE SubTotal $2,370,000
TOTAL (as of summer 2004) $9,750,000
Additional costs for the disposal of product, filter cake, and carbon that would
have been required during the pilot study were negotiated in the operation and
maintenance contract with SCS Engineers, which included the following line items for
waste disposal:
Line Item 15 - Spent carbon removal from storage, transfer to roll-off

containers, and transport for disposal in landfill. Unit price is $0.72 per lb.
Line Item 16 - Spent carbon removal from storage, transfer to roll-off

containers, and transport for disposal as F032/F034 Listed Waste for
incineration. Unit price is $1.84 per lb.
Line Item 19AA - Hazardous Waste disposal (Incineration) of filter cake

and other non-NAPL items. Unit price is $1.06 per lb.
Line Item 19AC - Hazardous Waste disposal (Incineration) of product

(NAPL). Unit price is $2.05 per lb.
Due to the low volume of product removed during the pilot study operation and
maintenance period, there were no expenditures for waste disposal from January 1, 2002
through April 15, 2003.
SECTION 8.0 CONCLUSIONS
The two goals of the pilot study were to assess the likelihood that a full-scale
thermal remediation would achieve the cleanup goals for the site and to provide
information for implementation of the potential full-scale thermal remediation. The pilot
study proceeded for approximately six months, from October 2002 to April 2003.
Operations were restricted by equipment problems, and the pilot study was terminated.
Limited progress was made towards achieving the specific objectives of the pilot study
relating to performance assessment, community and environmental impacts, and process.
A significant amount of data was collected during the five months of operations, and
important observations regarding system performance, impacts on the surrounding
environment, and in process design and operations were made that may be useful in
designing and implementing a new pilot study or full-scale remediation, if needed for the
Wyckoff site, or may be useful to others for implementing thermal treatment systems at
other sites.
8.1 PERFORMANCE ASSESSMENT
The pilot study was performed partly to determine the effectiveness of steam-
enhanced extraction to remove mobile NAPL and reduce contaminant concentrations in
groundwater and soil to acceptable risk levels. Data gathered during the pilot study show
that the mass removal rate increased dramatically when the site started heating.
Dissolved PAH and TOC concentrations in extracted liquids increased, and the vapor
extraction system experienced problems due to fouling and precipitation related to
extremely high contaminant concentrations in the vapor phase. However, the mass
removal that occurred during the pilot study was minor compared with what could have
been achieved with a fully functional SEE and treatment system.
The mass of contaminants removed from the subsurface was estimated as follows:
Recovered as NAPL: 340 gal.
Recovered in the dissolved phase: 2,600 gal.
Recovered as non-condensable vapor: The observation of naphthalene crystals in

the vapor lines suggest that significant contamination was removed in the vapor
phase; however some of the planned monitoring instruments were not installed
when the vapor system was in operation, and accurate vapor flow and
concentration measurements were not made.
In-situ destruction/degradation. Based on laboratory results obtained as part of

the Treatability Study, some oxidative destruction of the creosote could have been
expected to occur during the pilot study. However, adequate sampling protocols
were not in place during the pilot to measure the amount of oxidation that
occurred.
The total estimate for the amount recovered is 2,940 gal in the NAPL and
dissolved phase plus an unknown amount in the vapor phase or degraded in-situ. Prior to
the pilot, the average amount of NAPL extracted per month was approximately 320
gallons with an average of approximately 24 gallons per month in the dissolved phase.
Though the amount of NAPL removed did not show a marked increase during the pilot
study, the amount of contaminants removed in the dissolved phase increased
dramatically.
For comparison, steam injection was used by Southern California Edison to

recover creosote and pentachlorophenol at their Visalia Pole Yard in Visalia, California.
More than 1,200,000 pounds of creosote were recovered by the injection of
approximately 700 million pounds of steam over a three year period. This was a 3500-
fold increase in the extraction rate compared to the existing pump-and-treat system that
had been operating since 1975 and recovering approximately 10 pounds of creosote per
week. Most of the creosote was recovered as a separate phase emulsion, with significant
quantities also recovered in the vapor phase and dissolved in the groundwater. Carbon
dioxide in the extracted vapors indicated that some creosote was being oxidized in situ.
8.1.1 System Performance
Numerous problems were encountered during operation of the thermal

remediation system that contributed by varying degrees to the failure of the system to
perform as designed. These include:
Material incompatibility: Seals and gaskets in pumps, heat exchangers,

and other components failed, which required frequent attention and
replacement of parts, and led ultimately to the complete failure of HX-3,
the liquid-vapor heat exchanger.
Aeration basin adaptability: The activated sludge treatment system failed
to handle the contaminant load, either due to toxicity or lack of oxygen.
PAH precipitation/condensation: The formation of various waxes and

crystalline precipitates clogged valves and pipes, thereby limiting the
performance of extraction wells and restricting flow through liquid and
vapor conveyance lines.
Liquid in vapor lines: Condensation points and drip legs were inadequate
to handle the amount of liquid in the vapor lines, which led to slugs of
liquid reaching the liquid ring vacuum pumps. The liquid most likely
came from rain penetrating the membrane into the vapor collectors and
was transported along the vapor extraction pipeline to the liquid ring
vacuum pumps.
8.2 COMMUNITY AND ENVIRONMENTAL IMPACTS
Monitoring for impacts to the surrounding community and environment, including

noise, air emissions, odors, and off-site migration of contaminated groundwater, was
included in the pilot study design. Monitoring for noise, air emissions, and odors to the
surrounding community were, however, not completed because the necessary equipment
was not installed early enough during pilot study operations.
Upper and lower aquifer groundwater samples were collected and data were
reviewed. No significant changes to groundwater concentrations outside of the pilot test
area were observed that would indicate steam injection in the pilot study area was
adversely impacting site groundwater or mobilizing contaminants beyond site boundaries.
During the short period of time that vapors were passed through the combustion
chamber of the steam generator, stack sampling was not performed to determine if this
destruction method is appropriate for full-scale treatment. A proper evaluation would
have involved detailed stack profiling and analysis for destruction efficiency, and
presence of dioxins and furans in the stack effluent. Such testing was planned for
operation at the maximum rates. Such conditions were never reached, and the sampling
was therefore omitted.
8.3 PROCESS OPERATIONS AND DESIGN
Data collected during the pilot study were to be used to evaluate operational
approaches to removing NAPL, treatment plant performance, and the effectiveness of
microbial degradation, and to determine oxidation rates. Each of these is discussed
below.
8.3.1 Operations
The pilot study was not operated as designed and the subsurface was not heated
enough to demonstrate the overall potential of the technology. The shortcomings
included:
The designed injection and extraction rates were not achieved, due to limitations
of the process equipment. In particular, failures of the vacuum pumps and heat
exchanger prevented vapor extraction during the 2003 operations period. This
means that in the last four months of operations the dual-phase extraction system
was inactive, and only liquids were extracted.
Steam distribution was not satisfactory due to the lower than anticipated injection
pressures and rates, the short steam injection time, and the failure of the overall
process to be implemented as designed. This lead to uneven heat distribution in
the subsurface.
Approximately 1 pore volume of steam was injected into the site, an amount
insufficient to heat it to the target temperature. Two to three pore volumes of
steam were planned, as has been used at other wood-treater sites.
The operations period was shorter than would be expected for full-site treatment,
allowing less time for degradation reactions to assist in the reduction of
contaminant concentrations.
Several operational modes were defined for the period after site heating was
achieved (pressure cycling and mass removal optimization phases). Since
satisfactory heating was not achieved, these operational modes were never
implemented.
A good steam push across the aquitard surface was never achieved.
Due to these shortcomings, the optimal operations strategy for full-scale NAPL removal
and treatment could not be determined.
8.3.2 Heating Design
One of the objectives of the pilot study was to determine if the existing steam
injection design was adequate. Since the steam injection wells in the upper aquifer were
not operated as designed, it cannot be concluded whether satisfactory heating can be
achieved using the injection wells with single deep screens as designed for the pilot
study. However, several observations support a conclusion that an improved heating
strategy for the bottom of the aquifer (and the top of the aquitard) would be beneficial:
Observations of significant steam override at several locations within the pilot test
area, particularly in the northern, deep end.
Groundwater pumping tests and temperature observations indicate that a

substantial upward water flow through the aquitard takes place when the fluid
level is depressed by extraction. This upward migration of cold water will keep
cooling the base of the aquifer in the locations where water flows, reducing the
chance of satisfactory heating. This will act to impede deep steam penetration
across the top of the aquitard, where DNAPL may be accumulated.
Furthermore, if in the future there is increased focus on the potential for DNAPL
penetration into and through the aquitard, then there may be additional focus on the
aquitard, and potentially the aquitard could be included in the remediation volume. This
would mean that a robust heating strategy should be added to the design.
The pilot design was not properly or fully implemented, and thus conclusions
cannot be made about the adequacy of the heating system. It is likely that a supplemental
heating approach for the base of the upper aquifer would be a benefit. The three
competing techniques are:
1. Use direct electrical-resistive heating process, three-phase electrical heating, to

heat the upper 5-8 ft of the aquitard and the base of the upper aquifer. This would
involve 3-phase electrodes placed in a triangular pattern, with electrode separation
in the 20 to 30-foot range.
2. Use In-Situ Thermal Desorption (ISTD) to heat the upper 5-8 ft of the aquitard
and the base of the upper aquifer. This would include direct heating elements
installed in a triangular pattern with typical separations of between 12 and 25 ft.
3. Use steam injection below the aquitard to encourage upwards steam flow through
the aquitard, and to block upward water flow. This approach was used
successfully at Visalia Pole Yard, but a significant amount of characterization and
risk assessment would be required before trying this approach at the Wyckoff site
Further evaluation and discussion of these three alternatives are warranted before
a full-scale design is prepared, or a modified and improved pilot study is performed.
8.3.3 Maximum Treatment System Loading
Since the pilot study did not achieve satisfactory subsurface temperatures, the
designed steam break-through to the extraction wells, or maximum contaminant
extraction levels in the liquid and vapor systems, it would be difficult to attempt to use
the data to design a full-scale groundwater and vapor treatment system. The following
data are still needed for design of a system for full-scale treatment:
1. Estimates of the rates of NAPL recovery during and after site heating. The site
was never heated adequately, and the NAPL recovered from the pilot test area
was not separated from the NAPL recovered from the FPA wells to determine the
amount of NAPL recovered from the pilot test area.
2. NAPL behavior in extracted water. Density changes that change a DNAPL to an

LNAPL and/or emulsification are possible, but were not determined since the
liquid stream from the pilot test area was mixed with the liquid stream from the
FPA wells in order to better control the temperature of the liquid in the treatment
plant.
3. Maximum dissolved phase contaminant concentrations achieved during maximum

mass recovery times. While the TOC readings indicated more than a 10-fold
increase in dissolved organics, the maximum contaminant concentrations during
times where steam breakthrough to several extraction wells occurred could not be
measured because the TOC analyzer did not have adequate capacity.
4. Vapor phase flow rates and contaminant concentration. Instrumentation for

measuring vapor flow rates was not installed during the pilot test, and vapor
samples for chemical analysis were never obtained.
5. True extent of the PAH precipitation problem. It is possible that PAH fouling
(i.e., precipitation and crystallization of naphthalene and other PAH) may have
been reduced by full-capacity operation per design assumptions, which would
have lead to higher temperatures in the overall system. It is also possible that
precipitation problems could have been worse during full capacity operations.
6. Condensate flow rates and contaminant levels achieved during maximum mass
removal periods. Methods to separately measure condensate flow and
contaminant concentration were not included in the Pilot Study design.
8.3.4 Vacuum and Vapor Flow
The pilot test area included a vapor cap with horizontal vapor collectors as well as
vapor extraction from each of the seven extraction wells. The system only functioned for
about one month of the pilot study period, and was erroneously operated with most of the
flow originating from the surface vapor-collection system. Information sufficient to
support full-scale design of an adequate vapor extraction and treatment system is
therefore not available.
SECTION 9.0 LESSONS LEARNED
During analysis of the results of the study, technical details regarding contaminant
composition, design, construction, and operations were identified that may be useful in
implementing a new pilot study, if needed for the Wyckoff site, or may be useful to
others for implementing thermal treatment systems at other sites. Team structure,
scheduling, and budgeting were also determined to have had influences on the project.
This section includes these Lessons Learned with some suggestions on what to do
differently.
9.1 SITE AND CONTAMINANT COMPOSITION CHARACTERIZATION
Naphthalene concentration. The pilot study area contained much higher concentrations
of naphthalene than is typical for creosote. Baseline data collected in 2001 and 2002
from the pilot test area demonstrate this fact, however, the data were not completely
analyzed prior to the start of the pilot study. Also, the physical and chemical
characteristics of naphthalene were not well understood. A thorough understanding of
the creosote composition, naphthalene characteristics, its distribution in the subsurface,
and how it interacts with other components in the NAPL is needed prior to design and
operations. A better understanding of the site conditions may have resulted in system
design changes or helped to predict and/or address problems.
This problem has been solved at other creosote sites, such as the Alhambra, CA site
where thermal treatment has been implemented. The vapors are kept above
crystallization temperatures by insulation and heating of the manifold pipes. All the
extracted naphthalene is conveyed to the thermal oxidizer for destruction in the vapor
phase.
Hydrocarbon characterization. A significant fraction of the LNAPL is lighter end

hydrocarbons. This fraction was not routinely tested for during site characterization
activities leading up to and during the pilot study. A better understanding of what this
light end is composed of and how it potentially will interact with PAH through the
temperature ranges of interest during thermal remediation is needed to refine the design
and operations strategy.
9.2 DESIGN
Several issues were identified related to the original design and design changes
made during the pilot study that adversely affected the outcome of the pilot study. The
following section includes details on how the sequencing of certain events could be
improved to be more efficient and cost effective, as well as technical details about the
pilot test area location, the sheet pile wall, the liquid and vapor extraction systems, the
GWTP, and well and instrumentation spacing.
9.2.1 Schedule
Well Field Design. The original well field design was based on assumptions
regarding aquifer hydraulic conductivity. After construction was started on the well
field using the original design, a pumping test was completed on wells within the
pilot study area, and a thermal multi-phase numerical model was run using these
results. Modeling indicated that the well spacing (between injection and extraction
wells) was at the maximum estimated to be effective for thorough heating and
transfer of steam. Closer well spacing would have been better. Though no changes
to the well spacing were deemed necessary, if changes had been necessary, it would
have been costly to change the well spacing after construction had been partially
completed since all of the 10-inch diameter wells were already installed. The pilot
study itself would have determined if the well spacing was too far apart. The contract
had optional bid items to add additional injection and extraction wells during the
study.
Procurement. Procurement of materials and equipment (e.g., well screens) should not
be initiated until after the design has been finalized.
Liquid and Vapor Extraction System. The design of the liquid and vapor extraction
system should not be finalized until after the well field design has been completed.
An even better lesson would be to conduct vacuum tests on the wells before designing
the vacuum system. Maintaining a lag period in the schedule for the liquid and vapor
extraction system design can prevent unnecessary reworking of the extraction design
due to changes in the well field design.
9.2.2 Pilot Test Area Location
When attempting to determine from a pilot study if stringent remedial objectives can be
achieved, the pilot test area must be isolated from the rest of the site and the
contamination within the area must be treated. This is the most effective way to evaluate
the success of the technology and ensure that the pilot test area will not have to be
revisited during full-scale treatment. Because soil contamination and NAPL remain in
the area upgradient of the pilot test area and within the sheet pile wall, recontamination of
the treated area after steam injection activities are completed is expected. This will
compromise evaluation of the ability of the system to reach project cleanup goals as well
as the evaluation of the contribution of in-situ oxidation to the reduction of site
contaminants. It was intended that this area would be re-treated during full scale
operations.
A smaller-sized pilot area may have better matched the budget and still have been
sufficient in size to establish design criteria for a larger-scale project. The pilot area
needs to be big enough to collect data and not too large to increase the cost of operations.
The pilot objectives would also need to be revised.
9.2.3 Sheet Pile Wall
The Frodingham sheet pile wall used to contain the pilot test area worked well in
several respects:
The installation procedure worked well and an intact wall was installed at the
desired locations.
Horizontal water flow was arrested/reduced substantially, allowing for some

dewatering inside the pilot test area without significant horizontal recharge
through the wall.
Steam migration was effectively arrested by the wall, as evidenced by the low
temperatures measured outside of the pilot test area.
No negative effect of the steam injection on the wall integrity was observed,
although steam was observed venting to the surface at one of the joints.
The temperatures achieved in locations next to the sheet pile wall were as high or
above pilot test volume average temperatures, indicating that the presence of the
wall did not impede heating of the material next to it. This is consistent with the
Design Analysis thermal modeling results, and is favorable for full-scale
operations, since the treatment efficacy next to the wall (close to the receptor) is
important.
Other observations about the sheet pile wall include the following:
Weep holes in the sheet pile were installed at the gravel surface.
Recommend also installing additional weep holes at the liner elevation.
The sealing of the barrier liner (membrane) to the sheet pile wall needs to
be improved so there is no leakage of surface water into the vapor
collection system.
This indicates that a sheet pile wall of this type will work well for containing
steam (and contaminants) during full-scale treatment, without compromising the
treatment efficiency.
9.2.4 Liquid and Vapor Extraction Systems
Vapor stream treatment system. The vapor stream treatment system should have
adequately-sized air/liquid separators (i.e., knock-out tanks), condensate collection
vessels, and transfer pumps. Other thermal projects have used a robust knock-out vessel
equipped with a demister and level switches, designed specifically for condensate and
with excess capacity to account for surges and fluctuations. The actual equipment
installed relied on a wide vertical pipe to drop out the condensate before it reached the
liquid ring vacuum pumps. The design did not have enough of a safety factor as it might
have had with large condensate receiver tanks to handle a slug of water introduced into
the vapor line, which by design should not have occurred, but most likely did occur.
Vapor extraction system. Vapor extraction system must be capable of handling relatively
large quantities of condensate and preventing condensate from reaching the vacuum
pump. The condensate collection system should have a high-level alarm located
appropriately to provide enough time to shut down the vacuum pump if condensate
accumulates faster than it can be removed. The Visalia SEE project used water-seal
vacuum pumps, which are better able to accept water than oil-seal pumps without
compromising their function.
Vapor extraction system gauges. To better understand the vapor collection in the
subsurface, vacuum gauges should be installed on every branch of the system.
Heat exchangers. To account for crystallization and precipitation of naphthalene, two
heat exchangers should be installed in a parallel configuration in the vapor extraction
line. This would allow operation of one of the units to continue when the other unit must
be taken off-line for steam cleaning. In addition, a temperature controlled system for
removal of naphthalene along with the condensing of extracted vapors needs to be
incorporated into any re-design.
Vapor cap. Integrity of the impermeable membrane (material compatibility, sealing of

penetrations) in the vapor cap should be maintained to minimize infiltration of
precipitation. A quality assurance plan to verify membrane integrity and complete
grouting of all membrane penetrations should be implemented. A vacuum test should
also be performed to verify that the membrane performs as needed based on the capacity
of the selected vacuum pumps.
Liquid extraction line. Pneumatic pumps in the liquid extraction line need air-flow
regulators.
Aboveground piping system. As constructed, the aboveground piping layout associated

with the extraction pumps routed the exhaust lines outside the wellhead. During
operations, rapid cooling of vapors in the exhaust line caused precipitation of PAH and
fouling of the exhaust line. This resulted in significant degradation in pump performance
and required frequent maintenance to correct. The manufacturer had recommended the
exhaust line be inside the wellhead.
Cooling system. The capacity of the cooling system ended up being only 10% of the
original design. This needs to be re-evaluated to make sure the system is robust enough.
Steam regeneration of carbon. The original design required a steam-generated carbon-

adsorption system for use when the boiler was not in operation. There are issues with
using the boiler for steam regeneration, which were eliminated if the thermal oxidizer
were used in its place.
Material incompatibility. The chemical incompatibility between the seals and gaskets in
the equipment and the wastewater constituents led to mechanical failure of the equipment
on a daily basis. This issue required continuous maintenance and ultimately shut down
the pilot.
9.2.5 Groundwater Treatment Plant
NAPL Removal System. The treatment system designed for dissolved-phase

contaminants must be preceded by one or more processes for removal of NAPL. The
Wyckoff system included a dissolved-air floatation tank (DAFT) to separate NAPL from
dissolved-phase contaminants. The treatment plant also included a bioreactor along with
a final activated carbon polishing step which were the principle dissolved-phase
treatment processes. There are some questions regarding whether the DAFT was
operating properly. Removal efficiencies for oil and grease were generally poor, and
often highly variable. (The addition of polymer in the DAFT following the pilot study
improved its operation.) Some of the interior piping in the DAFT was found to be
corroded in the summer of 2003. Also, there were reports of build-up of naphthalene
precipitate on interior surfaces of the DAFT and its cover. It is not known when (or how
often) the interior of the DAFT was inspected, or whether the diffusers were operating
properly.
Naphthalene Removal. Dissolved air floatation is generally recognized as being one of

the best available processes for separating NAPL from the aqueous phase liquids.
However, it is not clear if this process can be expected to adequately remove the
naphthalene crystals that formed in the aqueous phase due to the extremely high
naphthalene concentrations in the pilot test area.
The extremely high naphthalene levels might have been a temporary condition, which
may have been because the project had only entered into the early stage of heat-up of the
pilot test area. If the pilot study had continued according to plan, it is possible that the
naphthalene-fouling problems would have dissipated when the site came up to
temperature. After the site was at temperature, the DAFT might have been capable of
providing adequate NAPL (and naphthalene) removal. Since the pilot study was
discontinued before the site came up to temperature, the duration of the naphthalene-
fouling problem and the relation to subsurface temperature remain largely unknown.
Treatment system capacity. An accurate evaluation of the true capacity of the existing
treatment system should have been done prior to the start of the pilot study. Problems
encountered here due to the lack of adequate treatment plant capacity could have been
avoided by providing a separate temporary treatment system for the liquids extracted
from the pilot test area.
Liquid conveyance system. Separate treatment and conveyance systems for the liquids
from the pilot system and the FPA would have allowed determination of the amount of
NAPL extracted from the pilot test area. However, these liquid conveyance systems were
combined to have some control over the temperature of the water entering the treatment
plant to minimize the risks to the microbial population.
Cooling capacity. The capacity of the effluent pipe down-gradient of treatment plant tank
303 to discharge the combined flow of treated water and non-contact cooling water was a
constraint on the steam injection rates. The capacity of the cooling system needs to be
adequate for the designed steam injection rate.
Oil/Water Separator. A properly-sized oil/water separator installed upstream of the

DAFT would help in the DAFT performance.
9.2.6 Well and Instrumentation Spacing
Without having operated the pilot study at the design rates, any evaluation of how
well the designed system would have worked will be inaccurate. However, the following
observations are offered based on a holistic evaluation:
The density of the subsurface temperature monitoring strings was sufficient to

document the heating that occurred within the pilot test area. Vertical profiles and
select transects and planar views were useful for the operators to make decisions
and to follow the progress. Because this was a pilot study, the level of monitoring
was higher than would normally be used during a full-scale implementation of the
technology. Since the operations never proceeded to the later stages where the
objective would be to heat the coolest areas, the full value of the high density of
monitoring locations was not realized.
During calculations of the energy balance for the pilot study volume, it was
observed that the heat loss through the sheet pile could not be calculated based on
a measured temperature gradient near the sheet-pile. If this is an objective, the
line of thermocouples outside of and perpendicular to the sheet pile wall should
be located closer to the wall and to each other.
Seven of the first set of pressure transducers were lost due to damage during
curing of the grout. Failure is believed to be the result of penetration of the
transducer membrane by silica flour used in the grout. Determination of the
failure mechanism has not been confirmed and would require excavation of the
transducers. Replacement transducers were installed just before the pilot started
in a way that prevented damage, and the density of the water level monitoring
sensors was sufficient.
The most appropriate spacing between injection and extraction wells for full-scale
remediation could not be determined because the pilot study system was not
operated at the design rates. However, several limitations to heating were noted:
o Significant steam override of cooler fluids occurred.
o Most heating was concentrated in the vadose zone, with only modest
heating near the base of the aquifer.
o Steam was observed to have different penetration depths and distances

across the pilot test area, indicating that spatial variations in permeability
control heat distribution.
o The base of the water-bearing zone stayed relatively cool, which was
partially attributed to upward flow of cold water through the aquitard. As
a consequence, a smaller average well spacing, combined with an
additional method for heating the base of the upper water-bearing zone
and the upper part of the aquitard, would add confidence to the design, and
allow for a more uniform and predictable heating than was observed
during the pilot study.
Overall, the well and instrument spacing used for the pilot study were appropriate
for the objectives. However, the full value was not realized due to the shortening of the
test and operation at injection rates that were much lower than the design rates.
9.3 CONSTRUCTION AND EQUIPMENT
Well development. Wells should be thoroughly developed with several hours of vigorous
surging until negligible sediment is removed even though they have filter packs.
Geo-membrane integrity. Holes were poked through the geo-membrane of the cap and
may not have been completely sealed. Water infiltrating the cap and the vapor collection
system may have been one reason for the inundation of the liquid ring vacuum pumps.
Contractor-requested design changes. Need design engineers to review any contractor-

initiated design changes and be present during commissioning of the equipment to verify
project was built as designed.
Equipment selection. Install only the equipment needed for the pilot test and do not
prematurely anticipate moving on to full scale. Investing in the larger boiler used up part
of the budget that could have been used to support some other aspects of the pilot study
that were eliminated.
9.4 OPERATIONS
Shakedown period. Allow for a proper shakedown period and do not start operations
until all equipment, including instrumentation, is installed, tested, and modified (if
necessary). Testing equipment in isolation does not necessarily provide reliable
indications of performance when all components of the system are operated together.
Some instrumentation equipment, which had been deleted during the bidding period on
the construction contract, was not purchased nor installed until after the pilot study had
started.
Changing conditions. Initial operation of the system should be according to the

management plan. A process should be developed for addressing changing conditions
that includes Internal Technical Review staff and system designers to deal with
conditions encountered in the field that necessitate changes.
Steam injection rate. Injecting steam at less than the maximum possible design rate has
several significant disadvantages. Maximum horizontal steam flow (minimum override)
requires maximum injection pressure, maximum vacuum, and maximum pumping rate. It
is possible that there will be less PAH precipitation and encrustation at maximum
capacities due to higher overall temperatures in the system. Design flows should be
reached as quickly as possible. Costs of labor and expenses are the same regardless of
level of operation, making reduced rate operations an inefficient use of money due to the
less-than-maximum effort.
Vapor extraction. With the decreased vapor extraction system size, the vacuum should
be reduced at the surface collectors, per design assumptions, so almost all of the vapor
extraction is from the wells.
Thermal Oxidizer. The thermal oxidizer was needed on the project for the period when
the boiler was not operating. Bringing the equipment to the project earlier than scheduled
in lieu of burning the non-condensable vapors in the boiler may have eliminated any
boiler stack emission issues.
Treating the aquitard. As previously noted in Section 8.3.2, a substantial upward water
flow through the aquitard occurred during extraction in the northeast corner of the pilot
area. This cooling will impede the treatment of the aquifer. Therefore, additional
treatment of the aquitard may be required to successfully remove the contaminants.
9.5 MONITORING PROGRAM
9.5.1 Chemical Monitoring Data Needs And Methods
Monitoring plan. The monitoring plan for the system chemistry was too intense and at a
micro-scale for the majority of the operations. The approach towards system
performance monitoring needs to meet the data and timing needs of operations and
performance assessment. For example, daily extraction well samples were unnecessary;
biweekly samples likely would have been adequate, until closer to the end of the injection
phase. As another example, while the TOC data clearly showed slugs of contaminants
and were instrumental in diagnosing problems with the aeration basin, the instrument was
too sensitive for its use during this test. Additionally, the timing of some of the
monitoring efforts needs to be re-evaluated. For example, measuring natural attenuation
parameters in hot extraction well water samples is unnecessary until after the steam
injection phase has ended.
Equipment installation schedule. The pilot study was started before all monitoring
equipment was installed and before methods for sample collection and analysis were
finalized. For example, systems were never installed for measuring non-condensable gas
flow and composition or stack emissions.
Review of monitoring system. The monitoring system should be reviewed and updated
to include all parameters needed for daily performance monitoring and to evaluate the
overall system performance after the completion of the pilot study. For example, better
measurement methods need to be established for determining which contaminants are
removed, how much, and in which phase. Additionally, if the in-situ degradation
pathway is to be monitored, an analysis of what the potential breakdown products will be
needs to be done and appropriate sample collection and analysis methods determined.
9.5.2 Automated Monitoring System
Geomation Operating Software (Intellution Fix). Due to the complexity of the iFix
software, it is recommended that simpler options be considered on future projects.
Geomation has one additional possibility on the market and is developing an easier
interface software package that should be available for purchase. Neither of these options
was available at the time of purchase of the equipment for this project.
Power Outages. The Geomation RTUs were AC powered and plugged into an adjacent
GFI receptacle. The GFI would occasionally trip due to moisture intrusion. It would
therefore be advantageous to have battery backup at the RTUs for future similar
installations. A rechargeable battery is a standard option supplied by Geomation.
Another way to handle the power outages would be to feed the power circuit for the
RTUs from a panel inside the building with a GFI breaker to eliminate the nuisance
tripping.
DTS Operating System. Communication between the controller and the PC would
occasionally lock up, requiring someone to close and restart the software and/or reboot
the controller. The graphics on the pc monitor would also occasionally not display
properly. The display problem was solved by reconfiguring the video mode selection for
pcAnywhere to compatibility mode. The communication problem was never solved even
after swapping out the laptop twice. This problem appeared to be related to the Windows
98 operating system. The DTS operating software has now been upgraded to run under
Windows 2000/XP, but has not been tested.
Additionally, by comparing the temperature in instrumentation boring T-7 it was

noted that temperatures read by the DTS were about 3 C higher than the temperatures
read by the thermocouples. To verify this manual temperatures were taken with a
portable probe in extraction well E-5 before it was sealed. The readings verified the 3 C
offset. Sensa explained that the calibration of the system could be modified to correct
this, but it was never done. The offset was constant and independent of temperature so it
was easy to correct for.
DTS versus Thermocouples (Geomation 3300). The USACE decided to operate two
different subsurface temperature-monitoring systems at the Wyckoff site. One was
composed of a series of thermocouples; the other was a DTS system using fiber optic
technology. This decision was based on the fact that since the Wyckoff project was a
pilot test site, it would be a good idea to test a promising new technology, DTS. The
DTS worked very well except for the communication problem discussed above.
Assuming this problem is solved with upgraded software, DTS is an excellent option for
monitoring temperatures in future projects. The thermocouples and Geomation 3300
system also worked very well. Although the installation was more complex, it functioned
almost perfectly. A few (6-10) thermocouple modules failed but were easily repaired or
replaced. The main advantage of the system is that it is also capable of reading other
types of instruments (i.e., 4-20mA, pulse counters, vibrating wire). Geomation 3300
systems also have the capability of communicating via radio to meet more complex site
requirements. A combination of the two systems would be an ideal solution for future
projects; DTS to monitor temperatures and Geomation 3300 to monitor everything else.
Silica Grout. The ITTAP panel recommended that silica grout be used in areas where
steam would be injected. Seven of the pressure transducers failed when installed with
silica grout. The use of standard grout or a sand filter pack around the transducer should
be investigated to see if either are acceptable alternatives.
Pump Stroke Counters. The pump stroke counters on the extraction wells manufactured
by Severn Trent did not function properly. Both the rotating dial and the digital output
failed routinely.
9.5.3 Data Management
Automated data collection. This part of the project worked extremely well. Little lag
time was noticed for those monitoring systems where data were generated automatically
and managed by the USACE. Generally the daily process was completed in about an
hour. The most time consuming task was the creation of the sections and slices showing
temperature distributions in GMS.
Non-automated data collection. There was a lag time between data collection and posting
to the web page. Data such as steam injection rates and well pumping rates were not
transferred to the webpage fast enough for daily decision making, therefore, the data were
compiled every morning and reported verbally to the team during the daily operations
call. When data became available, it was posted, making the webpage more effective as
an accessible data archive, but less effective as a real-time source of data. The manual
nature of this data gathering was inefficient. Better data collection tools need to be
designed for this part of the project.
Web page. Having data and communications logs posted to the project web page was an
effective method for team communication.
9.6 TEAM STRUCTURE AND COMMUNICATION
Having an appropriate team structure and effective communication methods are

vital to project success. Issues regarding team structure and communication surfaced
during design and operations. During analysis of the results of the study, it was
determined that these issues had had a strong influence on the project, beginning during
the design phase and continuing throughout the implementation of the pilot study. Issues
that were identified fell into four main categories: team continuity, the right resource for
the job, the decision making process, and the problem resolution process. This section
includes recommendations on characteristics of a team that should be considered when
implementing a pilot study.
When utilizing a complex, innovative technology such as steam injection,

technical people who are experienced with the technology should be utilized throughout
the project. Utilizing the experience and expertise of these people will reduce the time
needed for design, will increase the efficiency of the operation, and will likely reduce the
overall costs. For continuity through the project life cycle, managers, system designers,
outside technical advisors, and lead technical staff should be involved through design,
construction and operations. The team should maximize the use of industry expertise,
project knowledge and hard-won field experience in its team membership.
To implement a highly technical study, resources with the appropriate knowledge

and skill set should be selected and retained for the life cycle of the project. The
requirements for each project role should be defined early in the project and individuals
should be selected to best fit those roles. When appropriate resources are not available
within present staff, outside resources should be investigated.
The team should be structured in a way that allows decisions to be made at the
appropriate level. Major design changes and cost issues should be raised to appropriate
management prior to decision-making. Effort should be made to minimize bottlenecks in
decision making so day-to-day decisions can be made in a timely manner. Major
decisions affecting design, construction, and operations should involve and be approved
by the system designers. Once decisions are made, they should be accepted by the entire
team, unless changing conditions require re-analysis of the decision.
A problem and/or conflict resolution process should be established to resolve

issues in a timely manner. Communication channels should be open so ideas from all
team members will be heard, discussed, and acted upon as necessary. Failure to resolve
issues in a timely manner leads to internal team dissension and lack of clarity on the goals
of the project. In this situation, goals should be re-visited with appropriate management
to reach agreement prior to proceeding with the project.
9.7 SCHEDULE AND BUDGET
The project schedule and budgets should be based on the anticipated level of
effort needed to complete the scope of work and meet the project objectives.
Field investigation work should be completed before the design is started.
The project schedule should account for adequate time to test, and modify if
needed, all new equipment prior to starting the official pilot study.
Schedule and budget demands should not restrict the ability of the project team to
raise and resolve issues effectively.
Due to the inherent uncertainty in implementing the pilot study, a contingency

plan should have been in place to cover schedule changes and potential increases
in project costs.
SECTION 10.0 REFERENCES
Adenekan, A.E. Numerical Modeling of Multiphase Transport of
Multicomponent Organic Contaminants and Heat in the Subsurface, PhD thesis,
University of California at Berkeley, Berkeley, CA 94720, 1992.
Bamford, H. A., J. E. Baker, and D. Poster. 1998. Review of Methods and

Measurements of Selected Hydrophobic Organic Contaminant Aqueous Solubilities,
Vapor Pressures, and Air-Water Partition Coefficients, NIST Special Publication 928, US
Department of Commerce, March.
Davis, E. L., Steam Injection for Soil and Aquifer Remediation, USEPA Ground
Water Issue paper, EPA/540/S-97/505, January 1998.
Davis, E. L., How heat can enhance in-situ soil and aquifer remediation:
Important chemical properties and guidance on choosing the appropriate technique,
Ground Water Issue Paper, US Environmental Protection Agency, EPA/540/S-97/502,
April 1997.
Hunt, J. R., N. Sitar, and K. S. Udell, Nonaqueous phase liquid transport and
cleanup, 2. Experimental studies, Water Resources Research, 24(8):1259-1267, 1988.
Itamura, M. T., and Udell, K. S., Experimental Clean-Up of a Dense Non-

Aqueous Phase Liquid in the Unsaturated Zone of a Porous Medium Using Steam
Injection, Multiphase Transport in Porous Media 1993, ASME FED-Vol. 173/HTD-Vol.
265, pp. 57-62, 1993.
Lief, R. N., M. Chiarappa, R. D. Aines, R. L. Newmark, and K. G. Knauss, In situ

hydrothermal oxidative destruction of DNAPLs in a creosote contaminated site, in
Wickramanayake, G. B. and R. E. Hinchee, ed., Physical, Chemical, and Thermal
Technologies, proceedings of The First International Conference on Remediation of
Chlorinated and Recalcitrant Compounds, Monterey California, May 18-21, 1998, pp.
133-138.
Nitao, J J., Users Manual for the USNT Module of the NUFT Code,
Version 2.0, Lawrence Livermore National Laboratory, Livermore, CA,
UCR L- MA-130653, 1998.
Pruess, K., TOUGH User's Guide, Report NUREG/CR-4645, Nuclear Regulatory

Commission (also Report LBL-20700, Lawrence Berkeley Laboratory, Berkeley,
CA), 1987.
Udell, K.S. (1996): Heat and mass transfer in clean-up of underground toxic
wastes. In Annual Reviews of Heat Transfer, Vol. 7, Chang-Lin Tien, Ed.; Begell House,
Inc.: New York, Wallingford, UK, pp. 333-405.
Udell, K. S., and L. D. Stewart, Jr., Field study of in situ steam injection and
vacuum extraction for recovery of volatile organic solvents, Department of Mechanical
Engineering, University of Calf., June, 1989.
U.S. Army Corps of Engineers (USACE). 2002. On-Site Water Supply Well
Report, Well 010CT01, Wyckoff/Eagle Harbor Superfund Site, Bainbridge Island,
Washington, June 4, 2002.
U.S. Environmental Protection Agency (USEPA). 2000. Record of Decision:

Wyckoff/Eagle Harbor Superfund Site, Soil and Groundwater Operable Units,
Bainbridge Island, Washington. U.S. Environmental Protection Agency, Region 10,
Seattle, Washington.
In Situ Thermal Technology Advisory Panel, 1999, Meeting Minutes, June 2, 1999;
report by U. S. Army Corps of Engineers, for EPA Region 10.
Tables
TABLE 3.2-1 PHYSICAL CHARACTERISTICS AND THERMODYNAMIC PROPERTIES OF SOIL
SAMPLES
T-1 T-22 T-30
6-7 ft bgs 13.5-14 ft bgs 22-22.5 ft bgs
Silty sand & silt Clay Sand
Porosity (%) Total 30.8 50.5 28.6
Effective 2.39 9.27 23.67
Specific retention 92.23 81.65 17.17
Bulk density (g/cc) Before TC 2.066 1.663 2.082
After TC 2.187 1.753 2.146
Thermal at 30 oC 0.7186 0.5964 0.5114
Conductivity at 60 oC 0.7848 0.6640 0.6188
o
(BTU/hr-ft- F) at 90 oC 0.8458 0.7531 0.6976
at 120 oC 0.8635 0.7717 0.7503
Specific heat at 30 oC 0.399 0.382 0.334
capacity at 60 oC 0.307 0.290 0.236
o
(BTU/lb- F) at 90 oC 0.281 0.280 0.265
at 120 oC 0.338 0.306 0.282
Notes:
bgs below ground surface
TC thermal conductivity analysis
TABLE 3.5.1 EXTRACTION WELL BASELINE CHARACTERIZATION SAMPLING RESULTS
GW
E-1 E-2 E-3 E-4 E-4 E-5 E-6 E-6 E-7
Analyte Name Cleanup
(11/7/02) (11/7/02) (11/7/02) (11/5/02) (11/7/02) (11/7/02) (11/5/02) (11/7/02) (11/7/02)
Level
Naphthalene 83 2300 J 1700 J 2900 J 882 430 J 2 J 0.39 U 1100 J 1400 J

2-Methylnaphthalene 400 J 27 J 360 J 46.7 51 J 5 UJ 0.012 J 110 J 1000 J
Acenaphthylene 500 UJ 25 UJ 10 UJ 2.7 5 UJ 5 UJ 0.39 U 5 UJ 33.5 UJ
Acenaphthene 3 280 J 300 J 250 J 180 260 J 40 J 0.071 J 180 J 380 J
Anthracene 9 500 UJ 19 J 15 J 10.2 100 UJ 5 UJ 0.29 J 100 UJ 51 J
Phenanthrene 130 J 120 J 150 J 102 100 J 5 UJ 0.084 J 75 J 370 J
Fluoranthene 3 500 UJ 11 J 300 UJ 16 11 J 3 J 0.24 J 100 UJ 82 J
Pyrene 15 500 UJ 250 UJ 5 J 7.4 100 UJ 1 J 0.17 J 100 UJ 53 J
Benzo(a)anthracene 0.0296 500 UJ 25 UJ 10 UJ 0.43 U 5 UJ 5 UJ 0.39 U 5 UJ 11 J
Chrysene 0.0296 500 UJ 25 UJ 10 UJ 0.34 J 5 UJ 5 UJ 0.39 U 5 UJ 10 J
Benzo(a)pyrene 0.0296 500 UJ 25 UJ 10 UJ 0.39 U 5 UJ 5 UJ 0.032 J 5 UJ 33.5 UJ
Benzo(b)fluoranthene 0.0296 500 UJ 25 UJ 10 UJ 0.081 J 5 UJ 5 UJ 0.39 U 5 UJ 33.5 UJ
Benzo(k)fluoranthene 0.0296 500 UJ 25 UJ 10 UJ 0.39 U 5 UJ 5 UJ 0.39 U 5 UJ 33.5 UJ
Benzo(g,h,i)perylene 500 UJ 25 UJ 10 UJ 0.39 U 5 UJ 5 UJ 0.39 U 5 UJ 33.5 UJ
Indeno(1,2,3-cd)pyrene 0.0296 500 UJ 25 UJ 10 UJ 2U 5 UJ 5 UJ 2U 5 UJ 33.5 UJ
Dibenzo(a,h)anthracene 0.007 500 UJ 25 UJ 10 UJ 0.78 U 5 UJ 5 UJ 0.78 U 5 UJ 33.5 UJ
Pentachlorophenol 4.9 65 UJ 650 UJ 750 UJ 0.78 U 12.5 UJ 12.5 UJ 0.58 J 22 J 85 UJ
Notes:
Concentrations in g/L
Bold indicates detected concentration exceeds Groundwater Cleanup Level
TABLE 3.5-2 UPPER AQUIFER BASELINE CHARACTERIZATION GROUNDWATER SAMPLING
RESULTS
GW
Cleanup MW17 MW18 MW18 Dup MW19
Analyte Name Level (11/7/2002) (11/7/2002) (11/7/2002) (11/6/2002)
Naphthalene 83 2940 12600 12800 13.6
Naphthalene, 1-methyl- 166 779 800 0.74
Naphthalene, 2-methyl- 50.1 1080 1100 0.91
Acenaphthylene 3.2 9.8 8.9 0.079J
Acenaphthene 3 247 291 300 0.35J
Anthracene 9 10.2 13.8 12.7 0.56
Phenanthrene 36.7 102 101 0.097J
Fluoranthene 3 19.1 9.4 9.1 0.19J
Pyrene 15 8.9 4.9 4.9 0.2J
Benzo(a)anthracene 0.0296 0.87 0.71U 0.76 0.37U
Chrysene 0.0296 0.66 0.63 0.68 0.17J
Benzo(a)pyrene 0.0296 0.25J 0.18J 0.21J 0.15J
Benzo[b]Fluoranthene 0.0296 0.34J 0.31J 0.35J 0.28J
Benzo[k]fluoranthene 0.0296 0.15J 0.13J 0.11J 0.075J
Benzo(g,h,i)perylene 0.079J 0.39U 0.056J 0.12J
Indeno(1,2,3-cd)pyrene 0.0296 0.086J 2U 0.062J 0.13J
Dibenzo[a,h]anthracene 0.007 0.74U 0.78U 0.77U 0.74U
Pentachlorophenol 4.9 0.59J 1930 1920 1.2
Notes:
TABLE 3.5-3 LOWER AQUIFER BASELINE CHARACTERIZATION GROUNDWATER SAMPLING
RESULTS
GW
Cleanup CW05 CW09 CW15 MW02 MW04 CW05 Dup
Analyte Name Level (11/6/2002) (11/6/2002) (11/6/2002) (11/7/2002) (11/7/2002) (11/6/2002)
Naphthalene 83 0.49 8.6 98.8 0.38U 0.37U 0.55
Naphthalene, 1-methyl- 0.17J 1.1 16.4 0.38U 0.37U 0.16J
Naphthalene, 2-methyl- 0.23J 0.22J 10.1 0.38U 0.37U 0.21J
Acenaphthylene 0.36U 0.37U 0.36J 0.38U 0.37U 0.38U
Acenaphthene 3 0.43 1.3 41.3 0.021J 0.37U 0.35J
Anthracene 9 0.19J 0.21J 2.7 0.38U 0.034J 0.22J
Phenanthrene 1.1 2 38.9 0.38U 0.02J 0.85
Fluoranthene 3 1.4 0.98 9.6 0.11J 0.039J 1.6
Pyrene 15 0.76 0.61 5.2 0.12J 0.37U 0.94
Benzo(a)anthracene 0.0296 0.36U 0.37U 1.2 0.38U 0.37U 0.38U
Chrysene 0.0296 0.054J 0.16J 0.96 0.38U 0.37U 0.1J
Benzo(a)pyrene 0.0296 0.36U 0.045J 0.2J 0.38U 0.37 U 0.38U
Benzo[b]Fluoranthene 0.0296 0.36U 0.084J 0.41 0.38U 0.37U 0.38U
Benzo[k]fluoranthene 0.0296 0.36U 0.032J 0.14J 0.38U 0.37U 0.38U
Benzo(g,h,i)perylene 0.36U 0.37U 0.046J 0.38U 0.37U 0.38U
Indeno(1,2,3-cd)pyrene 0.0296 1.8U 1.9U 1.9U 1.9U 1.9U 1.9U
Dibenzo[a,h]anthracene 0.007 0.73U 0.74U 0.74U 0.75U 0.74U 0.75U
Pentachlorophenol 4.9 0.73U 0.95 0.54J 0.52J 0.74U 0.75U
Notes:
TABLE 4.1-1 DESIGN CRITERIA SUMMARY
System, Process or Original Design Redesign
Parameter
Treatment Duration Eight to 14 months of active steam injection, followed by contaminant Six to eight months (with an optional two-month steam
extraction for six to eight months injection period) followed by contaminant extraction for
six months (plus a two-month optional period)
Steam injection 0.5 psi per foot of soil above the center of the injection screen Unchanged
pressure
Extraction well 7.3 psi (0.5 atm) 3.7 psi. (0.25 atm)
vacuum
Vapor collector 1 psi 0.2 psi.
vacuum
Well field arrays Well field arrays should cover the entire NAPL-contaminated portion of the The southernmost row of arrays was eliminated leaving
Pilot Test Area, with the southern-most (upgradient) injection wells outside 16 injection wells and seven extraction wells
the NAPL area. Included 25 injection wells and 16 extraction wells.
Average maximum 2 gpm liquid equivalent (l.e.) per well (1000 lb/hr). Unchanged
steam injection rate
Total steam 44 gpm l.e. (22,000 lb/hr) 32 gpm l.e. (16,000 lb/hr).
requirement
Average liquid 5 gpm per extraction well, for total of 45 gpm Unchanged per well but with a total of 25 gpm
extraction rate
Average condensable 1.2 gpm l.e. per well (640 lb/hr), for a total of 16.2 gpm l.e. (7,780 lb/hr) Unchanged
vapor extraction rate
Non-condensable gas 350 scfm, about 40% of which was vapor cap leakage recovered from 250 scfm with only 13% from surface vapor collectors.
(primarily air) flow surface vapor collectors.
rate
Total enthalpy of 50% to 100% of injected enthalpy (ITTAP recommended that 30% be used 10%
extracted gas for design purposes)
Table 4.2-1 Liquid Conveyance and Treatment System
Equipment 1 Design Analysis Contract Initial Installation Final Installation Impacts on Pilot2
Liquid Conveyance
Extraction EWP1 11 gpm, 0.275 gal pump 11 gpm, 0.275 gal pump 11 gpm, 0.275 gal pump stroke, QED Hammerhead No impact
well pumps (DNAPL) stroke, QED Hammerhead stroke, QED Hammerhead pump
pump pump
EWP2 11 gpm, 0.275 gal pump Optional item, 11 gpm, 0.275 Not installed Delayed LNAPL recovery
(LNAPL) stroke, QED Hammerhead gal pump stroke, QED
pump Hammerhead pump
EWP3 11 gpm, 0.275 gal pump 11 gpm, 0.275 gal pump 11 gpm, 0.275 gal pump stroke, QED Hammerhead No impact
(DNAPL) stroke, QED Hammerhead stroke, QED Hammerhead pump
pump pump
pump pump
pump pump
(LNAPL) stroke, QED Hammerhead gal pump stroke, Blackhawk
pump pump
EWP9 11 gpm, 0.275 gal pump 11 gpm, 0.275 gal pump 11 gpm, 0.275 gal pump stroke, QED Hammerhead No testing of different
(DNAPL) stroke, QED Hammerhead stroke, Blackhawk pump pump pump performance could
pump occur
(LNAPL) stroke, QED Hammerhead gal pump stroke, Blackhawk
pump pump
EWP11 11 gpm, 0.275 gal pump 11 gpm, 0.275 gal pump 11 gpm, 0.275 gal pump stroke, QED Hammerhead No testing of different
(DNAPL) stroke, Blackhawk pump stroke, Blackhawk pump pump pump performance could
occur
EWP12 11 gpm, 0.275 gal pump Deleted Delayed LNAPL recovery
(LNAPL) stroke, Blackhawk pump
EWP13 11 gpm, 0.275 gal pump Renamed EWP12, 11 gpm, 11 gpm, 0.275 gal pump stroke, QED Hammerhead No testing of different
(DNAPL) stroke, Blackhawk pump 0.275 gal pump stroke, pump pump performance could
Blackhawk pump occur
EWP14 11 gpm, 0.275 gal pump Deleted Inability to test pump
stroke, Blackhawk pump performance at these site
EWP15 11 gpm, 0.275 gal pump conditions
stroke, Blackhawk pump
EWP16 11 gpm, 0.275 gal pump
Air RSAC-1 (for 89 CFM delivery, 100 psig, 89 CFM delivery, 100 psig, 80 CFM, 125 psig, 20 hp (with 100 SCFM air dryer) No impact
compressors extraction well 20 hp (with 100 SCFM air 20 hp (with 100 SCFM air
pumps) dryer) dryer)
RSAC-2 89 CFM delivery, 100 psig, Deleted No impact on pilot;
(backup) 20 hp (with 100 SCFM air however no backup
dryer)
Piping from well field 3-inch to 2-inch main line 3-inch main line No impact
Liquid Treatment
Treatment TPFP-1 78.5 gpm, 55 ft head, 2 hp 78.5 gpm, 55 ft head, 2 hp 78.5 gpm, 125 ft head, 5 hp, TACO VM0202B No backup. Material
Plant Feed Grundfos TP50-240 ODP Grundfos TP50-240 ODP imcompatibility in seals
Pumps exists. Installed pump has
TPFP-2 78.5 gpm, 55 ft head, 2 hp Deleted no alternative seals
(backup) Grundfos TP50-240 ODP available.
TPFP-3 20 gpm, 49 ft head, 0.75 hp 20 gpm, 49 ft head, 0.75 hp 20 gpm, 180 ft head, 0.75 hp TACO VM0202B No backup. Material
Grundfos CR8-20/IU ODP Grundfos CR8-20/IU ODP imcompatibility in seals
exists. Installed pump has
no alternative seals
available.
Heat HX-1 2.5 MBTU/hr 50 gpm, both sides 50 gpm, both sides; 50 gpm; EPDM gaskets Material incompatibility
Exchangers EPDM gaskets resulted in a bypass around
(Plate and HX-2 2.5 MBTU/hr 50 gpm, both sides 50 gpm, both sides; 50 gpm; EPDM gaskets equipment
Frame type) EPDM gaskets
Cooling tower Wet induced draft, 425 Deleted Limited the ability to use
gpm less groundwater and
reduced ability to reject
heat; however no impact on
the pilot.
Float skimmer Retrofitted at tank T-401 Deleted No impact
Dissolved air flotation tank New unit downstream of T- New unit, 120 gpm capacity, 120 gpm capacity, 85 sf flotation area with recycle No impact
(Replaced depurator) 401 85 sf floatation area with
recycle
Additional bioreactor Convert existing clarifier by Deleted Reduced treatment capacity
adding spargers
Dissolved air flotation tank New unit, downstream of Deleted Reduced treatment capacity
(DAF-205) additional bioreactor
Caustic chemical tank 1000 gallons (auto feed) 1000 gallons (manual feed) Not used therefore no
impact. However, manual
operation would have
required additional labor
Acid chemical tank 500 gallons (auto feed) 500 gallons (manual feed) Not used therefore no
impact. However, manual
operation would have
required additional labor
New Piping w/in Treatment 85 ft HDPE piping between T- 85 ft HDPE piping between T- 85 ft HDPE piping between T-402 and aeration basin No impact
Plant 402 and aeration basin (T- 402 and aeration basin (T- (T-203)
203) 203)
New Multimedia filter (T- 6-ft. diameter Deleted Reduced treatment capacity
206C)
Tank T-303 Replaced with new tank Repaired No impact
(damaged in 2/28/01
earthquake)
Flow Measurement
Flow meters 19 in extraction wells 13 in extraction wells (stroke 7 in extraction wells (stroke counters) Stroke counters not as
counters) accurate as flow meters
None specified for product 1 at inlet to product tank 1at inlet to product tank No impact
tank
1 in Condensate line None specified 1at HX-1 and HX-2 outlet; 1 at HX-3 condensate No impact
1: Equipment in process order except for flow meters

2: Impacts shown are due to design changes throughout project and not by impacts due to site conditions
Table 4.2-2 Steam Generation and Conveyance System
Equipment 1 Design Analysis Contract Initial installation Final Installation Impacts on Pilot2
Steam Generation
Water Well Pump (DWP-1) 400 gpm, 500 head, 5 in 400 gpm, 200 head, 5 in 225 gpm, 200 head, 4 in diameter column, 215 ft setting depth No impact
diameter column, 75 hp diameter column, 200 ft in well, 30 hp
setting depth in well, 30 hp
Deaerator DFP-1 109 gpm, 92 ft head, 5 hp, Deleted No backup; No impact on
Feed Pumps Grundfos HS150 5050 ODP pilot
DFP-2 50 gpm, 77ft head, 2 hp, 50 gpm, 77ft head, 2 hp, 80 gpm, 180 ft head, 7.5 hp, TACO VM0603B Operational problems due
Grundfos CR8-2OU Grundfos CR8-2OU to oversize pump
(increased labor for
operation and
maintenance)
DFP-3 10 gpm, 36 ft head, 0.5 hp, 10 gpm, 36 ft head, 0.5 hp, 80 gpm, 180 ft head, 7.5 hp, TACO VM0603B Operational problems due
Grundfos CR2-20/IU ODP Grundfos CR2-20/IU ODP to oversize pump
operation and
maintenance)
Feed Water Tank size 36X70 in. 36x72 in. 36x60 in. No impact
Softeners SFT-1 100 gpm and 8 psi drop at 97 gpm and 15 psi drop at 175 gpm and 15 psi drop at continuous flow. No impact
continuous flow continuous flow.
125 gpm and 25 psi drop at 250 gpm and 25 psi drop at peak flow.
peak flow.
SFT-2 100 gpm and 8 psi drop at 97 gpm and 15 psi drop at 175 gpm and 15 psi drop at continuous flow. No impact
continuous flow continuous flow.
125 gpm and 25 psi drop at 250 gpm and 25 psi drop at peak flow.
peak flow.
Brine tank 39x48 in tank; 2500 lb salt 39x48 in tank; 2050 lb salt 39x60 in tank; 1900 lb salt capacity No impact
capacity capacity
Deaerator 25,000 lbs/hr rating, 600 gal tank, 3714 lbs/hr steam flow No impact
Air BAC-1 (for Air deliver and pressure as Air deliver and pressure as 177 CFM, 125 psig, 40 hp Excessive operator time
compressors boiler) needed, 7.5 hp needed, 7.5 hp to maintain performance;
needed redesign
BAC-2 Air deliver and pressure as Deleted No impact; however no

(backup) needed, 7.5 hp backup
Table 4.2-2 Steam Generation and Conveyance System
Boiler Feed BFP-1 50 gpm, 160 ft head, 5 hp 50 gpm, 160 ft head, 5 hp 84 gpm, 227 ft head, 7.5 hp, EPDM O-rings, Grundfos CR16- Operational problems due
Pumps Grundfos CR8-4OU ODP Grundfos CR8-4OU ODP 4OU ODP to oversize pump
operation and
maintenance)
BFP-2 10.6 gpm, 181 ft head, 1.5 Deleted No impact
hp Grundfos CR2-5OU ODP
BFP-3 50 gpm, 160 ft head, 5 hp 10.6 gpm, 181 ft head, 1.5 hp 84 gpm, 227 ft head, 7.5 hp, EPDM O-rings, Grundfos CR16- Operational problems due
Grundfos CR8-4OU ODP Grundfos CR2-5OU ODP 4OU ODP to oversize pump
operation and
maintenance)
Boiler 25,000 lb/hr 800 hp, 27,600 lb/hr, 86% 800 hp, 27,600 lb/hr, 86% efficiency No impact
efficiency
Steam Conveyance
Piping to well field 6-inch to 3-inch black steel, 6-inch to 5-inch black steel, 6- inch black steel, insulated No impact
insulated insulated
Expansion Joints 47, 3-inch to 8-inch 35, 3-inch to 6-inch diameter 12, 3-inch to 6-inch diameter No impact
diameter.
Flow Measurement
Flow meters 25 at injection wellheads 16 at injection wellheads 16 at injection wellheads (manual combination calibrated flow Increased time for labor
(automated averaging pitot valves) to read and record.
tubes);
None specified at boiler inlet 1 at boiler inlet (Water into 1 at boiler inlet (automated meter on water softening sytem; No accurate measure of
boiler) (averaging pitot tube) however was not calibrated during pilot) water flow to boiler
None specified at boiler 1 at boiler outlet (steam to 1 at boiler outlet (orifice plate - could not calibrate during No accurate measure of
outlet injection wells) (pressure pilot and therefore not used) steam flow from boiler
compensated differential
pressure or orifice plate)
Table 4.2-3 Vapor Treatment System
Vapor cap
Vapor collection layer 8-inch gravel layer with 8-inch gravel layer with a 6-inch 8-inch gravel layer with a 6-inch protective soil layer Unknown impact;
a 6-inch protective soil protective soil layer Excessive perforation in
layer vapor barrier; Because no
Vapor collector laterals 4-inch diameter slotted 3-inch diameter slotted steel pipe 4-inch diameter slotted steel pipe soil partitions
steel pipe constructed, no
Vapor barrier HDPE geomembrane HDPE geomembrane with 12-inch HDPE geomembrane with 12-inch layer of select fill above experimentation of vapor
with a 12-inch layer of select fill above geomembrane. Geomembrane perforated due to instrument extraction rates occurred.
protective soil layer geomembrane and well installation occurring after vapor barrier placement
above geomembrane
Vapor cap partitions Soil partition cells Soil partition cells Not installed
Vapor conveyance
Piping from well field 8-inch to 6-inch main, 6-inch black steel with no insulation; 6-inch black steel with no insulation; piping sloped toward The combination of a
black steel with piping sloped toward each pipe one low point resulting in concurrent flow of condensate and single low point, a single
insulation support vapor. over-sized condensate
Vapor Line Condensate 4 at drip legs at main 4 at drip legs at main pipe anchors 1 at drip leg in vapor main (low point) near boiler building pump, and the pipe
Receiver pipe anchors sloping in the vapor
conveyance system
Condensate PDP-1, vapor 4, 0.5 gpm at 50 psi 4, Positive Displacement Pumps; 0.5 1, Centrifugal pump; 15 gpm, 18 psi head, 0.33 hp allowed vapor line to fill
pump extraction main head, 0.5 hp gpm, 50 psi head, 0.5 hp with water which entered
vapor treatment system.
Oil Sealed LRVP-1 450 ACFM capacity, 25 450 ACFM capacity, 25 hp 450 ACFM capacity, 25 450 ACFM capacity, 25 HP, Material incompatibility
Liquid Ring hp HP, EPDM seals Teflon seals (Seals replaced after impacted pump
Vacuum pilot) performance
Pumps
LRVP-2 450 ACFM capacity, 25 450 ACFM capacity, 25 hp (became 450 ACFM capacity, 25 450 ACFM capacity, 25 HP, Not operated during pilot;
hp backup) HP, EPDM seals EPDM seals no impact
LRVP-3 450 ACFM capacity, 25 Deleted No impact; however no
(backup) hp backup
Condensate PDP-3, vacuum 1, 0.5 gpm at 50 psi 1, Positive Displacement Pumps; 0.5 Not installed (piping configuration revised eliminating this Unknown impact; the
pump pump header head, 0.5 hp gpm at 50 psi head, 0.5 hp pump) addition of this pump
could have prevented
liquid from entering the
vacuum pump
LRVP condensate receiver Not in original design Not installed 15 gal tank; 10 gpm @ 25 ft head Unknown impact
(drains LRVP filters) (installed after pilot)
Table 4.2-3 Vapor Treatment System
Vapor Treatment
Heat HX-3 2.5 MBTU/hr Plate and Frame or Shell and Tube Plate and Frame Type, 50 Shell and Tube type, 50 gpm The following
Exchangers Type, 50 gpm liquid side, 2,500 gpm liquid side, 2,500 liquid side, 2,851 lb/hr vapor side contributed to the
lb/hr vapor side lb/hr vapor side, EPDM (installed after pilot study ended) eventual vapor system
gaskets shut down: Liquid in the
HX-4 (backup) 2.5 MBTU/hr Deleted vapor system; Material
incompatibility in the
Condensate PDP-2, heat 2, 30 gpm, 50 psi head, 2, Positive Displacement Pumps; 2, Centrifugal pumps; 30.0 gpm, 100 psi head, 1.5 hp (these heat exchanger and
pump exchanger (HX-3 hp 30.0 gpm, 50 psi head, 3.0 hp pumps became named CP-2A and CP-2B) pumps; condensate tank
3) too small for plate and
HX-3 condensate tank 2 ft x2 ft x2 ft (60 gal) 2 ft x2 ft x2 ft (60 gal) tank (initially 15 gal tank; 10 gpm @ 25 30-inch diameter x 48-inch length frame type heat
tank to be used as a condensate receiver) ft head (47 gal) steel vacuum rated exchanger.
(installed after pilot - to be used as
a separator (knockout tank))
Vapor GAC sorber Steam regenerated. Deleted No impact

Used only when boiler
is not in operation.
Thermal Oxidizer (replaced Not in original design 250 SCFM (not operated 250 SCFM (installed after pilot) No impact
GAC sorber) during pilot)
Flow Measurement
Flow meters 14 at extraction wells 8 at extraction wells (flow calculated 7 at extraction wells (flow calculated from pressure No flow information
from pressure differentials from differentials from pressure valves - not installed) ; 1 at EW-4 available during pilot
pressure valves);1 at EW-4 (automated averaging pitot tube);
(automated averaging pitot tube);
8 at vapor cap branch 8 at vapor cap branch (TASCO flow 8 at vapor cap branch (TASCO flow control valves) No impact
control valves);
1 for non-condensable 1 at HX-3 inlet (calibrated 1 at HX-3 inlet (averaging pitot tube); 1 at HX-3 outlet No flow information
gases combination flow control valve); 1 (averaging pitot tube) (HX-3 meters were installed after pilot; available during pilot
at HX-3 outlet (calibrated not calibrated.)
combination flow control valve)
ACRONYMS used in Tables 4.2-1 through 4.2-3
acfm Actual cubic feet per minute
BAC Boiler air compressor
BFP Boiler feed pump
cfm Cubic feet per minute
DFP Dearator feed pump
DNAPL Dense non-aqueous phase liquid
DWP Deep water pump
EPDM Ethylene propylene diamine monomer
EW Extraction well
EWP Extraction well pump
GAC Granular activated carbon
gal Gallons
gpm Gallons per minute
HDPE High density polyethylene
hp Horsepower
HX Heat exchanger
LNAPL Light non-aqueous phase liquid
LRVP Liquid ring vacuum pump
MBTU/hr Million British Thermal Units/hr
ODP Open drip proof
PDP Positive displacement pump
psi Pounds per square inch
RSAC Rotary screw air compressor
scfm Standard cubic feet per minute
sf Square feet
SFT Softener
TPFP Treatment plant feed pump
NOTE: Acronyms not defined here are manufacturer names
TABLE 4.2-4 MASS-AVERAGED COMPONENT PROPERTIES AT 25 C
Vapor pressure
Component group Solubility (mg/L) (Pa)
LPAH1 2910 14
HPAH2 2 3
Aliphatic 52 47
1
Light Polycyclic Aromatic Hydrocarbons - Includes the substituted-PAH
compounds.
2
Heavy Polycyclic Aromatic Hydrocarbons
TABLE 4.3-1 PILOT AREA WELL FIELD AUTOMATED MONITORING SUMMARY
Media Locations Parameter Frequency Implemented?
Subsurface Injection Wells (16) Temperature Automated Yes

Soil (Thermocouples) Collected
Instrument Strings (55) continuously
Stored daily
Extraction Wells (7) Temperature Automated Yes

[Fiber optic Distributed Collected
Instrument Strings (10) Temperature Sensor (DTS) continuously
System] Stored daily
Instrument Strings (1) Temperature Automated Yes

(Thermocouples and DTS) As above for
thermocouples and
DTS
Groundwater Instrument Strings (9) Not Water Level Automated Yes

co-located with DTS or (Vibrating Wire Pressure Collected
Thermocouples Transducer) continuously
Stored daily
Water Level Automated Yes

(Vibrating Wire Pressure Collected
Transducer) continuously
Stored daily
Extraction Wells (7)
Flow (Stroke Counters) Automated Yes
Strokes counted &
stored hourly
Flow calculated &
stored daily
TABLE 4.3-2 INJECTION AND EXTRACTION SYSTEM MONITORING SUMMARY
Steam Injection Wells (16) Pressure/Flow Daily Yes

(Pressure gauges
around orifice valve)
From boiler to Temperature Daily Yes

injection wells
Pressure/Flow To be Daily, Yes
(Orifice plate) but readings
unreliable
Extracted vapor Extraction Wells (7) Pressure/Flow Daily No

(Pressure gauges
Vapor Cap Branch Pressure/Flow Daily Yes

(Pressure gauges
Vapor inlet to HX-3 Pressure/Flow Daily Yes

(Pitot tube)
Vapor outlet from Pressure/Flow Daily Yes

HX-3 (Pitot tube)
Temperature Daily Yes
Condensed Condenser Total condensate Daily No

liquid production
Vapor line and HX- Combined flow Daily No

3 drip leg (flow meter)
condensate
HX-3 condensate to Flow (flow meter) Daily No

GWTP
Condenser (HX3) Temperature drop Daily Yes

across
Non- Between the heat PAH, PCP, total Weekly No

condensable exchangers and the hydrocarbons, CO2
gases vapor treatment and O2
system
Extracted liquid Combined liquid Total Organic Carbon Automated Yes

line in boiler (Originally
building before the every 20
condenser minutes;
changed to
every 30
minutes and
then every
hour)
Extracted liquid Extraction wells (7) Flow (Stroke Counter) Daily Yes
Semivolatile organics Daily Yes
Dissolved Oxygen and Daily Yes

Carbon Dioxide
Temperature Daily Yes
Inlet to Product Flow (Flow meter) Yes

Tank in boiler
building
Outlet of HX-1, Flow (Flow meter) Yes

HX-2 to GWTP
Inlet/outlet Temperature Yes

HX-1/HX-2
TABLE 4.3-3 TREATMENT SYSTEM MONITORING SUMMARY
NAPL Product Tanks: Volume Weekly Yes

T-105 and T-108
Treatment Plant SP-0: Treatment Oil & Grease Weekly Yes

Plant Inlet
SP-1: PAH/PCP Weekly

Equalization
Tank (T-401)
Outlet
SP-3: DAF-104 Oil & Grease Weekly

Outlet
SP-4: T-402 PAH/PCP Weekly

Effluent
Temperature Daily
Mass Flow Rate Daily
SP-5: Aeration ML TS & VS Weekly

Tank (T-203)
Digester TS & VS Weekly
RAS TS & VS Weekly
Temperature Daily
D.O. Daily
PH Weekly
SP-6: Clarifier TS Weekly

Effluent (T-204)
VS Weekly
COD Weekly
TPH Weekly
PAH/PCP Weekly
NH3 Weekly
Orthophosphate, dissolved Weekly
SP-8: Multi- TS Weekly

Media Filter
Effluent (T- TPH Weekly
206A, T-206B)
PAH/PCP Weekly
SP-9: Lead TPH Weekly

Carbon Filter
PAH/PCP Weekly
SP-10: Lag TPH Weekly

Carbon Filter
PAH/PCP Weekly
GWTP Effluent SP-11: Effluent PAHa Daily effluent Yes

storage tank sampling during
PCPa weeks 1 and 2.
Discharge Flow Rate Twice weekly

sampling for week
TSSa 2 to 3 months
TDS
Temperature
pH
Dissolved Oxygen
PAHa
Acute survival test - Menidia Annually

beryllina (Inland Silversides)
Chronic test - Mytilus Sp. (blue Quarterly

mussel) or Crassostrea gigas
(Pacific oyster)
Groundwater Upper Aquifer: Alkalinity One round of Yes

Monitoring MW-17, MW- sampling in
18, MW-19, Total Organic Carbon November 2002
EW-04, EW-06,
EW-08 Nitrate/Nitrite
Sulfate/Sulfide
Chloride
Total Metals (Ca, Mn, Mg, Na and

K)
Petroleum Hydrocarbons
Semivolatile Organics with TICs
Lower Aquifer: Alkalinity Three rounds of Yes

CW-05, CW-09, sampling in
CW-15, 99CD- Total Organic Carbon November 2002,
MW02, 99CD- December 2002,
MW04, 02-CD- Nitrate/Nitrite and January 2003
MW-01
Sulfate/Sulfide
Chloride
Petroleum Hydrocarbons
Semivolatile Organics with TICs
Key to Parameters
PCP = Pentachlorophenol VS = Volatile solids
COD = Chemical oxygen demand NH3 = Ammonia as nitrogen
TSS = Total suspended solids TPH = Total Petroleum Hydrocarbon
D.O. = Dissolved oxygen pH = Hydrogen ion
PAH = Polynuclear aromatic hydrocarbon ML = Mixed liquor
RAS = Return activated sludge
a
24-hour composite sample.
TABLE 4.3-4 WORKER SAFETY AND PERIMETER MONITORING SUMMARY
Noise Various: on and Four times during No
off site dBA first 6 mos of
operations
Air Quality 2 monitoring Total Suspended Particles Samples will be No
stations will be collected every 24
placed around hours. Baseline
the perimeter of PCP sampling followed
the treatment by three monitoring
plant (one PAH events once
upwind of groundwater
operational Naphthalene temperatures
activity and one stabilize.
downwind of the
area).
Worker Safety Treatment Plant PAH and PCP Samples to be Yes
Air Quality DAF-104 collected once at
startup and once
Treatment Plant
during operations
T-203
Treatment Area
Boiler Air Stack Dioxins/Furans, PAHs, Volatile Up to 12 samples No
Emissions Organics, Semivolatile Organics, throughout
Total Hydrocarbons, Hydrogen operations
chloride, Chlorine, Particle Size
Joint J09, J10, J11 Semivolatile organics (including Within three No
Observation naphthanols and quinones) months of sheet
Well pile wall
Groundwater installation and
after active steam
injection
TABLE 5.2-1 SUPPORTING GROUPS AND CONTRACTORS
Support Groups/
Subcontractor Service
Bay West, Inc. Installation of Sheet Pile Walls
SCS, Inc./OMI Operations and maintenance contractor
MarVac, Inc. Construction of the boiler building and the subsurface

(belowground) portion of the pilot studys elements,
including the installation of the water supply well, and the
steam injection and extraction wells
Pease Construction, Inc. Pilot system construction and start up
URS Corporation Sampling and analysis of perimeter, source and fugitive air
emissions not implemented
SteamTech Environmental Services Expert consultant services for steam injection operations
Sensa, Inc. Installation support for DTS
Holt Soil boring and monitoring well installations

FASP/Techlaw
Shannon & Wilson/ESN Northwest
Environmental Resource Associates Performance evaluation samples
USACE Subsurface instrumentation and ADAS
USEPA Region 10 CLP Groundwater analysis for water supply well and extraction
well sampling
USEPA Region 10 Manchester Analysis of Groundwater Treatment Plant Monitoring

Environmental Laboratory samples
USEPA Office of Research and Soil analysis for microbiological baseline testing (microcosm
Development (Battelle Memorial Institute studies and phospholipid fatty acid (PLFA) analyses)
and Microbial Insights, Inc.)
TABLE 5.2-2 PROJECT WEB PAGE CONTENTS
Data Source Data Type
Subsurface Data: Temperature: 3D Views
Vertical Sections
Horizontal Slices
Individual Well Profiles
Pressure: Water Level Contours
Time History
Injection Well Data: Injection Rate Summary Table
Rate and Mass of Steam
Cumulative Rate and Mass of Steam
Extraction Well Data: Vapor Flow Rate
Temperature
Dissolved Oxygen and Carbon Dioxide of Extracted Liquid
PAH/PCP
Automated Pumping Rates
Contractor Pumping Rates
Extraction Rate Summary Table
Vapor Collector Lines: Pressure
Flow
Heat Exchanger: Flow
Temperature
Thermal Remediation Totals: Total Organic Carbon
Liquid Flow Rate
Contaminants
Reports: Operations Report This report included the daily hours of
operation for the boiler, treatment plant, vapor extraction
system, injection wells, and extraction wells. This report
also included meteorological data.
Operations Log Daily operations meeting minutes were
posted on this message board by the Team Coordinator.
Additionally, updates on the status of particular systems
and repairs were posted by technical team members.
Message Board Technical discussions on monitoring or
system performance were posted on this message board by
technical team members.
Links shown as bold italics were never activated, either because instruments were
not installed (or not monitored) or a format was not developed for posting to the website
before operations ceased.
TABLE 6.1-1 AVERAGE STEAM INJECTION RATES FOR EACH INJECTION WELL DURING THE
TWO MAJOR INJECTION PERIODS
Oct 30-Dec 14 Jan 12-Mar 22

Well (lb/hr) (lb/hr)
I-1 193 67
I-2 233 132
I-3 244 81
I-4 311 102
I-5 275 84
I-6 224 111
I-7 327 210
I-8 366 136
I-9 373 220
I-10 266 203
I-11 160 80
I-12 143 189
I-13 193 199
I-14 378 213
I-15 302 198
I-16 199 150
Table 6.6-1. Groundwater Treatment Plant Data Summary
SP-3:
SP-0: Combined Extracted DAFT SP-6: Clarifier SP-8: Multi-Media Filter Carbon Treatment
Liquid Influent to GWTP Compliance Levels DAFT Efficiency Aeration Basin Efficiency
Effluent SP-4: T-402 Effluent Effluent Effluent SP-9: Lead Carbon SP-12: Mid Carbon SP-10: Lag Carbon SP-11: Treated Effluent System Efficiency
2113 2102 2103 2104 2105 2106 2116 2107 2108
Week Total PAH PCP O&G O&G Total PAH PCP Total PAH PCP PCP Total PAH PCP Total PAH PCP Total PAH PCP Total PAH PCP Temp. D.O. Total PAH PCP Temp. D.O. O&G Removal PAH Removal PAH Removal PCP Removal PAH PCP
Date Number (ug/L) (ug/L) (mg/L) (mg/L) (ug/L) (ug/L) (ug/L) (ug/L) Total PAH (ug/L) (ug/L) (ug/L) (ug/L) (ug/L) (ug/L) (ug/L) (ug/L) (ug/L) (ug/L) pH (deg. C) (mg/L) (ug/L) (ug/L) pH (deg. C) (mg/L) (%) (%) (%) (%) Removal Removal
10/7/02 41 9443.59 300 6.21 5.71 499.09 460 0.858 7.2 0.796 3.3 0.14 0.12 -- -- 0.01 0.074 0.01 0 7.96 15.6 3.9 20 6 6-9 < 25 >6 8.05 94.72 99.83 98.43 99.99 99.98
10/14/02 42 11028.33 340 6.41 5.86 618.57 340 0.35 22 0.211 12 0.21 0.27 -- -- 0.01 0.074 0.01 0.0088 8.24 13.6 3.9 20 6 6-9 < 25 >6 8.58 94.39 99.94 93.53 99.95 99.99
10/21/02 43 9803.71 220 5.65 5.93 717.56 290 0.929 7.3 0.444 3.2 0.19 0.11 -- -- 0.052 0.074 0.01 0.0048 7.92 15.8 3.8 20 6 6-9 < 25 >6 -4.96 92.68 99.87 97.48 99.88 99.98
10/28/02 44 13266.02 310 6.75 5.75 398.22 320 1.871 6.6 0.526 2.2 0.273 0.11 -- -- 0.05 0.074 0.01 0.074 7.91 13.2 4.1 20 6 6-9 < 25 >6 14.81 97.00 99.53 97.94 99.90 99.97
11/4/02 45 12106.72 330 5.6 5.41 424.65 400 2.615 14 0.35 6.7 0.243 0.23 -- -- 0.121 0.074 0.01 0.074 8.07 10.5 4.7 20 6 6-9 < 25 >6 3.39 96.49 99.38 96.50 99.65 99.99
11/11/02 46 11648.97 270 10.7 6.1 4041.61 290 3.992 9.8 1.634 4.1 0.26 0.0059 -- -- 0.091 0.0069 0.01 0.074 8.26 13.8 4.2 20 6 6-9 < 25 >6 42.99 65.31 99.90 96.62 99.94 100.00
11/18/02 47 36467 260 34.9 22.9 21301.200 240 7.344 11 3.404 9.1 0.59 0.0064 -- -- 0.327 0.074 0.15 0.074 7.79 13.8 4.7 20 6 6-9 < 25 >6 34.38 41.59 99.97 95.42 99.90 99.99
11/25/02 48 24697.4 220 14.9 10.8 11830.800 250 5.498 14 2.1 10 0.33 0.014 -- -- 0.01 0.0084 0.01 0.074 7.75 15.8 3.8 20 6 6-9 < 25 >6 27.52 52.10 99.95 94.40 100.00 100.00
12/2/02 49 2628477.5 220 1570 50.7 32181.600 260 5.021 15 2.989 13 0.23 0.012 -- -- 0.01 0.074 5.84 0.074 7.73 13.5 3.8 20 6 6-9 < 25 >6 96.77 98.78 99.98 94.23 100.00 99.99
12/9/02 50 133908 190 205 92.2 86415.700 250 365.2 84 628.99 79 24.12 7 -- -- 10.114 0.23 2.459 0.083 7.38 13.8 4.1 20 6 6-9 < 25 >6 55.02 35.47 99.58 66.40 99.98 100.00
12/16/02 51 27069.800 270 28.6 13.1 75339 290 1945.04 190 0.52 180 132.97 52 -- -- 5.187 0.29 6.55 0.36 7.31 14.8 3.1 20 6 6-9 < 25 >6 54.20 -178.31 97.42 34.48 90.03 100.00
12/23/02 52 24244.4 347 31.8 140 5037.02 385 4.095 154 3.788 146 0.01 0.0895 -- -- 0.01 0.0681 0.01 0.0851 7.74 11.3 4.2 20 6 6-9 < 25 >6 -340.25 79.22 99.92 60.00 100.00 100.00
12/30/02 1 25924.9 440 17 7.38 7839.67 440 3.21 220 3.1 0 0.613 0.051 -- -- 0.526 0.029 0.542 0.034 7.92 10.4 5.8 20 6 6-9 < 25 >6 56.59 69.76 99.96 50.00 99.83
1/6/03 2 109017.9 610 103 729 52132.4 510 7.31 120 1.997 110 0.209 0.029 -- -- 0.087 0.021 0.93 0.031 7.91 12.1 4.8 20 6 6-9 < 25 >6 -607.77 52.18 99.99 76.47 99.96 100.00
1/13/03 3 42316.6 570 54.4 13.3 33866.1 660 8.95 81 4.69 73 1.297 0.22 -- -- 0.01 0.047 0.17 0.064 7.54 14.2 3.4 20 6 6-9 < 25 >6 75.55 19.97 99.97 87.73 100.00 100.00
1/20/03 4 24,876.30 680 20.1 30.5 32190.3 700 8.79 15 6.31 0 3.296 0.42 -- -- 0.635 0.036 2.8 0.063 7.52 15.1 4 20 6 6-9 < 25 >6 -51.74 -29.40 99.97 97.86 99.90
1/27/03 5 31589.4 760 34.5 21.5 31017 560 5.649 18 3.15 10 1.238 0.82 -- -- 0.39 0.046 1.59 0.041 7.55 16.7 5.5 20 6 6-9 < 25 >6 37.68 1.81 99.98 96.79 99.88 100.00
2/3/03 6 78047.6 1000 76.8 694 66412.1 930 5.94 11 5.94 11 1.73 0.98 -- -- 1.25 0.053 2.276 0.066 7.45 17 3.7 20 6 6-9 < 25 >6 -803.65 14.91 99.99 98.82 99.79 100.00
2/10/03 7 29700.8 550 18.9 19.5 32846.39 970 5.24 9 3.723 4.8 2.154 0.97 -- -- 1.36 0.053 0.68 0.058 7.56 14.5 4.3 20 6 6-9 < 25 >6 -3.17 -10.59 99.98 99.07 99.63 99.99
2/17/03 8 64357.7 900 90 17.2 33835.3 1100 3.84 9.2 2.561 7.5 1.001 1.4 -- -- 0.7 0.053 0.69 0.062 7.7 14.1 4.7 20 6 6-9 < 25 >6 80.89 47.43 99.99 99.16 99.73 99.99
2/24/03 9 51327.2 920 50.3 34.1 41505 1000 3.35 9.7 0 6.7 1.288 0.11 -- -- 0.32 0.014 1.4 0.051 7.81 13.7 4.9 20 6 6-9 < 25 >6 32.21 19.14 99.67 99.03 100.00
3/3/03 10 113134.4 1000 67.4 26.7 35742.7 910 7.254 9.7 5.645 5.8 2.438 0.066 -- -- 1.895 0.016 0.01 0.0063 7.72 14.1 4.9 20 6 6-9 < 25 >6 60.39 68.41 99.98 98.93 99.66 100.00
3/10/03 11 35476.3 720 36.6 21.3 30386 1100 3.606 5.9 2.963 5.7 1.496 0.073 -- -- 0.237 0.0061 0.29 0.073 7.78 15.3 4.5 20 6 6-9 < 25 >6 41.80 14.35 99.99 99.46 99.92 100.00
3/17/03 12 55207.69 630 87.8 394 65211.1 850 9.973 3 6.93 3.2 2.17 0.12 -- -- 0.7 0.0084 1.12 0.074 7.49 16.2 4.1 20 6 6-9 < 25 >6 -348.75 -18.12 99.98 99.65 99.90 100.00
3/24/03 13 36445.4 1000 46.9 21.8 33601.2 950 4.1 14 3.49 12 2.09 0.13 -- -- 1.54 0.0043 1.2 0.074 7.61 16.8 4.7 20 6 6-9 < 25 >6 53.52 7.80 99.99 98.53 99.56 100.00
3/31/03 14 65675.9 970 101 113 26952 770 4.651 9.5 3.109 9 3.58 0.23 -- -- 1.24 0.011 1.25 0.073 7.69 18.7 4 20 6 6-9 < 25 >6 -11.88 58.96 99.98 98.77 99.60 100.00
4/7/03 15 28350.2 970 20.6 46.4 24080.4 920 2.74 24 3.17 25 2.55 0.35 -- -- 0.33 0.074 2.374 0.079 7.94 15 5.1 20 6 6-9 < 25 >6 -125.24 15.06 99.99 97.39 99.90 100.00
4/14/03 16 28378 700 23.5 21.6 22390.9 870 2.52 20 2.67 17 1.68 0.21 -- -- 0.85 0.004 0.88 0.02 7.65 16 4.1 20 6 6-9 < 25 >6 8.09 21.10 99.99 97.70 99.68 100.00
4/21/03 17 27519.8 520 21.9 10.9 13816.8 660 3.587 13 1.72 17 1.89 13 -- -- 0.01 0.073 0.818 0.074 8.05 15.6 4 20 6 6-9 < 25 >6 50.23 49.79 99.97 98.03 99.99 100.00
4/28/03 18 29030.8 630 23.2 10.1 12580.92 880 1.4 9.2 1.07 5.2 0.63 0.0055 -- -- 0.54 0.074 0.01 0.074 7.98 15.7 4.4 20 6 6-9 < 25 >6 56.47 56.66 99.99 98.95 99.50 99.99
5/5/03 19 18903.1 930 14.5 8.9 13700 1200 0.24 3.1 1.32 1.3 0.0077 0.0077 -- -- 0.074 0.074 0.01 0.074 7.79 16.3 4.8 20 6 6-9 < 25 >6 38.62 27.53 100.00 100.00 99.94 99.94
5/12/03 20 21411.1 530 30.1 10.4 10839.5 780 1.77 1.9 1.83 0.81 0.73 0.064 -- -- 0.01 0.073 0.2 0.073 7.57 16.9 3.5 20 6 6-9 < 25 >6 65.45 49.37 99.98 99.76 99.99 99.91
5/19/03 21 23816.4 620 16 8 7561.72 990 0.81 2.5 0.46 1 0.29 0.38 1.39 0.0066 0.01 0.074 0.01 0.073 7.81 15.4 4.3 20 6 6-9 < 25 >6 50.00 68.25 99.99 99.75 99.98 99.93
5/26/03 22 17818.3 560 57.9 6.8 3691.74 770 0.3 3.1 1 1.9 0.53 0.049 0.09 0.074 0.01 0.074 0.01 0.025 7.74 18 4.4 20 6 6-9 < 25 >6 88.26 79.28 99.99 99.60 99.99 99.96
6/2/03 23 21475.2 600 22.6 7 6814.81 610 4.185 13 0.87 12 0.46 0.048 0.14 -- 0.01 0.074 0.01 0.0039 7.72 18.4 4.5 20 6 6-9 < 25 >6 69.03 68.27 99.94 97.87 99.99 99.99
6/9/03 24 37119.8 600 23.8 5.9 5094.05 600 0.86 2.5 1.03 1.6 0.19 0.057 0.01 0.0079 0.01 0.074 0.01 0.0078 7.88 20.4 4.5 20 6 6-9 < 25 >6 75.21 86.28 99.98 99.58 99.99 99.95
6/16/03 25 11091.4 450 13.9 5.8 1514.83 600 1.663 2.8 0.559 1.8 0.01 0.046 0.22 0.0041 0.01 0.074 0.01 0.0076 7.84 19.3 4.5 20 6 6-9 < 25 >6 58.27 86.34 99.89 99.53 99.98 99.96
6/23/03 26 32063.6 560 14.6 5.3 675.82 500 0.53 2.5 0.733 1.8 0.379 0.032 0.18 0.004 0.12 0.074 0.41 0.0093 7.75 18.4 5.8 20 6 6-9 < 25 >6 63.70 97.89 99.92 99.50 99.84 99.96
6/30/03 27 19238.2 430 13.7 7.3 810.57 510 1.78 4.8 0.49 2.1 0.031 0.031 0.13 0.074 0.075 0.075 0.973 0.0094 7.74 20.1 4.5 20 6 6-9 < 25 >6 46.72 95.79 99.78 99.06 99.85 99.96
7/7/03 28 17212.1 450 9.6 5 660 670 2.06 4.4 0.66 1.9 0.22 0.031 0.13 0.0044 0.17 0.074 0.088 0.0096 7.81 19.7 4.1 20 6 6-9 < 25 >6 47.92 96.17 99.69 99.34 99.74 99.96
7/14/03 29 17768.5 560 18.8 9.4 688.03 530 2.4 2.8 0.41 1.6 0.21 0.03 9.17 0.011 0.14 0.074 0.01 0.0035 7.78 21 4.2 20 6 6-9 < 25 >6 50.00 96.13 99.65 99.47 99.66 99.95
7/21/03 30 14477.77 430 9.1 5.1 652.54 600 1.96 1.7 0.4 1.5 0.21 0.047 0.14 0.073 0.01 0.074 0.01 0.074 7.84 21.8 4 20 6 6-9 < 25 >6 43.96 95.49 99.70 99.72 99.98 99.95
7/28/03 31 13737.4 720 13.3 5.1 589.61 650 1.37 2.7 0.14 2 0.037 0.037 0.01 0.0055 0.074 0.074 0.24 0.074 7.79 21.1 3.9 20 6 6-9 < 25 >6 61.65 95.71 99.77 99.58 99.47 99.96
8/4/03 32 20452.6 550 28.6 6.4 556.42 650 1.19 4 0.13 3.9 0.01 0.033 0.01 0.073 0.01 0.073 0.01 0.073 7.9 20.4 5.1 20 6 6-9 < 25 >6 77.62 97.28 99.79 99.38 99.92 99.98
8/11/03 33 152153.2 490 8.4 6.6 491.35 540 0.34 4.5 0.6 3.2 0.22 0.039 0.16 0.0046 0.15 0.075 0.15 0.074 7.9 18.9 6.9 20 6 6-9 < 25 >6 21.43 99.68 99.93 99.17 99.75 99.98
8/18/03 34 13661.9 450 31.2 3.2 602.44 450 1.92 11 0.64 9.6 0.83 0.032 0.31 0.0067 0.16 0.073 0.14 0.0087 7.8 21.1 4 20 6 6-9 < 25 >6 89.74 95.59 99.57 97.56 99.75 99.99
8/25/03 35 20460.1 510 18.3 7.1 682.84 550 1.18 9.8 0.35 8 0.12 0.04 0.01 0.011 0.01 0.0076 0.01 0.074 8 20 3.6 20 6 6-9 < 25 >6 61.20 96.66 99.83 98.22 99.97 100.00
9/1/03 36 12767.1 470 10.2 8.6 751.63 420 1.36 11 0.24 7.9 0.11 0.043 0.17 0.012 0.01 0.014 0.01 0.0072 7.9 20 4.3 20 6 6-9 < 25 >6 15.69 94.11 99.82 97.38 99.96 100.00
9/8/03 37 23331.6 450 30.4 4.2 633.36 390 0.62 14 0.28 9.8 0.11 0.046 -- -- 0.01 0.074 0.01 0.007 7.9 20 3.7 20 6 6-9 < 25 >6 86.18 97.29 99.90 96.41 99.96 99.99
9/15/03 38 13470.2 340 11.6 5 563.43 420 1.02 22 0.15 21 0.01 0.063 0.01 0.015 0.01 0.074 0.01 0.0052 8 17 3.3 20 6 6-9 < 25 >6 56.90 95.82 99.82 94.76 99.93 100.00
9/22/03 39 77840.8 480 146 5 840.8 370 0.38 16 0.36 13 0.23 0.075 0.01 0.074 0.21 0.0049 0.8 0.0056 7.9 17 4.7 20 6 6-9 < 25 >6 96.58 98.92 99.95 95.68 99.42 100.00
9/29/03 40 12309.17 340 7 5 566.27 380 0.45 18 0.036 16 0.01 0.099 0.01 0.0095 0.01 0.0076 0.24 0.074 7.8 19 3.2 20 6 6-9 < 25 >6 28.57 95.40 99.92 95.26 99.72 100.00
10/6/03 41 30697.5 470 16.5 7.4 521.47 410 0.44 14 0.1 14 0.01 0.17 0.01 0.011 0.01 0.074 0.01 0.074 8 17 3.5 20 6 6-9 < 25 >6 55.15 98.30 99.92 96.59 99.90 99.99
10/13/03 42 9394.94 340 5 5 281.86 450 0.01 19 0.11 15 0.01 0.28 0.17 0.0043 0.01 0.073 0.01 0.073 7.9 20 6.6 20 6 6-9 < 25 >6 0.00 97.00 100.00 95.78 99.91 100.00
10/20/03 43 20228.2 470 10.9 5 752.34 440 0.24 16 0.21 14 0.01 0.29 0.01 0.074 0.01 0.0055 0.01 0.074 8 17 3.2 20 6 6-9 < 25 >6 54.13 96.28 99.97 96.36 99.95 100.00
10/27/03 44 40244.7 780 52.2 5.2 1050.04 750 0.01 32 0.47 24 0.18 0.43 0.01 0.0047 0.01 0.074 0.27 0.0048 7.8 16 4.3 20 6 6-9 < 25 >6 90.04 97.39 100.00 95.73 99.98 100.00
11/3/03 45 18928.9 740 25.8 5 714.57 660 0.9 96 0 83 0.01 3.4 0.01 0.012 0.01 0.0093 0.01 0.0049 8.1 12 4.1 20 6 6-9 < 25 >6 80.62 96.22 99.87 85.45 100.00
11/10/03 46 27564 560 34.9 7.9 2781.3 650 0.61 82 0.7 65 0.47 3.2 0.01 0.012 0.01 0.009 0.01 0.009 7.7 14 4.1 20 6 6-9 < 25 >6 77.36 89.91 99.98 87.38 99.99 100.00
11/17/03 47 20720.7 480 20.5 5 576.84 480 1.63 67 0.73 83 0.22 3.9 0.01 0.0092 0.01 0.0056 0.21 0.0074 7.91 15 3.7 20 6 6-9 < 25 >6 75.61 97.22 99.72 86.04 99.99 100.00
11/24/03 48 20768.9 670 27.9 7.1 953.49 840 0.81 180 0.33 180 0.36 21 0.18 0.033 0.01 0.012 0.01 0.0075 7.88 13 5.5 20 6 6-9 < 25 >6 74.55 95.41 99.92 78.57 99.97 100.00
12/1/03 49 25305.1 1000 51.8 6.8 1704.3 860 0.78 85 0.809 99 0.19 9.1 0.21 0.026 0.42 0.075 0.01 0.0069 7.8 12 5.3 20 6 6-9 < 25 >6 86.87 93.26 99.95 90.12 99.48 100.00
12/8/03 50 14521.1 650 25.8 5 14468.33 870 0.01 88 0.98 69 0.34 7.3 0.33 0.029 0.01 0.011 0.39 0.0064 7.7 12 4.7 20 6 6-9 < 25 >6 80.62 0.36 100.00 89.89 99.99 100.00
12/15/03 51 22055.9 810 25.9 11.9 3236.8 820 0.11 72 1 50 0.59 5.8 0.71 0.034 0.28 0.011 0.26 0.037 7.8 12 4.5 20 6 6-9 < 25 >6 54.05 85.32 100.00 91.22 99.72 100.00
12/22/03 52 19708.2 660 15.9 8.9 1201.61 730 2.35 28 0.15 31 0.25 3.8 0.18 0.033 0.13 0.062 0.27 0.022 7.7 13 4.1 20 6 6-9 < 25 >6 44.03 93.90 99.80 96.16 99.13 100.00
12/29/03 1 16597.9 790 22.6 7 957.74 680 1.5 29 0.65 25 0.24 3.7 0.57 0.028 0.26 0.0088 0.1 0.00084 7.8 11 5.2 20 6 6-9 < 25 >6 69.03 94.23 99.84 95.74 99.60 100.00
1/5/04 3 21332.2 700 15.4 5 991.75 900 0.78 110 0.33 85 0.16 17 0.52 0.11 0.093 0.018 0.14 0.0003 7.7 12 4 20 6 6-9 < 25 >6 67.53 95.35 99.92 87.78 99.72 100.00
1/12/04 4 29478.4 600 14.3 6.7 1994.07 620 1.05 16 1.07 14 0.49 2.8 0.16 0.053 0.16 0.0055 0.01 -- 7.68 14 3.7 20 6 6-9 < 25 >6 53.15 93.24 99.95 97.42 99.85 100.00
Note: Total PAH does not include those values with a U qualifier
Figures
Figure 1.0-1. Site Vicinity Map.
Figure 3.2-1. Location map showing profiles through the Pilot Study Area.
229420
99SE13 99SE39
99SE14
229400
A'
I-01 E-02
00PP21 C'
229380 E-01
99SE16 I-02
I-0500PP20
I-06 99SE38
99SE17
229360
E-06A
I-03 I-04 E-06B
She
99SE26
229340 99SE18
e
E-04
t Pi
I-07
le W
229320
B E-07
E-03
all
97AP19 I-11 I-09 99SE19
l
al
W I-10
00PP19
229300 I-08
I-12
ile
I-16
tP
I-13 E-05
ee
E-08
Sh
229280
97AP28 B'
E-10
00PP22 E-09
E-13 I-14 I-15
99SE45
I-17 I-18
229260 I-20
97AP20
I-19 E-12
I-24 I-25 E-11

229240 I-21
I-23
97AP21
229220
97AP36 I-22
229200
0 15 30 45 60 97AP37
A 00PP09 C
229180
1228900 1228920 1228940 1228960 1228980 1229000 1229020 1229040 1229060 1229080 1229100 1229120 1229140 1229160 1229180 1229200 1229220 1229240 1229260
Figure 3.5-1. Wyckoff/Eagle Harbor Superfund Site Monitoring Well Network
229850
CW07
CW06
229800
229650
OB-3-2
MW14
RPW3
CW05
229750
CW15 OB-3-1 EW12
229640
CW08
229480
EWC3
229700 MW15
229475 OB-2-2
RPW2
RPW1 OB-2-1 CW04
229630
229470 OB-3-2 1229270 1229280 1229290 1229300 1229310
229650 RPW3
OB-3-1
EW12
OB-1-2 EWC3
229465 MW16
RPW5
OB-1-1 PO01
229600 229460
1229050 1229060 1229070 1229080 CW09
229550
PW8
99CD-MW02
PO03
229500 PO13
PO09 MW15
RPW1 229300
CW14 OB-1-2
OB-1-1
Northing (NAD 83)
CW13
PW9
229450 CW10
PO05 OB-4-4
229290
99CD-MW04
MW17 CW12 OB-4-3
229400 E-02
PO04 RPW4
229280 OB-4-2
E-01
EW03 EW07
RPW6
MW18 OB-4-1
02CD-MW01
E-06
229350 229270
E-04
E-07 E-03
1229370 1229390 1229410
229300 E-05 OB-4-4
OB-4-3
EW08 RPW4
OB-4-2
OB-4-1
RPW7
EW11
PO18
EWC2
CW02
229250 CW03
229270
229200 7
11
W
RP EW
229260
229150 18
PO
MW23 02 C2
CW EW
MW19 MW21
229100 229250
03
CW
229050
229240
1229440 1229450 1229460 1229470
0 30 60 90 120
01CT01
229000
1228400 1228450 1228500 1228550 1228600 1228650 1228700 1228750 1228800 1228850 1228900 1228950 1229000 1229050 1229100 1229150 1229200 1229250 1229300 1229350 1229400 1229450 1229500 1229550
Easting (NAD 83)
Figure 4.0-1 Construction Timeline
Prime Month/Year 000-July 200 Aug-01 Sep-01 Oct-01 Nov-01 Dec-01 Jan-02 Feb-02 Mar-02 Apr-02 May-02 Jun-02 Jul-02 Aug-02 Sep-02 Oct-02
Contractor Task Date All 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22 1 8 15 22
Site Infrastructure
Bay West Sheet Pile Wall 11/01/00-2/14/01
MarVac Vapor Cap Installation 8/29/01-9/24/02
MarVac Boiler Room Access Pad 1/2/02-1/7/02
MarVac Boiler Building Erection 1/2/02-2/27/02
MarVac Fuel Storage Tank Pads 1/10/02-1/16/02
MarVac Water Storage Tank Pad 1/10/02-1/16/02 BEGIN
MarVac Piping to TP 1/18/02-1/23/02 STEAMIN
MarVac Pump House Shed 1/27/02-2/24/02
MarVac Water Well Utilities 10/17/01-4/1/02
MarVac Building Foundation Slab 11/6/01-12/23/01
MarVac Electrical Utilities 12/21/01-1/7/02
Pease PSE Power to Mechanical Bldg 9/10/2002
Pease Transformer 6/25/2002
Steam Generation/Conveyance
MarVac Water Well Installation/Development 8/20/01-1/14/02
MarVac Water Well Pump 8/20/01-1/14/02
MarVac Injection Wells 10/31/01-1/28/02
MarVac Remove Exposed West Dock Pilings 11/15/01-11/17/01
Pease Water Tank 6/13/02-7/11/02
Pease Expansion Joints 6/25/02-7/24/02
Pease Air Compressors 6/25/02-7/29/02
Pease St. Inj. Piping to Well Field 6/26/02-7/24/02
Pease Steam Injection Piping 6/26/02-7/24/02
Pease Boiler 6/26/02-8/8/02
Pease Blow Down Tank 7/18/02-9/11/02
Pease Water Well/Feed Piping 7/23/02-8/16/02
Pease Boiler Feed Pumps 7/30/02-7/31/02
Pease Deaerator Feed Pumps 7/30/02-7/31/02
Pease Treatment Plant Feed Pumps 7/30/02-8/9/02
Pease Boiler Stack 8/6/02-9/10//02
Pease Deaerator 8/12/02-8/16/02
Pease Feed Water Softeners 9/2002
Installed at
various times
prior to and
Pease Flow Meters during pilot
Liquid Conveyance/Treatment
MarVac Extraction Wells 9/2/01-5/6/02
Pease Tank T-303 2/7/02-2/15/02
Pease Piping from Well Field 4/16/02-9/10/02
Pease Extraction Well Pumps 6/10/02-6/17/02
Pease New TP Piping 7/19/02-8/22/02
Pease Heat Exchangers 8/1/02-9/4/02
Pease Caustic Chemical Tank 8/14/02-8/21/02
Pease Acid Chemical Tank 8/14/02-8/21/02
Pease DAF-104 (Dissolved Air Flot. Tank) 8/23/02-9/16/02
Pease Multimedia Filter 9/2002
Installed at
various times
prior to and
Pease Demolish T-101, Sep-102 in TP 8/9/2002
Vapor Treatment
Pease Vapor Collector Laterals 5/9/02-7/22/02
Pease Piping from Well Field 4/16/02-7/29/02
Pease Sour Gas Line Condensate Receiver 7/12/02-7/17/02
Pease Sour Gas Piping 7/12/02-8/23/02
Pease Condensate Pumps 7/30/02-7/31/02
Pease Heat Exchanger (HX-3) 8/1/02-9/4/02
Pease HX-3 Condensate Tank 8/1/02-9/4/02
Pease Oil Sealed LRV Pumps 6/28/02-7/30/02
Pease Thermal Oxidizer 8/16/02-8/23/02
Installed at
various times
prior to and
Pease Pilot Area Piping 4/19/02-8/20/02
Water
Deaerator Boiler
Softeners
To
Water from well
Injection
Wells
Steam Generation Process Flow Diagram
Thermal
Emissions
Oxidizer
From vapor
extraction wells Heat Exchanger
and vapor HX3
collector
Boiler Emissions
Condensate to WTP
Vapor Treatment Process Flow Diagram
FIGURE 4.2-5 STEAM AND VAPOR

PROCESS FLOW DIAGRAMS
Phase I
Systems 1-Oct-02 2-Oct-02 3-Oct-02 4-Oct-02 5-Oct-02 6-Oct-02 7-Oct-02 8-Oct-02 9-Oct-02 10-Oct-02 11-Oct-02 12-Oct-02 13-Oct-02 14-Oct-02 15-Oct-02 16-Oct-02 17-Oct-02 18-Oct-02 19-Oct-02 20-Oct-02 21-Oct-02 22-Oct-02
Liquid (gpm) 5 0 1 6 5 5 5 5 5 5
Extraction
Vapor
Steam
Boiler Records incomplete; sporadic operations
Injection (lbs/hr)
Treatment Plant On
DTS On On On Off Off On On On On On On On On On Off Off On On On Off On
Thermocouples On On On Off Off On On On On On On On On On Off Off On On On On On
Monitoring
TOC
Pressure Off On On On Off Off On Off On On On On On On On Off Off On On On On On
Liquid
Extraction
Vapor
Boiler
Steam
Injection
402 pump seized
402 pump seized

Treatment Plant
DTS
Thermocouples
Monitoring TOC
Pressure
Phase I Phase II
Systems 23-Oct-02 24-Oct-02 25-Oct-02 26-Oct-02 27-Oct-02 28-Oct-02 29-Oct-02 30-Oct-02 31-Oct-02 1-Nov-02 2-Nov-02 3-Nov-02 4-Nov-02 5-Nov-02 6-Nov-02 7-Nov-02 8-Nov-02 9-Nov-02 10-Nov-02 11-Nov-02 12-Nov-02 13-Nov-02 14-Nov-02 15-Nov-02
Liquid (gpm) 4 3 4 2 4 9 8 8 4 5 6 13 16 19 7 10 14 26 24
Extraction
Vapor Training Training Training Training Training Training Training On On On Off-On Off Off-On On On Off-On
Boiler Training Training Training Training Training Training Training On On On Off-On Off Off-On On On On
Steam
Injection (lbs/hr) 778 5893 7649 1195 4955 958 7134 6264 8100 11935 0 2574 5266 5287 4323
Treatment Plant On On On On On On On On On On On On On On On On On On
DTS Off On On On On On Off On On On On On On Off On On On On Off Off On On On On
Thermocouples On On On On On On On On On On On On On On On On Off On On On On On On On
Monitoring
TOC On On On On On On Off
Pressure On On On On On On On On On On On On On On On On Off On On On On On On On
Liquid
Extraction
vacuum pumps
receiver failed
receiver failed
Condensate
Condensate
Liquid in
Vapor
Feed water pump trip; boiler

Feed water pump trip
chem feed system

Boiler
Steam
Injection
Treatment Plant
problems
Software
DTS
Thermocouples
Programming
issue
Monitoring TOC
sensor inop
sensor inop
sensor inop
sensor inop
E-3 temp
E-3 temp
E-3 temp
E-3 temp
Pressure
Phase II
Systems 16-Nov-02 17-Nov-02 18-Nov-02 19-Nov-02 20-Nov-02 21-Nov-02 22-Nov-02 23-Nov-02 24-Nov-02 25-Nov-02 26-Nov-02 27-Nov-02 28-Nov-02 29-Nov-02 30-Nov-02 1-Dec-02 2-Dec-02 3-Dec-02 4-Dec-02 5-Dec-02 6-Dec-02 7-Dec-02 8-Dec-02 9-Dec-02
Liquid (gpm) 19 21 22 21 20 20 16 22 20 21 14 13 14 15 15 20 21 20 17 20 20 20 22
Extraction
Vapor Off-On Off-On On On Off On On On Off Off Off Off-On Off-On Off-On On On On Off-On Off Off-On On On On Off-On
Boiler Off-On Off-On On On On On On On Off Off Off Off-On Off-On Off-On On On On Off-On Off Off-On On On On Off-On
Steam
Injection (lbs/hr) 0 3942 5809 3864 3863 3837 3846 5182 0 0 0 5036 0 0 4988 5189 5264 5261 0 0 5671 7276 7276 7609
Treatment Plant On On On On On On On On On On On On On On On On On On On On On On On Off-On
DTS On Off On On On On On On On On On On On On On Off On On On On On On On On
Thermocouples On On On On On On On On On On On On On On On Off On On On On On On On On
Monitoring
TOC Off Off Off On On On On On On On On On On On On Off On On On On On On On On
Pressure On On On On On On On On On On On On On On On Off On On On On On On On On
E-1 fouled with solidified

Crystal formation in E-1
Waxy PAH fouled cone
screen in product tank
E-3 and E-5 pumps

E-1 and E-7 pump
maintenance
exhaust port
lowered 2 ft
PAH
Liquid
Extraction
Field condensate
Liquid in vapor
vacuum system
Crystal fouling
differential in
in condenser
pump seals
Pressure
system
failed
Vapor
Sticking sour gas valve due to

Condensate pumps failed due
chem incomp; fuel imbalance
Condensate pumps failed due

damages gaskets in boiler
Sticking valve in sour gas
Sticking valve in sour gas

to chem incompatibility
to chem incompatibility

Boiler
Steam
Injection
Mixing chamber between AB and

clarifier clogged; solids spilled
Some die-off in aeration basin
downstream
Treatment Plant
DTS
Thermocouples
Monitoring TOC
Pressure
Phase II Phase III
Systems 10-Dec-02 11-Dec-02 12-Dec-02 13-Dec-02 14-Dec-02 15-Dec-02 16-Dec-02 17-Dec-02 18-Dec-02 19-Dec-02 20-Dec-02 21-Dec-02 22-Dec-02 23-Dec-02 24-Dec-02 25-Dec-02 26-Dec-02 27-Dec-02 28-Dec-02 29-Dec-02 30-Dec-02 31-Dec-02 1-Jan-03 2-Jan-03
Liquid (gpm) 4 4 11 18 23 22 7 0 0 0 0 0 0 0 0 0 0
Extraction
Vapor Off Off-On Off-On On On Off-On Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off
Boiler Off Off-On Off-On On On Off-On Off Off Off Off Off Off Off Off Off Off Off Off-On Off Off Off Off Off Off-On
Steam
Injection (lbs/hr) 0 0 7847 7795 8364 0 0 0 0 0 0 0 0 0 0 0 0
Treatment Plant Off Off-On On On On On Recycle Recycle On On On On On On On On On On On On On On On On
DTS On On On On On Off On On On On On On On On On Off On On Off Off On On Off On
Thermocouples On On On On On On On On On On On On On On On On On On On On On On On On
Monitoring
TOC On On On On On On On Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off
Pressure On On On On On On On On On On On On On On On On On On On On On On On On
Liquid
Extraction
Field condensate
pump air-locked
Seals in vacuum
exchanger (HX-
LRVP seal torn

Gaskets in heat
pumps failed;
3) failed
Vapor
Condensate receiver pump
Condensate receiver pump

failed
failed
Boiler
Steam
Injection
Low DO in AB; recycle mode with no
Recycle mode with no discharge to

Nearly complete die-off in AB
discharge to outfall
Treatment Plant outfall

problems
Sporadic
software
Loop 2
down
DTS
Thermocouples
Monitoring TOC
T-22 pressure
T-22 pressure
T-22 pressure
No readings
all sensors
inop
inop
inop
Pressure
Phase III Phase IV
Systems 3-Jan-03 4-Jan-03 5-Jan-03 6-Jan-03 7-Jan-03 8-Jan-03 9-Jan-03 10-Jan-03 11-Jan-03 12-Jan-03 13-Jan-03 14-Jan-03 15-Jan-03 16-Jan-03 17-Jan-03 18-Jan-03 19-Jan-03 20-Jan-03 21-Jan-03 22-Jan-03 23-Jan-03 24-Jan-03 25-Jan-03 26-Jan-03
Liquid (gpm) 4 3 16 11 29 31 4 30 17 25 22 20 23 17 18 16 16 14 15 19 19 20
Extraction
Vapor Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off
Boiler Off Off Off Off Off Off Off Off-On Off Off Off-On On On On On On On On On On On On On On
Steam
Injection (lbs/hr) 0 0 0 0 0 0 0 0 3253 3223 3192 3146 2787 2801 2833 2830 2778 2789 2751 2772 2837 2788
Treatment Plant On On On On On On On On On On On On On On On On On On On On On On On On
DTS Off Off Off On On On On On Off Off On On On On On Off Off Off On On On On Off Off
Thermocouples Off Off Off On On On On On On On On On On On On On On On On On On On On On
Monitoring
TOC Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off On On On On On On On
Pressure Off Off Off On On On On On On On On On On On On On On On On On On On On Off
E-7 exhaust continues to

E-3 pump fouled with
Stroke counters not

counting properly
crystalline PAH
foul
Liquid
Extraction
Vapor
Boiler
Steam
Injection
Depressed biological activity
Depressed biological activity

Declining DO
Declining DO
Treatment Plant
DTS
Thermocouples
Monitoring TOC
T-22 pressure
T-22 pressure
T-22 pressure
No readings
No readings
No readings
all sensors
all sensors
all sensors
inop
inop
inop
Pressure
19-Feb-03
TOC failed due

2374
T-29 pressure
Off
On
On
On
On
On
On
20
E-3 exhaust air/vapor lock T401 flow valve not working to chemical
inop
incompatability
18-Feb-03
E-3 exhaust plugged with

2265
T-29 pressure
Off
On
On
On
On
On
On
17
PAH? Suspected E-5 T401 flow valve not working

inop
discharge tube rupture
17-Feb-03
2248
T-29 pressure
Off
Off
On
On
On
On
On
12
inop
16-Feb-03
TOC pump
2137
T-29 pressure
Off
Off
On
On
On
On
On
13
clogged w/
inop
PAH
15-Feb-03
2143
T-29 pressure
Off
Off
On
On
On
On
On
12
inop
14-Feb-03
2261
T-29 pressure
Off
On
On
On
On
On
On
15
inop
13-Feb-03
2266
Product tank and liq heat TOC clogged T-29 pressure

Off
On
On
On
On
On
On
13
Feed water pump trip T401 flow valve not working

exchangers clogged w/ PAH crystals inop
12-Feb-03
2271
T401 flow valve not working; clarifier T-29 pressure

Off
On
On
On
On
On
On
14

pump not working inop
11-Feb-03
2294
T-29 pressure
Off
On
On
On
On
On
On
18
Feed water pump trip Clarifier pump not working

inop
10-Feb-03
2570
T-29 pressure
Off
On
On
On
On
On
On
19
inop
9-Feb-03
2359
Heat exchanger plugged Feed water flow control valve T401 flow meter plugged with PAH T-29 pressure
Off
Off
On
On
On
On
On
19
with PAH failed; feed water pump trip crystals; clarifier pump not working inop
8-Feb-03
2359
T-29 pressure
Off
Off
On
On
On
On
On
19
Phase IV
inop
7-Feb-03
2548
T-29 pressure
Off
On
On
On
On
On
On
20

inop
6-Feb-03
2545
T-29 pressure
Off
On
On
On
On
On
On
20
Feed water pump trip Clarifier pump failed

inop
5-Feb-03
2498
T-29 pressure
Off
On
On
On
On
On
On
23
T401 flow meter not working

inop
4-Feb-03
Crystals
2796
T-29 pressure
Off
On
On
On
On
On
On
23
T401 flow meter not working forming in TOC

inop
well
3-Feb-03
2764
T-29 pressure
Off
On
On
On
On
On
On
26
inop
2-Feb-03
2699
T-29 pressure
Off
Off
On
On
On
On
On
24
inop
1-Feb-03
2672
T-29 pressure
Off
Off
On
On
On
On
On
25
inop
31-Jan-03
2750
Off
On
On
On
On
On
On
25
30-Jan-03
2785
Off
On
On
On
On
On
On
26
Unstable DO
29-Jan-03
2726
Off
On
On
On
On
On
On
28
28-Jan-03
2726
Off
On
On
On
On
On
On
23
27-Jan-03
2824
E-7 exhaust continues to

Off
Off
On
On
On
On
On
18
foul
Injection (lbs/hr)
Thermocouples
Thermocouples
Liquid (gpm)
Injection
Pressure
Pressure
Liquid
Boiler
Boiler
Vapor
Vapor
TOC
TOC
DTS
DTS
Treatment Plant
Treatment Plant
Systems
Monitoring
Monitoring
Extraction
Extraction
Steam
Steam
15-Mar-03
2295
T-29 pressure
Off
On
On
On
On
On
On
22
inop
14-Mar-03
2257
T-29 pressure
Off
On
On
On
On
On
On
21
inop
13-Mar-03
2263
T-29 pressure
Off
On
On
On
On
On
On
21
DAF recycle pump clogging

inop
12-Mar-03
2267
T-29 pressure
Off
On
On
On
On
On
On
21
inop
11-Mar-03
2278
T-29 pressure
Off
On
On
On
On
On
On
21
inop
10-Mar-03
2247
T-29 pressure
Off
On
On
On
On
On
On
21
inop
9-Mar-03
2156
T-29 pressure
Off
Off
On
On
On
On
On
20
inop
8-Mar-03
2156
T-29 pressure
Off
Off
On
On
On
On
On
21
E-3 exhaust air/vapor lock

inop
7-Mar-03
2255
T-29 pressure
Off
On
On
On
On
On
On
19
Corrosion noted in DA tank

inop
6-Mar-03
Off-On
Off-On
T-29 pressure
Off
On
On
On
On
8
inop
5-Mar-03
2291
T-29 pressure
Off
On
On
On
On
On
On
20
DAF pump system clogged

inop
4-Mar-03
2299
T-29 pressure
Off
On
On
On
On
On
On
19
Liquid line to TP plugged T401 pumping efficiency decrease TOC clogged

Phase IV
inop
3-Mar-03
2303
T-29 pressure
Off
On
On
On
On
On
On
18
T401 pumping efficiency decrease

inop
2-Mar-03
2311
TOC sampler T-29 pressure

Off
Off
On
On
On
On
On
20
clogged inop
1-Mar-03
2328
T-29 pressure
Off
Off
On
On
On
On
On
20
inop
28-Feb-03
2298
T-29 pressure
Off
On
On
On
On
On
On
20
inop
27-Feb-03
2271
Seals failed on heat T-29 pressure

Off
On
On
On
On
On
On
20
exchanger (HX-2) inop

26-Feb-03
2264
Skimmers in clarifier not keeping up TOC drain part T-29 pressure

Off
On
On
On
On
On
On
20

with solids production disintegrated inop
25-Feb-03
2267
T401 flow valve clogged with waxy TOC sampling T-29 pressure
Off
On
On
On
On
On
On
20

substance arm clogged inop
24-Feb-03
Waxy buildup decreasing efficiency of

2266
T-29 pressure
Off
On
On
On
On
On
On
20
Feed water pump trip DAF pump; T401 flow valve cleaned
inop
but still inop
23-Feb-03
2187
T-29 pressure
Off
Off
On
On
On
On
On
20
inop
22-Feb-03
2187
T-29 pressure
Off
Off
On
On
On
On
On
20
inop
21-Feb-03
2266
T-29 pressure
Off
On
On
On
On
On
On
20
T401 flow valve not working

inop
20-Feb-03
Backwash pump P206 motor failed;

2270
T-29 pressure
Off
On
On
On
On
On
On
21
T401 flow valve inop; DAF

inop
pumps/valves fouling with PAH
Injection (lbs/hr)
Thermocouples
Thermocouples
Liquid (gpm)
Injection
Pressure
Pressure
Liquid
Boiler
Boiler
Vapor
Vapor
TOC
TOC
DTS
DTS
Treatment Plant
Treatment Plant
Systems
Monitoring
Monitoring
Extraction
Extraction
Steam
Steam
Phase IV Phase V
Systems 16-Mar-03 17-Mar-03 18-Mar-03 19-Mar-03 20-Mar-03 21-Mar-03 22-Mar-03 23-Mar-03 24-Mar-03 25-Mar-03 26-Mar-03 27-Mar-03 28-Mar-03 29-Mar-03 30-Mar-03 31-Mar-03 1-Apr-03 2-Apr-03 3-Apr-03 4-Apr-03 5-Apr-03 6-Apr-03 7-Apr-03 8-Apr-03
Liquid (gpm) 22 23 22 24 24 21 20 20 18 20 19 18 18 18 18 17 10 10 8 12 11 10 11 10
Extraction
Vapor Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off Off
Boiler On Off-On Off-On Off-On On On On Off-On Off-On Off-On Off-On Off Off Off Off Off Off Off-On Off-On Off Off Off Off Off
Steam
Injection (lbs/hr) 2252 0 2240 2246 2206 2222 2169 0 0 0 2222 0 0 0 0 0 0 0 0 0 0 0 0 0
Treatment Plant On On On On On On On On On On On On On On On On On On On On On On On On
DTS On On On On On On On On On On On On On On On On On On On On On On On On
Thermocouples On On On On On On On On On On On On On On On On On On On On On On On On
Monitoring
TOC On On On On On On On On On On On On On On On On On On On On On On On On
Pressure On On On On On On On On On On On On On On On On On On On On On On On On
2" slit in E-5 discharge

No discharge at E-5
tube
Liquid
Extraction
Vapor
Low water alarm trip; air/fuel

ratio problem causing boiler
Compressor line air leak
Boiler fuel igniter inop

Compressor oil leak
panting
Boiler
Steam
Injection
DAF recycle pump completely clogged
with NAPL wax
Treatment Plant
DTS
Thermocouples
disintegrated
TOC PVC
drain part
Monitoring TOC
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-29 pressure
T-22 pressure
T-22 pressure
T-22 pressure
T-22 pressure
T-22 pressure
T-29 pressure
T-29 pressure
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
inop
Pressure
Phase V
Systems 9-Apr-03 10-Apr-03 Notes
Liquid (gpm) 10 9 Records incomplete
Extraction
Vapor Off Off 5 Records incomplete (gpm operating condition if known)
Boiler Off Off-On Training Training
Steam
Injection (lbs/hr) 0 1804 9 System operating (gpm or lbs/hr operating condition)
Treatment Plant On On On System operating
DTS On On Off System not operating
Thermocouples On On 0 System not operating (0 gpm or lbs/hr operating condition noted)
Monitoring
TOC On On Off-On System turned on or off during day
Pressure On On 1647 System turned on or off during day (gpm or lbs/hr operating condition)
E-3 exhaust line fouling

Liquid
Extraction
Vapor
Boiler
Steam
Injection
Treatment Plant
DTS
Thermocouples
Monitoring TOC
T-22 pressure
inop
Pressure
Figure 6.1-1. Total steam production rates calculated based on water usage, fuel
usage, and well-head measurements
Energy injected based on fuel usage Energy injected based on feed water flow Energy injected based on well steam flows
14
Phase I Phase II Phase III Phase IV Phase V Phase VI
12
10
Energy injection rate (million BTU/hr)
0
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-2. Steam injection rates for all 16 injection wells combined
14,000
12,000
Total steam injection rate (lbs/hr)
10,000
8,000
6,000
4,000
2,000
0
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-3. Steam injection rates for the 16 injection wells during 2002
operations. No steam was injected after 12/16/02 that year
I-01 I-02 I-03 I-04 I-05 I-06 I-07 I-08
I-09 I-10 I-11 I-12 I-13 I-14 I-15 I-16

1,000
Phase I Phase II Phase
III
900
800
Steam injection rate (lbs/hr)
700
600
500
400
300
200
100
0
10/22 10/27 11/1 11/6 11/11 11/16 11/21 11/26 12/1 12/6 12/11 12/16
Figure 6.1-4. Steam injection rates for the 16 injection wells during 2003
operations
I-01 I-02 I-03 I-04 I-05 I-06 I-07 I-08
I-09 I-10 I-11 I-12 I-13 I-14 I-15 I-16

300
Phase IV Phase V Phase VI
250
Steam injection rate (lbs/hr)
200
150
100
50
0
1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30
Figure 6.1-5. Cumulative steam injection volumes for each of the 16 injection
wells
I-01 I-02 I-03 I-04 I-05 I-06 I-07 I-08
I-09 I-10 I-11 I-12 I-13 I-14 I-15 I-16

900

800
Cumulative energy injected (MM BTU)
700
600
500
400
300
200
100
0
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-6. Liquid extraction rates from each of the seven extraction wells.
E-1 E-2 E-3 E-4 E-5 E-6 E-7

8

7
6
Pumping rate (gpm)
0
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-7. Cumulative water volumes extracted from each extraction well
E-1 E-2 E-3 E-4 E-5 E-6 E-7
900,000

800,000
700,000
Cumulative volume (gallons)
600,000
500,000
400,000
300,000
200,000
100,000
0
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-8. Total liquid extraction rates from the seven wells
25
20
Total liquid pumping rate (gpm)
15
10
0
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-9. Water levels observed in extraction wells during operation
E-1 E-2 E-3 E-4 E-5 E-6 E-7
15
10
Water level (ft MLLW)
-5
-10
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-10. Water levels observed in monitoring well locations during
operation
T-12 T-14 T-22 T-25 T-27 T-29 T-40 T-46 T-63
15

13
11
9
Water level (ft MLLW)
-1
-3
-5
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-11. Water flow rates for the pilot test area
Steam injection rate Liquid extraction rate Condensate production rate
Total extraction rate Net extraction rate
25
20
15
Flow rate (gpm)
10
-5
-10
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-12. Cumulative water balance for the pilot test area
Water injected as steam Water extracted Condensate produced from vapor
Total extracted Net extraction Water in pilot test volume
Water net inflow
3,500,000

3,000,000
Estimated from water level measurements
2,500,000
Cumulative volume (gallons)
2,000,000
1,500,000
1,000,000
500,000
-500,000
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-13. Enthalpy fluxes for the pilot test area during operation.
Injected Extracted in water Extracted as steam Non-condensable gas

Total extracted Net enthalpy flux Total heat loss
14

12
10
Enthalpy flux (million BTU/hr)
-2
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-14. Energy balance for the pilot test area during operation
Steam energy injected Energy extracted in water
Energy in extracted steam Energy in extracted non-condensable vapor
Total energy extracted Net energy added to site
Energy stored based on measured temperature Total heat loss
10,000

9,000
8,000
Cumulative energy (million BTU)
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.1-15. Cumulative energy losses estimated based on temperature
gradients at the boundaries of the pilot test volume
Through top Through bottom Through sheet-pile Total heat loss
3,500
3,000
Cumulative energy (million BTU)
2,500
2,000
1,500
1,000
500
0
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.2-1. Temperatures of the liquid extracted from six of the seven extraction
wells. No temperature measurements were made for E-4
E-1 E-2 E-3 E-5 E-6 E-7 All EWS

180
160
140
Temperature (F)
120
100
80
60
40
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.2-3. Comparison of temperature profiles from DTS and
thermocouples for instrument string T-7.
Instrument String T-7

Oct 2, 2002 Dec 14, 2002
Elevation (ft MLLW)
Elevation (ft MLLW)

25
20
Temperature (deg C)
15
10
5
0
20
15
10
-10
-15
-20
5
-5
Elevation (ft MLLW)
Thermocouple DTS
Figure 6.2-4. Interpolated temperature maps for pseudo-horizonal Slices A-D and horizontal Slice E.
Phase I Phase II Phase III Phase IV
28 Oct 02 04 Nov 02 11 Nov 02 18 Nov 02 25 Nov 02 02 Dec 02 10 Dec 02 16 Dec 02 23 Dec 02 30 Dec 02 06 Jan 03 13 Jan 03 21 Jan 03 27 Jan 03
Collector
Layer
Slice E
Vadose Zone
Slice D
Slice C
Aquifer
Slice B
Aquitard Slice A
03 Feb 03 10 Feb 03 17 Feb 03 24 Feb 03 03 Mar 03 10 Mar 03 17 Mar 03 24 Mar 03 31 Mar 03 07 Apr 03
Not
Slice E Available
Temp (deg C)
Slice D 110
100
90 Phase I Commissioning
80 Phase II Continuous steam injection
70
Phase III Engineering evaluations
Slice C 60
Phase IV Low-level steam injection
50
40 Phase V Sporadic low-level steam injection
30 Phase VI Liquid extraction (not shown)
20
10
Slice B
Slice A
Phase IV (cont.) Phase V

Figure 6.2-5. Interpolated temperature profiles along Sections C and F.
Section C
Phase I Phase II Phase III
C 28 Oct 02 C' 04 Nov 02 11 Nov 02 18 Nov 02 25 Nov 02 02 Dec 02 10 Dec 02 16 Dec 02 23 Dec 02
Phase III (cont.) Phase IV

30 Dec 02 06 Jan 03 13 Jan 03 21 Jan 03 27 Jan 03 03 Feb 03 10 Feb 03 18 Feb 03 24 Feb 03

03 Mar 03 10 Mar 03 17 Mar 03 24 Mar 03 31 Mar 03 07 Apr 03
Section F
Phase I Phase II Phase III
F 28 Oct 02 F' 04 Nov 02 11 Nov 02 18 Nov 02 25 Nov 02 02 Dec 02 10 Dec 02 16 Dec 02 23 Dec 02
Phase III (cont.) Phase IV

30 Dec 02 06 Jan 03 13 Jan 03 21 Jan 03 27 Jan 03 03 Feb 03 10 Feb 03 18 Feb 03 24 Feb 03

03 Mar 03 10 Mar 03 17 Mar 03 24 Mar 03 31 Mar 03 07 Apr 03 F'
C
Temp (deg C)
110
100
90
80
Phase I Commissioning 70
Phase II Continuous steam injection 60
Phase III Engineering evaluations C' 50
Phase IV Low-level steam injection F 40
30
Phase V Sporadic low-level steam injection 20
Phase VI Liquid extraction (not shown) 10
Figure 6.2-6. Temperature profiles and water levels for extraction wells E-4 and E-6.
Elevation (ft MLLW)
Elevation (ft MLLW)

Elevation (ft MLLW)
11 Dec 02 9 Dec 02
E-6 16 Dec 02 15 Apr 03 14 Apr 03
14 Dec 02
Elevation (ft MLLW)
Elevation (ft MLLW)
Elevation (ft MLLW)

E-4
11 Dec 02 16 Dec 02 15 Apr 02

Water Levels 9 Dec 02 14 Dec 02 14 Apr 02
Figure 6.2-7. Collector layer temperatures on December 10 and 12, 2002.
December 10, 2002
Cold temperature
trough
Surface Cap
Temp (deg C)
Vapor Collectors
110
100
90
80
70
60
December 12, 2002 50
40
30
20
Cold temperature 10
trough
Surface Cap
Vapor Collectors
Figure 6.2-8. SeaTac airport precipitation and collector layer
temperature at instrument string T-52.
December 2002
1.4 100
90
1.2
80
Precipitation (inches)
Temperature (deg C)
1 70
0.8 60
SeaTac
50
T-52
0.6 40
0.4 30
20
0.2
10
0 0
1-Dec-02 8-Dec-02 15-Dec-02 22-Dec-02 29-Dec-02
March 2003
1.4 100
90
1.2
80
Precipitation (inches)
Temperature (deg C)
1 70
0.8 60
SeaTac
50
T-52
0.6 40
0.4 30
20
0.2
10
0 0
1-Mar-03 8-Mar-03 15-Mar-03 22-Mar-03 29-Mar-03
Figure 6.2-9. Temperature sensor locations in the vapor collector layer
exhibiting a temperature drop of greater than 20 degrees C.
December 10-12, 2002
E-2
E-1
E-6
E-4
E-7 E-3
E-5
Temperature Sensors
Extraction well
Sensor with less than 20 deg C change
Sensor with greater than 20 deg C change
March 12-13, 2003
E-2
E-1
E-6
E-4
E-7 E-3
E-5
Temperature Sensors
Extraction well
Sensor with less than 20 deg C change
Sensor with greater than 20 deg C change
Elevation (ft MLLW) Elevation (ft MLLW)
Figure 6.2-10. Instrument string temperature profiles along the inside of the sheet-pile wall for October-December 2002.
Elevation (ft MLLW)

Elevation (ft MLLW)
E-6
E-5
E-2
E-4
E-1
Elevation (ft MLLW)
E-3
E-7
Elevation (ft MLLW)
Elevation (ft MLLW)
Elevation (ft MLLW)
14-Dec-02
22-Oct-02
12-Nov-02
19-Nov-02
3-Dec-02
2-Oct-02
6-Nov-02 Elevation (ft MLLW) Elevation (ft MLLW)
-20
Figure 6.2-11. Instrument string temperature profiles along the inside of the sheet-pile wall for January-May 2003.
Elevation (ft MLLW)

Elevation (ft MLLW)
Elevation (ft MLLW)
E-6
E-5
Elevation (ft MLLW)

E-2
E-4
E-1
E-3
E-7
Elevation (ft MLLW)
Elevation (ft MLLW)
Elevation (ft MLLW)
Elevation (ft MLLW)
15-May-03
26-Feb-03
19-Mar-03
13-Jan-03
5-Feb-03
9-Apr-03
-15 Elevation (ft MLLW) Elevation (ft MLLW)
Figure 6.2-12. Instrument string temperature profiles for the E-4 array for October-December 2002.
Elevation (ft MLLW)
Elevation (ft MLLW)

Elevation (ft MLLW)
Elevation (ft MLLW)
Elevation (ft MLLW)

Elevation (ft MLLW)
T-30
120
2-Oct-02
Temperature (deg C)
100
22-Oct-02
Elevation (ft MLLW)
E-2
E-1 80 6-Nov-02
E-6
60
E-4 12-Nov-02
E-7 E-3
40 19-Nov-02
E-5
3-Dec-02
20
14-Dec-02
0
20 10 0 -10 -20
Elevation (ft MLLW)
Figure 6.2-13. Instrument string temperature profiles for the E-4 array for January-May 2003.
Elevation (ft MLLW)
Elevation (ft MLLW)

Elevation (ft MLLW)
Elevation (ft MLLW)
Elevation (ft MLLW)

Elevation (ft MLLW)
Elevation (ft MLLW)
T-53
20 40 60 80 100 120
Temperature (deg C)
13-Jan-03
5-Feb-03
E-2
26-Feb-03
E-1
19-Mar-03
E-6
E-4 9-Apr-03
E-7 E-3
15-May-03
E-5
0
20
15
10
-10
-15
5
-5
Elevation (ft MLLW)

Figure 6.3-1. Calculated average temperatures based on subsurface temperature
sensors
Measured average temperature Top sensor Bottom sensor Sensors inside and adjacent to sheet-pile
80
70
60
Average temperature (C)
50
40
All sensors (inside and
outside sheet-pile wall)
30
20
10
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.3-2. Comparison of the measured average pilot test area temperature to
the average temperature calculated from energy balance data and an assumed
volume and heat capacity
Net energy added to site Measured average temperature Average temperature (calculated w/o loss)
5,000 250

4,500 225
Net energy added to site (million BTU)
4,000 200
Calculated average temperature (F)

3,500 175
3,000 150
2,500 125
2,000 100
1,500 75
1,000 50
500 25
0 0
10/2 10/12 10/22 11/1 11/11 11/21 12/1 12/11 12/21 12/31 1/10 1/20 1/30 2/9 2/19 3/1 3/11 3/21 3/31 4/10 4/20 4/30 5/10 5/20
Figure 6.3-3. Cumulative volume of water (in thousand gallons) extracted from each E well during steam
pilot
500 750
250
100 777
53
816
727
19 172 750
621
100
250
500
Figure 6.3-4. Average steam injection rate (in lbs/hr) for each of the 16 injection wells in the
period between 10/30/02 and 12/14/02
200
200
300
193
233
275
224
244 311
200
327
160 368 366

143 266
199 193
378 302
200 300
Figure 6.4-1. Extracted Liquid PAH Concentrations
(November 7, 2002 through January 27, 2003)
E01 E02 E03 E04 E05 E06 E07
1400000
1200000
1000000
Concentration in ug/L
800000
600000
400000
200000
0
2
2
2
03
03
3
/0
/0
/0
/0
4/
1/
8/
2/
9/
6/
2/
9/
/7
/5
16
23
/1
/2
/2
/1
/1
/2
1/
1/
11
12
1/
1/
11
11
11
12
12
12
Date
Figure 6.4-2. Extracted Liquid PAH Concentrations
(January 28, 2003 through May 12, 2003)
E01 E02 E03 E04 E05 E06 E07

1200000
1000000
800000
600000
400000
200000
0
03
03
03
03
03
3
3
/0
/0
/0
/0
/0
/0
/0
/0
/0
/0
4/
4/
1/
8/
6/
28
11
18
25
11
18
25
15
22
29
2/
3/
4/
4/
5/
1/
2/
2/
2/
3/
3/
3/
4/
4/
4/
Date
Figure 6.4-3. Extracted Liquid PCP Concentrations
(November 7, 2002 through January 27, 2003)
E01 E02 E03 E04 E05 E06 E07
2000
1800
1600
1400
1200
1000
800
600
400
200
0
11/7/02 11/14/02 11/21/02 11/28/02 12/5/02 12/12/02 12/19/02 12/26/02 1/2/03 1/9/03 1/16/03 1/23/03
Date
1/
28
0
100
200
300
400
500
600
700
800
/0
1/ 3
30
/0
3
2/
1/
03
2/
3/
03
2/
5/
03
2/
7/
03
2/
9/
0
2/ 3
11
/0
2/ 3
13
/0
2/ 3
15
/0
2/ 3
E01
17
/0
2/ 3
19
/0
2/ 3
E02
21
/0
2/ 3
23
/0
E03
2/ 3
25
/0
2/ 3
27
/0
3
E04
3/
Date
1/
03
3/
3/
E05
03
3/
5/
03
3/
7/
E06
03
3/
9/
0
3/ 3
11
E07
/0
(January 28, 2003 through May 12, 2003)
3/ 3
13
/0
3/ 3
15
Figure 6.4-4. Extracted Liquid PCP Concentrations
/0
3/ 3
17
/0
3/ 3
19
/0
3/ 3
21
/0
3/ 3
23
/0
3/ 3
25
/0
3/ 3
27
/0
3
Figure 6.4-5. Average daily TOC and cumulative TOC (estimated where no data). Also shown are maximum average daily TOC, average
TOC removal rate per day, and average subsurface temperature and average extraction rate for each operational phase.
2,500 12,000
Phase II Phase III Phase IV Phase V
Avg. TOC 757 mg/L No data 860 mg/L 566 mg/L
Max. TOC 1962 mg/L No data 1401 mg/L 731 mg/L
Cum. TOC 2557 kg 166 kg (est.) 6261 kg 849 kg
Avg. TOC/day 67 kg/day 6 kg/day (est.) 91 kg/day 45 kg/day
Avg. site temp. 46.6 deg C 62.1 deg C 72.8 deg C 72.3 deg C 10,000
2,000 Avg. extr. rate 16 gpm 1.6 gpm 20 gpm 14 gpm
8,000
1,500
TOC (mg/L)
TOC (kg)
Average
6,000
Cumulative
1,000
4,000
500
2,000
0 0
7-Nov-02 5-Dec-02 2-Jan-03 30-Jan-03 27-Feb-03 27-Mar-03
Figure 6.6-1. Total PAH and PCP Concentrations in Treatment Plant Influent
Prior to and at Start of Pilot Study
Total PAH PCP

10,000,000
Pilot Start Date: 10/1/02

1,000,000
100,000
10,000
1,000
100
10
1
0 10 20 30 40 50 60
Week of 2002
Figure 6.6-2. Total PAH Concentrations Into (SP-0) and Out of (SP-4) DAFT
SP-0 SP-4
200,000
2,630,000 ug/L on 12/2/02

180,000
160,000
Total PAH Concentrations in ug/L
140,000
120,000
100,000
80,000
60,000
40,000
20,000
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-3. Naphthalene Concentrations Into (SP-0) and Out of (SP-4) DAFT
SP-0 SP-4
200,000
(1,150,000 ug/L on 12/2/02)
180,000
160,000
Naphthalene Concentration in ug/L
140,000
120,000
100,000
80,000
60,000
40,000
20,000
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-4. PCP Concentrations Into (SP-0) and Out of (SP-4) DAFT
SP-0 SP-4
1,400
1,200
1,000
PCP Concentration in ug/L
800
600
400
200
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-5. NAPL Removal per Month and Cumulative Total NAPL Removal
Cumulative T-108 Only Monthly T-108 Only

T-108 Cumulative Trend Cumulative PA
25,000
20,000
15,000
Gallons
10,000
5,000
0
99
99
99
00
00
00
01
01
01
02
02
02
03
8
2
99
99
00
00
00
19
19
19
20
20
20
20
20
20
20
20
20
20
/1
/1
/2
/2
/2
1/
1/
1/
1/
1/
1/
1/
1/
1/
1/
1/
1/
1/
/1
/1
/1
/1
/1
3/
6/
9/
3/
6/
9/
3/
6/
9/
3/
6/
9/
3/
12
12
12
12
12
Month
Figure 6.6-6. Total PAH Concentrations Into (SP-4) and Out of (SP-6) Aeration Basin
SP-4 SP-6
100,000
90,000
80,000
Total PAH Concentration in ug/L
70,000
60,000
50,000
40,000
30,000
20,000
10,000
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-7. Naphthalene Concentrations Into (SP-4) and Out of (SP-6) Aeration Basin
SP-4 SP-6
100,000
93,700 ug/L on 12/15/02
90,000
80,000
Naphthalene Concentration in ug/L
70,000
60,000
50,000
40,000
30,000
20,000
10,000
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-8. PCP Concentrations Into (SP-4) and Out of (SP-6) Aeration Basin
SP-4 SP-6
1,400
1,200
1,000
PCP Concentrations in ug/L
800
600
400
200
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-9. Aeration Basin Total PAH and PCP Removal Effectiveness
PAH Removal (%) PCP Removal (%)
120
100
80
Percent Removed
60
40
20
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-10. Dissolved Oxygen and Total PAH Concentration In Aeration Basin
SP-4 Dissolved Oxygen (mg/L) SP-4 Total PAH (ug/L)
8.00 100,000
90,000
7.00
Dissolved Oxygen Concentration in mg/L
80,000
6.00

70,000
5.00
60,000
4.00 50,000
40,000
3.00
30,000
2.00
20,000
1.00
10,000
0.00 0
9/1/02 9/21/02 10/11/02 10/31/02 11/20/02 12/10/02 12/30/02 1/19/03 2/8/03
Date
Figure 6.6-11. Total PAH Concentration Into (SP-8) and Out of (SP-10) Carbon
Treatment System
SP-8 SP-10
200
629 ug/L on 12/9/02
180
160
140
120
100
80
60
40
20
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-12. PCP Concentration Into (SP-8) and Out of (SP-10) Carbon Treatment
System
SP-8 SP-10
200
180
160
140
PCP Concentration in ug/L
120
100
80
60
40
20
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-13. Carbon Treatment System Total PAH and PCP Removal Effectiveness
PAH Removal (%) PCP Removal (%)
102.00
100.00
98.00
Percent Removed
96.00
94.00
92.00
90% on 12/16/02
90.00
88.00
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-14. Total PAH and PCP Concentrations in GWTP Effluent
Total PAH (ug/L) PCP (ug/L)
25
Total PAH Compliance Concentration = 20 ug/L

20
15
10
PCP Compliance Concentration = 6 ug/L
0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-15. pH Concentration in GWTP Effluent
SP-11
10
9.5
Discharge Limit < 9
9
8.5
8
pH
7.5
6.5
Discharge Limit > 6

6
5.5
5
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Figure 6.6-16. Temperature and Dissolved Oxygen in GWTP Effluent
Temperature Dissolved Oxygen

30 18
Temperature Discharge Limit < 25 deg. C

16
25
14
Dissolved Oxygen in mg/L

20 12
Temperature in Deg. C
10
15
8
Dissolved Oxygen Discharge Limit > 6 mg/L
10 6
4
5
2
0 0
9/1/02 10/21/02 12/10/02 1/29/03 3/20/03 5/9/03 6/28/03 8/17/03
Date
Appendices
Appendix A Site Cleanup Levels
Appendix B Pilot Test Area Selection Memorandum
DEPARTMENT OF THE ARMY
SEATTLE DISTRICT, CORPS OF ENGINEERS
P.O. BOX 3755
SEATTLE, WASHINGTON 98124-2255
REPLY TO
ATTENTION OF
CENWS-PM
MEMORANDUM FOR RECORD
RE: Wyckoff/Eagle Harbor Superfund Site, Soils and Groundwater

Operable Units, Addendum to Thermal Remediation Pilot Test Area
Location Selection Memorandum dated April 21, 2000
DATE: August 9, 2000
1. This memorandum provides an addendum to the Thermal Remediation Pilot Test

Area Location Selection Memorandum, describing the recent changes made to the
pilot study location.
2. The changes were made in consultation between EPA Region 10 and the U.S. Army
Corps of Engineers (USACE), as well as USACEs expert consultants for the pilot
study design: Roger Aines and Robin Newmark of Lawrence Livermore National
Laboratory, Gorm Heron of SteamTech, Inc., and Kent Udell of the University of
California Berkeley.
3. The following changes were made to the pilot location:
2.2 Design Data Total NAPL Removal and Aquitard Heating are limited to
the upper aquifer. NAPL contained in the aquitard will not be remediated as part
of the pilot study, based upon discussions between Hanh Gold of EPA Region 10
and Kathy LeProwse of USACE in January 2000. Aquitard heating will be
limited to heating at and near the upper surface of the aquitard, to allow NAPL
located at the bottom of the aquifer (e.g., sitting on top of the aquitard) to be
remediated.
5.1 Selection Criteria Criteria 7 (Presence of sufficient contaminated fine-

grained materials (aquitard or non-marine clays) for testing of electrical resistance
heating) and 8 (Presence of NAPL in the aquitard, to allow testing of thermal
methods for remediation of the aquitard) are not critical to the scope of the pilot
study. Use of electrical resistance heating would only be used if necessary to
meet the three pilot study objectives. EPA does not require evaluation of this
technology during the pilot study. See the Pilot Study Final Conceptual (10%)
Design, Section 5.6, for more details regarding the use of this technology.
5.2 Comparison of Alternative Test Areas See the change to 5.1 relative to
selection criteria 7 and 8.
1 of 2
5.3 Preferred Alternative Test Area The proposed pilot study location has been
modified to reduce the total volume of soil to be treated from approximately
40,000 yd3 to approximately 25,000 yd3. The volume reduction was made by
moving the western portion of the pilot study sheet pile wall eastward from the
alignment proposed in the Pilot Study Location Selection Memorandum. The
primary purpose for this size reduction was to allow successful completion of the
pilot study within the EPAs budget. Due to this change, the estimated NAPL
volumes decreased from 17,000 gallons of LNAPL and 41,000 gallons of DNAPL
to 13,000 gallons of LNAPL and 29,000 gallons of DNAPL. EPA has indicated
that the size of the pilot study will not be less than 25,000 yd3 and, if funds allow,
the size will be larger.
Figures 1, 2, and 3 Area C does not extend as far west as shown in this figure,
based on the changes described for 5.3 above. Please refer to Pilot Study
Conceptual (10%) Design, Figure 4, to see the revised alignment for Area C.
4. This memorandum is submitted in lieu to revising the Thermal Remediation Pilot

Test Area Location Selection Memorandum.
5. Any questions related to this addendum or the Thermal Remediation Pilot Test Area
Location Selection Memorandum should be directed to Ms. Kathy LeProwse of
USACE (206-764-3505).
END OF RECORD
2 of 2
Thermal Remediation
Pilot Test Area Selection
Memorandum
Groundwater and Soil Operable Units

Bainbridge Island, Washington
FINAL
Prepared by
Department of the Army
U. S. Army Corps of Engineers

Seattle District
21 April 2000
FINAL
TABLE OF CONTENTS
TABLE OF CONTENTS ............................................................................................................ II

LIST OF TABLES ......................................................................................................................III
LIST OF FIGURES ....................................................................................................................III
LIST OF ACRONYMS ..............................................................................................................III
1. INTRODUCTION..............................................................................................................1
1.1 Purpose............................................................................................................................1
1.2 Background ....................................................................................................................1
1.3 Site Data..........................................................................................................................1
2. PILOT TEST OBJECTIVES............................................................................................2
2.1 Performance ...................................................................................................................2
2.2 Design Data.....................................................................................................................2
3. SITE DESCRIPTION........................................................................................................3
3.1 Fill....................................................................................................................................3
3.2 Marine Sand and Gravel. ..............................................................................................3
3.3 Non-Marine Clay. ..........................................................................................................3
3.4 Marine Silt. .....................................................................................................................3
3.5 Glacial Clay, Silt, and Sand. .........................................................................................4
4. DESCRIPTIONS OF ALTERNATIVE TEST AREAS .................................................4
4.1 Area A .............................................................................................................................4
4.2 Area B .............................................................................................................................5
4.3 Area C .............................................................................................................................6
5. DEVELOPMENT OF PREFERRED ALTERNATIVE ................................................7
5.1 Selection Criteria ...........................................................................................................7
5.2 Comparison of Alternative Test Areas ........................................................................7
5.3 Preferred Alternative Test Area...................................................................................9
6. RECOMMENDATIONS...................................................................................................9
7. REFERENCES.................................................................................................................10
ii
FINAL
LIST OF TABLES
Table 1 Alternative Pilot Area Data

Table 2 Alternatives Compared by Selection Criteria
All tables are located at the end of the document text.
LIST OF FIGURES
Figure 1 Site Plan and Proposed Pilot Locations

Figure 2 Non-Marine Clay Thickness
Figure 3 LNAPL Thickness
Figure 4 DNAPL Thickness
Figure 5 Location of NAPL in the Aquitard and Solid-Phase PAH Deposits
All figures are located at the end of the document text.
LIST OF ACRONYMS
DNAPL dense non-aqueous phase liquid
EPA U. S. Environmental Protection Agency
FS Feasibility Study
ITTAP Insitu Thermal Technologies Advisory Panel
LNAPL light non-aqueous phase liquid
MLLW Mean Lower Low Water
NAD83 North American Datum
NAPL non-aqueous phase liquid
NGVD National Geodetic Vertical Datum
OU operable unit
ppm parts per million
ppt parts per thousand
RI Remedial Investigation
USGS United States Geological Survey
WSPCS Washington State Plane Coordinate System
iii
FINAL
Pilot Test Area Selection Memorandum
1. INTRODUCTION
1.1 Purpose
The purpose of this memorandum is to document the decision process involved in the selection
of the Thermal Remediation Pilot Test location at the Wyckoff Superfund Site. The
memorandum will also include descriptions of each test area, as well as a comparison of the test
areas based on the test objectives, and a proposal for a preferred pilot area.
1.2 Background
Due to the site-specific nature of thermal processes, EPA has determined that a pilot test must be
completed prior to implementing a full scale thermal remediation. The pilot test area will be
separated from the rest of the site by a sheetpile wall, to prevent recontamination by inflow of
groundwater and NAPL from adjacent untreated soil, and to allow an objective evaluation of the
results of the thermal treatment. Three possible locations for a pilot test were discussed at the
ITTAP meeting on 19 October 1999:
A. Upgradient (south) end of site, east shoreline
B. Upgradient (south) end of site, west shoreline
C. Downgradient (north) end of site, east shoreline
Areas A and B were considered to be the most representative of the project site, since the sheet
pile enclosure can be open at the upgradient end to allow recharge by oxygenated groundwater.
Area C was eventually eliminated from further consideration due to its downgradient location,
and subsequent evaluations have focussed on upgradient areas. Three upgradient alternatives
will be compared in this memorandum, including the two original upgradient areas (See Figure
1):
A. East shoreline
B. West shoreline
C. Central Area
All three areas have the potential to be configured to the approximate shape of the full site, and
the surrounding sheetpile walls can be terminated upgradient, in relatively uncontaminated soils,
to allow lateral groundwater recharge into the treatment zone.
1.3 Site Data

The evaluations described in this memorandum are based on all available site data, including
information derived from the following investigations:
Investigations performed from 1972 through preparation of the Final RI Report (CH2M
HILL, June 1997). A complete list can be found in the RI Report.
Pre-design drilling and sampling for a proposed slurry wall, performed in 1997 and 1998 (U.
S. Army Corps of Engineers, 1998b and 1998c). Although these were primarily geotechnical
and geological explorations, considerable data on NAPL extent was obtained.
Pre-design drilling, probing and sampling in 1999, for the currently proposed sheet pile wall
and thermal treatment (U. S. Army Corps of Engineers, 2000a).
1
FINAL
Additional push-probe explorations in 2000, delineating the upgradient extent of the
proposed pilot test locations, and filling data gaps revealed during conceptual thermal design
(U. S. Army Corps of Engineers, 2000b).
2. PILOT TEST OBJECTIVES

The objectives of the pilot test fall into two broad categories: a) To assess the likelihood that a
full-scale thermal remediation will achieve the cleanup goals for the site; b) To provide
information for implementation of the potential full scale thermal remediation. Pilot area
selection criteria are based on these objectives.
2.1 Performance
EPA has developed performance expectations for the pilot test, which will demonstrate the
ability of the thermal technologies to meet the cleanup goals:
1. All mobile NAPL is removed from the pilot test area.
2. NAPL component concentrations are reduced sufficiently to allow site groundwater to be
fully remediated by continuing thermally-enhanced natural processes, after the active
treatment is completed. EPA will use site-specific laboratory testing and modeling to
predict the degradation rates which will occur after treatment, and the conditions required for
the residual contamination to be reduced to concentrations that are protective of marine water
quality, surface water quality, and sediment standards in Puget Sound.
3. Contaminant levels in surface soils (0-15 depth) are reduced to MTCA Method B cleanup
levels.
If expectation 3. is not fully achieved, other measures such as soil capping may be implemented.
Failure to meet expectations 1. and 2., however, could result in institution of the containment and
pump-and-treat alternative instead of full-scale thermal remediation.
2.2 Design Data

The pilot test will also provide vital information which can be used for planning and design of
the proposed full-scale thermal treatment:
Community impacts
Potential adverse effects to Eagle Harbor
Vapor cap performance
Dioxin removal from site soil
Total NAPL removal
Steam migration patterns
Aquitard heating
Treatment plant performance
Microbial populations and contaminant oxidation rates before and after treatment
2
FINAL
Operational approaches
Engineering data such as well diameter, depth, and spacing; monitoring requirements; vapor
cap thickness; extraction and injection rates; surface temperatures; sheetpile permeability;
construction material requirements; etc..
3. SITE DESCRIPTION
The Wyckoff property occupies approximately 57 acres, including a spit with about 0.8 miles of
shoreline extending northward into Eagle Harbor. The spit has been extended and filled at least
twice prior to the 1950s, and was the location of wood treatment activities that have caused the
current soil and groundwater contamination. The spit covers approximately 8 acres, and
contains the majority of the on-site contamination.
The Wyckoff Soil and Groundwater Operable Units (OUs) occupy a relatively flat lowland (the
spit and the log peeler area) and intertidal area, bounded by a densely vegetated bluff on the
south. The lowland area has an average elevation of approximately 15 feet MLLW, while the
hillside area rises to elevations above 200-feet. The north and west portions of the spit are
bounded by Eagle Harbor, and by Puget Sound along the eastern margin. A summary of the
geologic materials present in the onshore portion of the Soil and Groundwater OUs are provided
below.
3.1 Fill.
Fill material, imported from nearby sources, was placed on the property to extend the shoreline
into Eagle Harbor and Puget Sound. The fill consists of silt and fine-grained sand similar in
physical characteristics to the underlying marine sand and gravel unit, making it difficult to
differentiate between the two units. Locally, the fill consists of fine brown sand containing
bricks, broken glass, metal fragments, and other anthropogenic material, and generally lacks
shell fragments.
3.2 Marine Sand and Gravel.

The marine sand and gravel unit is a nearshore marine/beach deposit present over nearly all of
the Wyckoff facility. The unit underlies the fill or non-marine clay unit and is generally
continuous to the top of the marine silt or the glacial deposits. The marine sand and gravel unit
consists of loose to dense, poorly graded, gray to dark gray, fine to medium sand with shell
fragments throughout. Gravel zones are common in this unit, and cobble zones have been
observed along the east side of the site.
3.3 Non-Marine Clay.

The non-marine clay is either uplands colluvium or imported fill, generally lying
stratigraphically above the marine sand and gravel. The non-marine clay consists of gray to
brown, very soft to medium clay to brown clayey fine sand. Occasional plant fibers, wood
fragments, and roots are present, as well as iron oxide staining.
3.4 Marine Silt.

The marine silt layer is a nearshore lagoon, tide-flat, or marsh deposit which occurs below the
marine sand and gravel in the northern and eastern portions of the facility. The unit generally
consists of a distinctive olive-colored silt or clay or silty sand with abundant shells and shell
3
FINAL
fragments and occasional red-brown wood debris and other organic material. The marine silt
often overlies the glacial clay, silt, and sand unit, but may also be separated from the glacial
deposits by thin layers of marine sand.
3.5 Glacial Clay, Silt, and Sand.

The glacial clay, silt and sand unit lies stratigraphically beneath the marine silt and the marine
sand and gravel deposits. The glacial deposits consist mainly of gray to brown, stiff to hard clay
and silt, with some sand and gravel, and no organic matter. Sand intervals within the glacial
aquitard have been observed in the central portion of the site. The glacial portion of the aquitard
has been divided into three facies: brown-gray silty sand, brown-gray sandy, gravelly silt and
gray clayey silt. The gray facies is only found in the southwest corner of the site, and appears to
transition to the brown-gray facies in the central portion of the site. The glacial unit acts as an
aquitard between the upper and lower aquifers.
4. DESCRIPTIONS OF ALTERNATIVE TEST AREAS

Locations of the proposed test areas are shown on Figure 1. The horizontal extent of each of the
test areas is based on a total treatment volume of 40,000 to 50,000 cubic yards.
4.1 Area A
Pilot Test Area A is located on the east side of the project area, directly north of the treatment
plant.
4.1.1 Geology
The upper aquifer at Area A consists of fill material and marine sand and gravel totaling 26 to 53
feet thick. The fill material is 5 to 10 feet thick over most of Area A and thins slightly to the
south. Marine sand and gravel underlies the fill material and is approximately 16-feet to 43-feet
thick. Marine silt is present in the northern half of Area A, lying directly below the marine sand
and gravel unit, and directly above the glacial clay, silt and sand unit. The maximum thickness
of the marine silt is 6 feet (99CD05), thining southward and pinching out beneath the treatment
plant. The glacial aquitard is composed of the brown, silty sand facies.
Two separate cobble zones up to six-feet thick have been identified in the marine sand and
gravel within Area A. In addition, cobbles have been observed at the glacial contact with the
marine silt (99CD05, RPW7 and 97AP22) and within the glacial aquitard (99CD05, RPW7 and
CW02).
4.1.2 Contamination
Mobile NAPL is primarily restricted to the fill and marine sand and gravel units at Area A.
Some small blobs of a waxy form of contaminant were noted associated with a heavy sheen
within the marine silt unit in contact with the glacial aquitard at 99CD05. The maximum
thicknesses of LNAPL and DNAPL in Area A are approximately 9 feet and 18 feet, respectively,
at boring 97AP01.
The existence of NAPL beneath the treatment plant is a primary concern in Area A. Ten feet of
DNAPL was observed at CW03 approximately 20 feet north of the treatment plant pad. Boring
00PP01, just north of the treatment pad, showed a total of 5.5-feet of mobile NAPL at various
4
FINAL
intervals. Boring 00PP05, drilled through the treatment plant pad, revealing about 1 foot of
residual NAPL. Boring 00PP04, approximately midway along the seaward side of the plant,
encountered 0.5-foot of mobile LNAPL as well as significant residual NAPL.
No contaminant has been observed within the aquitard at Area A. The results from 00PP16 and
99SE25 indicate that the area upgradient of Area A is free of contamination.
4.2 Area B
4.2.1 Geology
The upper aquifer at Area B consists of fill material, non-marine clay, and marine sand and
gravel. These units are underlain by the glacial clay, silt and sand unit. Upper aquifer thickness
ranges from 16 to 36 feet. Fill material averages about 10 feet thick, reaching a maximum
thickness of 22 feet. Most of the fill in Area B consists of a sequence of a few feet of basalt
cobbles, overlying silty sand which has been placed over non-marine clay fill. The maximum
thickness of the non-marine clay fill is approximately 17 feet, thinning northward and pinching
out at the north end of the Area (see Figure 2).
Marine sand and gravel underlies the fill material and varies from 3 to 26 feet thick. Several thin
gravel deposits exist at Area B but no cobble zones have been observed. Organic-rich materials
underlie the clay fill in the north and east portions of the Area (3 feet in boring 97AP18); these
have been interpreted as lagoon deposits, and are mapped as either marine sand and gravel or
surficial marine sediment. Both the marine sand and gravel and the lagoon deposits directly
overly the glacial clay, silt and sand unit.
The glacial aquitard is composed of both the brown, sandy, gravelly silt and gray, clayey silt
facies within Area B. No cobbles have been observed in association with the glacial aquitard in
this area of the Wyckoff site.
4.2.2 Contamination
The maximum thickness of LNAPL and DNAPL in Area B is approximately 1.5 and 14.5 feet,
respectively (borings 99SE47 and 99SE40). The extent of LNAPL is limited, and has only been
detected at borings 97AP41, 99SE42, and 99SE47. Mobile NAPL is present in all units at Area
B, including the glacial aquitard. In general, the thickest NAPL-contaminated intervals occur in
the marine sand and gravel unit; however boring 97AP29, in the extreme northern portion of the
Area, revealed a 6-foot NAPL-saturated zone in the lagoon bottom sediments. The northeastern
edge of the Area may contain solid-phase PAH deposits associated with lagoon sediments
(borings 00PP13, 00PP14, 00PP30, see Figure 5).
Up to 3 feet of NAPL has been observed in both the marine silt and the non-marine clay unit.
Approximately 6 feet of DNAPL is in contact with the aquitard in the vicinity of boring 97AP29,
and LIF data indicate that NAPL is present within the aquitard at 99SE40, 99SE46 and 99SE47.
The maximum contaminant thickness in the glacial aquitard may be as much as 5 feet,
penetrating as deep as 11 feet below the top of the aquitard.
Results from borings 00PP24, 99SE21 and 99SE22 indicate that the area upgradient of Area B is
5
FINAL
free of NAPL; however, a pilot soil boring 24 feet north of MW19 noted a naphthalene odor and
petroleum sheen at approximately 20 feet below ground surface. A sample collected from boring
00PP24 contained 160 mg/kg diesel range hydrocarbons at the interface between the upper
aquifer and the underlying glacial material. The analysis of PAHs in this sample revealed the
presence of phenanthrene (33 mg/kg) and fluoranthene (16 mg/kg). Other PAHs including
naphthalene, acenaphthene, fluorene, anthracene, pyrene, benzo(a)anthracene, chrysene and
benzo(k)fluoranthene were each detected at concentrations less than 10 mg/kg. In addition,
small pieces of asphalt-like material were visible in soils collected from 4 to 8 feet in depth.
4.3 Area C
4.3.1 Geology
Surface soils at Area C is consists of fill material and non-marine clay. The fill material consists
primarily of silty sand with occasional basalt cobbles, averaging about 7 feet thick and reaching
a maximum thickness of about 14 feet in the northern portion of the Area. In the southwest half
of Area B, the silty sand overlies as much as 10 feet of non-marine clay; the clay thins northward
and pinches out on the northeast margin.
The upper aquifer in Area C consists of marine sand and gravel, underlying the fill and
nonmarine clay, with thicknesses from of 2 to 33 feet. Several thin, fine gravel deposits exist in
the Area, but no cobble zones have been observed. Lagoon deposits appear to be absent from
Area C, although organic debris was noted in the west side, indicating the probable eastern limit
of the former lagoon. The marine sand and gravel directly overlies the glacial clay, silt and sand
unit at Area C. The glacial aquitard is composed of both the brown, sandy, gravelly silt facies
and the gray, clayey silt facies. No cobbles have been observed in association with the glacial
aquitard in Area C.
4.3.2 Contamination
The maximum thickness of mobile LNAPL and DNAPL in Area C is approximately 13 and 9
feet respectively; both maximums were encountered at boring 00PP21. The thickest NAPL
zones occur at the north end of the Area, thinning to the south as the aquitard becomes
shallower. Mobile NAPL is present in all units, including the glacial aquitard. Most of the
NAPL occurs in the marine sand and gravel unit; NAPL in the non-marine clay is primarily in
the lower 1 foot of the unit, at the contact with the marine sand and gravel (borings 00PP22 and
97AP28). In the vicinity of boring 00PP21, approximately 6 feet of DNAPL lies directly on the
aquitard. At the southwest corner of the Area, LIF data from boring 99SE45 indicate the
existence of 5 feet of NAPL located 11-feet below the top of the aquitard.
Twelve direct-push borings have been completed upgradient of Area C to delineate the southern
extent of contamination. With the exception of boring 00PP09, each of the pushes contained
some level of residual contamination, including asphalt or charcoal debris. Analytical results
indicate two broad categories of residual contamination upgradient of Area C:
The first category is characterized by a higher concentration of motor-oil-range hydrocarbons

relative to diesel-range hydrocarbons, along with a greater fraction of HPAH relative to
LPAH. LPAH concentrations were below the laboratorys reporting limit in many samples.
6
FINAL
This pattern of contamination occurred in samples from pushes 00PP08, 00PP10, 00PP18,
00PP27 and 00PP29; with the highest concentrations occurring at 4 to 8 feet depth in push
00PP10. Hydrocarbon contamination in this sample included 2200 mg/kg motor-oil-range
TPH and 20 mg/kg total HPAH; total LPAH was below the laboratorys reporting limit
(USACE, 2000b).
The second category of contamination more closely resembles chemical patterns observed
throughout the Wyckoff site, associated with creosote. This pattern is characterized by higher
concentrations of diesel-range hydrocarbons relative to motor-oil-range hydrocarbons, and
high concentrations of LPAH relative to HPAH. The LPAH component is comprised mostly
of naphthalene. These characteristics are present at pushes 00PP23, 00PP25 and 00PP26.
The highest concentrations occurred in boring 00PP25 at a depth of about 5 feet, at the
interface between the nonmarine clay and the glacial unit. The analysis of this sample
revealed 3000 mg/kg diesel-range hydrocarbons, 1400 mg/kg motor-oil-range hydrocarbons,
and 1156 mg/kg total LPAHs (USACE, 2000b). Contamination with these same
characteristics was also detected in sand and gravel zones within the glacial aquitard, in
pushes 00PP23 and 00PP26.
5. DEVELOPMENT OF PREFERRED ALTERNATIVE
5.1 Selection Criteria

The following criteria have been identified for selection of the best pilot test area, based on
performance and design data goals listed in Paragraph 2.
1. Access available to all required drilling locations for extraction and injection wells
2. Upgradient location to allow construction of an open-ended sheetpile enclosure, allowing
lateral groundwater recharge.
3. Absence of mobile NAPL upgradient of pilot area.
4. Presence of both LNAPL and DNAPL.
5. Soil characteristics and stratigraphy similar to overall site.
6. Presence of DNAPL directly overlying the aquitard, to allow evaluation of steam
effectiveness at full aquifer depth.
7. Presence of sufficient contaminated fine-grained materials (aquitard or non-marine clay).for
testing of electrical heating.
8. Presence of NAPL in the aquitard, to allow testing of thermal methods for remediation of the
aquitard.
9. Shoreline location, to allow monitoring of thermal impacts in the intertidal zone.
The above criteria are listed in order of perceived priority. Criteria 1 through 4 are considered to
be critical to the success of the pilot test or for generation of meaningful results, while Criteria 5
through 9 are desirable.
5.2 Comparison of Alternative Test Areas
7
FINAL
The following comparison of proposed pilot test areas is based on the four critical criteria and
five desirable criteria outlined in Paragraph 5.1. Results of the comparison are summarized in
Table 2.
5.2.1 Access to Drilling Locations

Access to about 1/3 of Area A is restricted by the presence of the treatment plant. Explorations
have shown significant thicknesses of NAPL under the plant, however it would not be feasible to
install wells or electrodes on the required spacing in this area. There are no drilling access
problems at Areas B and C. A large portion of Area B is covered by an asphalt pad, however
this would not prohibit well installation.
5.2.2 Upgradient Location

All of the proposed pilot test areas are located at the upgradient margin of the project area.
5.2.3 Absence of NAPL Upgradient of Test Area

LNAPL is present upgradient of the current Area A location, under the treatment plant, as shown
in Figure 3. The boundaries of Area A would have to be adjusted southward to include the
treatment plant area; however this would result in a large total treatment volume if the northern
boundary is maintained near boring 99CDO5 to include a small zone of contamination in contact
with the aquitard. No mobile NAPL exists upgradient of Areas B and C. Small amounts or
residual NAPL were discovered upgradient of Area C; this contamination could easily be
removed by excavation to prevent any interference with the pilot study.
5.2.4 Presence of LNAPL and DNAPL

Both LNAPL and DNAPL are present in all 3 proposed test areas (Figures 3 and 4). At least 5
feet of both LNAPL and DNAPL are present over a large portion of Areas A and C. Area B
contains mostly DNAPL, however, with relatively little LNAPL.
5.2.5 Soils and Stratigraphy Representative of the Total Project

Stratigraphy at Area A is representative of the majority of the Wyckoff site, with all of the
onshore geologic units represented except the non-marine clay. Area B is dominated by the non-
marine clay unit, which occurs only in the southwest corner of the project area. In addition, Area
B contains organic-rich lagoon sediments, which are restricted to the western margin of the
project area. Area C contains all of the geologic units except the marine silt; relative volumes of
the geologic units are representative of the overall project area.
5.2.6 DNAPL directly overlying the Aquitard

No significant DNAPL is in contact with the glacial aquitard at Area A. Large portions of Areas
B and C contain several feet of DNAPL directly overlying the aquitard.
5.2.7 Contaminant in Fine-Grained Material

Neither the marine silt nor the glacial unit at Area A appear to contain significant NAPL
contamination. Areas B and C contain NAPL contamination in both the non-marine clay and in
the glacial unit.
5.2.8 Presence of NAPL in the Aquitard

No NAPL has been observed within the aquitard at Area A. Areas B and C contain NAPL in the
aquitard; in both instances the NAPL occurs as deep as 11 feet below the top of the aquitard.
8
FINAL
5.2.9 Shoreline Location
The east margin of Area A consists of approximately 150-feet of intertidal shoreline. Area B is
bounded on the west by about 250-feet of bulkheads, with subtidal sediment surfaces. Area C is
at least 250 feet from the nearest shoreline.
5.3 Preferred Alternative Test Area

Area C appears to be best choice for a pilot test location. Area A is ruled out by restricted access
to drilling locations, and also lacks NAPL on the aquitard or in fine-grained soils. Area B
contains disproportionate volumes of non-marine clay and lagoon-bottom sediments, and is
LNAPL-deficient.
No structures which would restrict drilling activity are present in Area C. Approximately 17,000
gallons of LNAPL and 41,000 gallons DNAPL are present, with no mobile NAPL detected
upgradient of the Area. The stratigraphy is similar to the overall project area, including about
7% (by volume) non-marine clay in the southwest corner (see Figure 2). DNAPL is in contact
with the glacial aquitard over a large area, primarily in the north end of the Area. Area C also
provides an opportunity to test electrical heating technology, with NAPL present in both the
glacial aquitard and in the non-marine clay.
6. RECOMMENDATIONS
It is recommended that Area C be accepted as the Pilot Test Area for thermal remediation at
the Wyckoff Superfund Site.
Although Area C exhibits all of the critical characteristics for pilot testing, it does not have
an intertidal shoreline which would allow evaluation of thermal impacts to the marine
environment. Thermal monitoring during the pilot study should be used to refine the
thermal parameters employed for pilot test design, and these improved parameters should
then utilized to make better predictions of heat transfer along the shoreline.
A limitation common to all upgradient test areas is the shallow depth of testing, compared to
average aquifer depths over the project site. Well installation costs, injection pressures,
steam temperatures, and extraction rates will all increase as deeper portions of the aquifer are
treated. These differences need to be considered when extrapolating pilot test results to the
full-scale remediation.
Residual hydrocarbons detected upgradient of Area C may be a source of contaminated
groundwater which could affect the interpretation of the pilot test results. This issue should
be evaluated, and the contaminated soils should be removed as necessary.
Although sufficient explorations exist to allow selection of a test area, well screen and
electrode locations can only be estimated at this time. The exact locations should be
determined in the field, utilizing data obtained during drilling for the wells, electrodes, and
thermal probes.
9
FINAL
7. REFERENCES
Berkeley Hydrotechnique, 1989, Groundwater Extraction System for the Wyckoff /Eagle Harbor
Site, Bainbridge Island, Washington: Report to EPA Region 10.
CH2M HILL, 1999a, Draft groundwater flow and transport modeling report, Wyckoff/Eagle
Harbor Superfund Site, Bainbridge Island, Washington: Report to EPA Region 10.
CH2M HILL, 1999b, Draft parameter estimation and sensitivity analysis of the Wyckoff
groundwater flow model, Wyckoff/Eagle Harbor Superfund Site, Bainbridge Island,
Washington: Report to EPA Region 10.
CH2M HILL, 1997, Final Remedial Investigation (RI) report for the Wyckoff Soil and
Groundwater Operable Unit, Wyckoff/Eagle Harbor Superfund Site, Bainbridge Island,
Washington: Report to EPA Region 10.
CH2M HILL, 1995, Extraction Well Construction Report, Wyckoff Facility and Groundwater
Operable Units, Wyckoff/Eagle Harbor Superfund Site, Kitsap County, Washington: Report to
EPA Region 10.
Freeze, A. F, Cherry, R. A., 1979, Groundwater: Englewood Cliffs NJ, Prentice-Hall, Inc.
Incropera, P. F., DeWitt, D. P., 1996, Introduction to heat transfer: New York NY, John Wiley
and Sons, Inc.
Western Regional Climate Center, 1999, Bremerton, Washington monthly average temperature:
website http://www.wrcc.dri.edu/summary/climsmwa.html
National Oceanographic and Atmospheric Administration, 1999a, National Oceanographic Data

Center coastal water temperature guide: website http://www.co-ops.nos.noaa.gov/bench.html
National Oceanographic and Atmospheric Administration, 1999b, National Ocean Service

Center for Operational Oceanographic Products and Services: website
http://www.nodc.noaa.gov/dsdt/wtg12.html
Sellmeijer, J. B. and Shouten, C. P., 1995, Hydraulic resistance of steel sheet pile joints: Journal
of Geotechnical Engineering, v. 121, no. 2.
Sellmeijer, J. B. and Shouten, C. P., 1993, Steel sheet pile seepage resistance: Proceedings of the
Fourth International Landfill Symposium, Cagliari, Italy.
Todd, D. K., 1980, Groundwater hydrology: New York NY, John Wiley and Sons, Inc.
10
FINAL
U. S. Army Corps of Engineers, 2000a, Wyckoff NAPL field exploration draft comprehensive
report, Soil and Groundwater Operable Units, Wyckoff/Eagle Harbor Superfund Site: Report to
EPA Region 10.
U. S. Army Corps of Engineers, 2000b, Supplemental NAPL field exploration report, Soil and
Groundwater Operable Units, Wyckoff/Eagle Harbor Superfund Site: Report to EPA Region 10.
U. S. Army Corps of Engineers, 1999a, Draft conceptual design, Soil and Groundwater Operable
Units, Wyckoff/Eagle Harbor Superfund Site: Report to EPA Region 10.
U. S. Army Corps of Engineers, 1999b, Wyckoff/Eagle Harbor Superfund Site Groundwater OU

Thermal Effects Study management plan: Report to EPA Region 10.
U. S. Army Corps of Engineers, 1999c, Wyckoff/Eagle Harbor Superfund Groundwater and Soil
Operable Units thermal modeling protocol: Report to EPA Region 10.
U. S. Army Corps of Engineers, 1999d, Wyckoff/Eagle Harbor Superfund East Harbor Operable
Unit Intertidal Area natural recovery study, LNAPL flux calculations: Report to EPA Region 10.
U. S. Army Corps of Engineers, 1998a, East Harbor Operable Unit Intertidal Area natural
recovery study, Phase 1: Report to EPA Region 10.
U. S. Army Corps of Engineers, 1998b, Onshore field investigation report for the barrier wall
design project; Wyckoff Groundwater Operable Unit: Report to EPA Region 10.
U. S. Army Corps of Engineers, 1998c, Offshore field investigation report for the barrier wall
design project; Wyckoff Groundwater Operable Unit: Report to EPA Region 10.
11
FINAL
TABLES
Pilot Test Site Selection Memorandum Wyckoff/Eagle Harbor Superfund Site, Bainbridge Is., WA
FINAL
Table 1. Alternative Pilot Area Data
PARAMETER (AVERAGE) AREA A AREA B AREA C

Ground surface elevation 17 16 17
Depth to groundwater 10 8 10
Aquifer thickness 35 26 26
Non-marine clay thickness 0 8 5
Aquitard thickness 17 30 26
LNAPL thickness 4 0.7 6
DNAPL thickness 9 7 5
Table 2 Comparison of Alternative Pilot Test Areas by Selection Criteria
CRITERION AREA AREA AREA

A B C
1 Access to drilling locations X X
2 Upgradient location X X X
3 Absence of NAPL upgradient of test area X X X
4 Presence of both LNAPL and DNAPL X X
5 Soil characteristics and stratigraphy similar to X X
overall site
6 Presence of DNAPL on the aquitard X X
7 Presence of sufficient contaminated fine-grained X X
materials for testing of electrical heating
8 Presence of NAPL in the aquitard X X
9 Shoreline location X X
FINAL
FIGURES
Appendix C Well E4 Pumping Test Results
mme 11/16/01
Wyckoff Superfund Site

Thermal Remediation Pilot Test
WELL E4 PUMPING TEST SUMMARY
1. TEST PROCEDURES.
1.1 General
Pumping tests at extraction well E4 were conducted to provide estimates of subsurface hydraulic
parameters for thermal treatment design and operations. The tests consisted of: 1) an 8-hour
step-discharge test with overnight recovery; 2) followed by a 3-day constant-discharge test and
3-day recovery. Well locations are shown on Figure 1, and well data are shown on Table 1.
Discharges and durations for the pumping tests are shown on Table 2.
1.2 Equipment
Pumping was performed using a 3.75-inch Goulds submersible pump driven by a Franklin 2-
horsepower electric motor. The pump and motor combination were designed to deliver a
maximum of 30 gpm at the maximum total dynamic heads which would be incurred with
anticipated water levels along with the pipeline and treatment plant.
A check valve was mounted just above the pump, and the pump and motor assembly was
suspended on a 1.25-inch ID flexible polyethylene column long enough to position the pump
intake at the bottom of the well screen. A 1-inch PVC stilling pipe was clamped alongside the
column pipe, extending to a few feet above the pump intake. The flexible column pipe curved
from vertical to horizontal as it exited the well, and was laid on the ground surface. A spigot for
sample extraction was mounted on the horizontal portion of the pipe, followed in the down
stream direction by two flow meters and a globe valve. The column and discharge pipe
extended 150 feet to a Baker tank, where the discharge was collected before pumping to the site
groundwater treatment plant.
Flows were measured using two 2-inch totalizer meters: a Precision meter measuring cubic feet
per minute, and a Master Meter measuring gallons per minute. The meters were calibrated and
sealed by the manufacturer prior to shipping, and were found to agree within 1% during
simultaneous testing before the step-discharge test. The meters were also checked every 8 hours
during each pumping test, and showed continuing agreement within 1%. Totalizer readings were
used to calculate average discharges for each test period.
1.3 Water level monitoring

Water levels were recorded in the pumping well, 5 extraction wells within a 100-foot radius of
the pumping well (see Figure 1), and in 3 monitoring wells outside the Pilot Area, using 9 Druck
down hole transducers connected to three Geomation data loggers. Electronic data were
collected from 3 days before the step-discharge test through the recovery period, and manual
data collection continued for 3 days after recovery. Tide data were recorded by a fourth data
logger, from a transducer placed in a stilling pipe in the Log Rafting Area. During the pumping
tests and recovery, water levels were recorded whenever a change of 0.05 foot was sensed, or at
minimum 1-hour intervals. Barometric pressure readings were also recorded by data logger, and
all data continuously downloaded to a desktop computer in the project office. Manual backup
water-level readings were taken every 8 to 16 hours from the wells containing transducers. The
background wells for the test were: MW19, 1040 feet southwest of well E4, in the upper aquifer;
PO5, 470 feet northeast, in the upper aquifer, and 9CDMW04, 300 feet north in the lower aquifer.
There was no significant precipitation during the testing, until the second day of the constant
discharge test (see Table 3).
1.4 Step-Discharge Test

The step-discharge test began at 11:25 AM on 15 October 2001, pumping at 3 gpm. The water
level was fairly stable at about 3.4 feet of drawdown throughout the 2-hour step. The pumping
rate was then increased to 6.1 gpm, and drawdown increased to 8.5 feet, continuing to increase
to 9.2 feet over the next 2 hours. After the pumping rate was increased to 9.5 gpm, the water
level dropped to within 3 feet of the pump intake, and the pumping rate was gradually reduced to
8.1 gpm over a period of about hour. Finally, the pumping rate was reduced to 6.7 gpm and
continued for another 1.5 hours. During this time, drawdown was steady at about 13.8 feet.
1.5 Constant-Discharge Test

The constant-discharge test was conducted at 2.9 gpm, less than of the maximum sustainable rate
determined from the step-discharge test. The test started on 16 October 2001 at 0930 hr. Initially
the drawdown was 4.8 feet, gradually increasing to over 6 feet during the next 72 hours. During
the test period, the water level in the pumping well appeared to fluctuate over a range of about
0.4 feet in response to tides.
2. AQUIFER TEST ANALYSES
2.1 Procedures.
2.1.1 General.
Data logger files for the 72-hour constant discharge pumping tests were converted to separate
computer files of elapsed time and drawdown for each well, which were imported to customized
EXCEL spreadsheets. The spreadsheets were used to make data corrections, to plot graphs, and
to perform data analyses. Neuman (1975) analyses and numerical model data fits were
performed for all of the observation wells in the Pilot Area except E5 and E6; these data were
not analyzed due to strong tidal effects. The Neuman method is the standard analysis for
unconfined aquifer test data, however review of the Neuman results indicated the need for
additional work with the existing numerical model. Pumping well (E4) data were not analyzed
because head losses and uncertainties regarding effective diameters for pumping wells preclude
accurate estimates of formation parameters (Driscoll, 1986; Fetter, 1980; Todd, 1980; U.S. Dept.
of Interior, 1985). Well E4 data was used, however to perform well efficiency calculations.
2.1.2 Data Corrections.

The background water-level trend was determined by calculating the average daily static water-
level drop from the pre-test background period through the post-test background period. Water
levels in the wells appeared to increase in response to rainfall over the last two days of the post-
test background period (see Table 3); therefore these data were not included in the trend
calculation. Background corrections for the observation wells ranged from 0.03 to 0.06 feet/day.
Using these corrections, the background water-level decline since the start of pumping was
calculated for, and subtracted from each drawdown measurement.
Barometric effects were not discernable in the water-level data, primarily because of tidal and
trend effects, therefore no barometric corrections were made. An attempt was made to correct
data from wells E5 and E6 for tidal effects, however the relative amplitude of well fluctuations
was different from the marine tidal fluctuations, preventing effective scaling and superposition of
well and tide data. All observation well data was corrected for unconfined conditions by the
method of Jacob (1963).
2.1.3 Neuman Analyses.

Pumping test data from observation wells E1, E2, and E3 were analyzed by Neuman's (1975)
method for partially penetrating wells in unconfined aquifers, modified for dual-boundary
conditions (Easterly, 1995). The modification of Neumans method is implemented by locating
image wells which simulate impermeable boundaries, and generating type curves by
superimposing dimensionless drawdowns for the pumping well and image wells. Image well
locations were derived from two intersecting linear boundaries, which were defined by linear
regressions through the sheet pile wall surrounding the Pilot Area. Custom type-curve data were
generated for each observation and pumping well pair by the computer program DELAY2.2.
The type curves and the test data were both plotted on EXCEL (Microsoft, 1999) spreadsheets
and hard copies were printed of both plots. A preliminary manual fit was performed with the
printed plots, and a final fit was obtained on the computer screen with the test data overlaid on
the type curve plot. Aquifer parameters were automatically calculated by the spreadsheet, based
on the match points used for the fit.
2.1.4 Efficiency Calculations.

The well efficiency for the pumping well (E4) was estimated using the method proposed by
Todd (1980) (see Table 2). The method consists of plotting specific capacities for each step of
the step-discharge test, and comparing the estimated laminar head loss (formation head loss)
with the maximum drawdown (formation head loss plus well head loss) for each step.
2.1.5 NUFT Model Data Fits.

The Neuman analyses appeared to produce anomalous estimates of storativity, specific yield, and
vertical anisotropy (see Table 4). The vertical anisotropy results were of particular concern
because they seem to contradict field observations which show only moderate stratification in
the upper aquifer sands. The discrepancies between the Neuman results and expected or typical
formation parameters are consistent with findings by Nwankwor et al (1992), Akindunni and
Gillham (1992), and Halford (1997):
Application of Neumans fully-saturated-vertical-flow assumption to fine-grained
materials with significant capillary retention can cause overestimation of storativity and
horizontal hydraulic conductivity, and underestimation of vertical hydraulic conductivity
and specific yield.
A 3-dimensional numerical multiphase model is the most reliable analysis tool for test
data from unconfined aquifers.
For these reasons, additional data analyses were performed by calibrating an existing numerical
model to the well drawdown data.
The numerical model had been previously prepared as an independent project with no cost to the
Government, for the purpose of simulating steam injection operations. It utilizes the multiphase
code NUFT, and contains the same stratigraphy, material and fluid parameters as the NUFT and
MODFLOW models used in the 90% Design Analysis (USACE, 2001). The Pilot Area and
surrounding vicinity was simulated by a 3-dimensional grid consisting of 50,000 nodes, with a
horizontal grid spacing of 2 meters inside the Pilot Area, and a vertical grid spacing of 1.5
meters in the upper aquifer. Fixed head boundaries were used on the north and south sides of the
domain, to create average northward flow conditions at the site. This model provides a valid
representation of partially-saturated delayed-yield conditions, as well as an accurate portrayal of
the sloping aquitard and irregularly-shaped sheet pile wall. Some error is inherent in the use of a
numerical model for pumping test analyses, however, because wells can only be located at the
centers of cells. Since the Pilot Area NUFT model has a 2-meter horizontal grid spacing, wells
might be simulated at locations as far as 3 feet from their actual locations.
2.2 Results.
2.2.1 General.
Results of the test data analyses are shown in Table 4. Data points were scattered over early
portions of the drawdown curves, probably due to a combination of instrument sensitivity and
heavy equipment operations near the test area. Nevertheless, data from wells E1, E2, and E3
generated drawdown curves of good quality that matched well with the Neuman type curves.
Matches with NUFT model output were fair to good, and could be improved with additional
model runs. Wells E5 and E6 were strongly affected by tides, and reliable fits could not be
obtained. For all curve fits, emphasis was placed on late time data, since the early data may be
affected by borehole conditions. The results of pumping test analyses for both pumping tests are
shown on Table 4. In general, results from Neuman analyses show higher horizontal hydraulic
conductivities and vertical anisotropy ratios (Kr/Kz) than the NUFT model fits.
2.2.2 Background Wells

Water levels in the background wells did not appear to be influenced by the pumping test. All of
the background wells were affected by tidal fluctuations, with amplitudes of 0.7 feet at well
MW19 (upper aquifer, upgradient), 0.8 feet at well PO5 (upper aquifer, downgradient), and 4.5
feet at well 99CD04 (lower aquifer, downgradient).
2.2.3 Neuman Analyses.

Test data fits to the Neuman type curves are shown in Figures 2, 3, and 4. The hydraulic
conductivity for the three wells which were analyzed ranged from 29.7 at well E2 to 66.8
feet/day at well E3, averaging 46.3 ft/day. Specific yields range from 0.033 at well E2 to 0.007
at well E3, averaging 0.019. These specific yield values are lower than the expected range for
sands (10% to 35%; Fetter, 1988), suggesting that saturated soils are not draining completely as
the phreatic surface is lowered by pumping. Vertical anisotropy ratios (Kr/Kz) varied from 27.9
to 388.7, with a average of 156.2. These values are normally indicative of a high degree of
stratification, which has not been apparent in Pilot Area soils. Because of the anomalous results
for both specific yield and vertical anisotropy, the Neuman method appears to be inappropriate
for the fine to medium silty marine sands in the Pilot Area.
2.2.4 NUFT Model fits
Test data fits to the NUFT model predictions are shown in Figures 5, 6, and 7. NUFT model
output did not fit the test data as well as the Neuman type curves, primarily because of the large
number of model runs required to produce exact fits. Early time data-fitting was not attempted
because NUFT does not have a storativity parameter, and elastic storage simulated in the early
portion of the drawdown curve is solely a function of the compressibility of water.
Twenty-five model runs were performed, with horizontal hydraulic conductivities ranging from
10 feet/day to 31 feet/day, and vertical anisotropy ratios ranging from 2 to 55. The best data fits
were for model runs with 26 feet/day and vertical anisotropy values ranging from 3 to 7,
averaging 4.7. All model runs employed the porosity and residual saturation values used for
Pilot Test design: 0.28 and 0.15 respectively, or a specific yield of 0.13. The best-fit modeled
drawdown curves are still offset slightly from the test data. This offset may be caused by
discretization error, or by differences in the formation specific yield versus the modeled storage
capacity. The groundwater storage capacity in the model can be adjusted as a function of
porosity and pressure-saturation-permeability characteristics, however several more model runs
would be required to obtain exact fits with the test data.
3. CONCLUSIONS
The NUFT model data-fitting appears to be the most appropriate method of analysis for
the test data, because it incorporates partially-saturated vertical drainage, sheet pile
boundaries, and 3-dimensional stratigraphy.
The average horizontal hydraulic conductivity in the Pilot Area appears to be about 26
feet/day, greater than the value of 10 feet/day used for Pilot Test design.
The average vertical hydraulic conductivity in the Pilot Area appears to be about 5
feet/day; i.e. the average vertical anisotropy ratio is about 5, considerably less than the
value of 20 used for Pilot Test design.
Pumping test results are consistent with the porosity value used for Pilot Test design
(0.28) and also with the pressure-saturation-permeability relationships used for design.
Pumping test data show strong tidal influence at well E6, and moderate influence also at
well E5. This phenomenon suggests that the aquitard is thin on the east side of the Pilot
Area, particularly in the deepest portion around well E6.
4. RECOMMENDATIONS
Because of the unique conditions at the Wyckoff site, upper aquifer pumping test
analyses should be performed using a 3-dimensional numerical model. The most reliable
results will be obtained from a multiphase model.
Since the average permeability of soils in the Pilot Area is greater than anticipated, the
injection well screen length can be reduced to 5 feet. Injection rates for the reduced
screen length will be similar to design assumptions for 10-foot screens.
The low vertical anisotropy values predicted from pumping test data also support a
reduced injection screen length. Conditions appear to be favorable for upward steam
migration into near-surface soils, the shorter screens should not significantly inhibit
vertical steam flow.
The steam injection model used for Pilot Test design should be revised using the aquifer
parameters predicted from the pumping test data.
Future site operations should emphasize vertical hydraulic control in the well E6 vicinity.
The aquitard may be thin or absent at this location, and the lower aquifer may be
vulnerable to contamination from above.
5. REFERENCES
Akindunni, F. F., and Gillham, R. W., 1992, Unsaturated and Saturated Flow in Response to
Pumping of an Unconfined Aquifer: Numerical Investigation of Delayed Drainage:
Groundwater, v. 30, n. 6, p. 873-884.
Driscoll, F. G., ed., 1986, Groundwater and wells: St. Paul MN, Johnson Filtration Systems Inc.
Easterly, M. E, 1995, Analysis of pumping test data using a dual-bounded, unconfined,

analytical aquifer model, Santa Ana River Canyon, San Bernardino County, California:
California State University Los Angeles, Master of Science thesis.
Fetter, C. W., 1988, Applied geohydrology: Columbus OH, Merrill Publishing Co..
Jacob, C. E., 1963, Determining the permeability of water-table aquifers, in Bentall, R., ed.,
Methods of determining permeability, transmissibility and drawdown: U. S. Geological Survey
Water-Supply Paper 1536-I, p. 245-271.
Halford, K. J., 1997, Effects of Unsaturated Zone on Aquifer Test Analysis in a Shallow
Aquifer System: Groundwater, v. 35, n. 3, p. 512-522.
Microsoft Corporation, 1999, EXCEL 2000: Redmond WA, Microsoft Corporation.
Neuman, S. P., 1975, Analysis of pumping test data from anisotropic unconfined aquifers
considering delayed gravity response: Water Resources Research, v. 11, n. 2, p. 329-342.
Nwankwor, G. I., Gillham, R. W., van der Kamp, G., Akindunni, F. F., Unsaturated and
Saturated Flow in Response to Pumping of an Unconfined Aquifer: Field Evidence of Delayed
Drainage: Groundwater, v. 30, n. 5, p. 690-700.
Todd, D. K., 1980, Groundwater hydrology: New York NY, John Wiley and Sons, Inc..
U. S. Army Corps of Engineers, 2001, Final Design Analysis, Thermal Remediation Pilot Study,
Soils and Groundwater Units, Wyckoff/Eagle Harbor Superfund Site, Bainbridge Island,
Washington: Report to EPA, in progress.
U. S. Department of the Interior, Bureau of Reclamation, 1981, Ground water manual:
Washington DC, U. S. Government Printing Office.
TABLES
Table 1.
Pumping Test Well Data
Coordinates Elevation (MLLW) Hole Screen Static water level Aquifer base Completion
Well Diameter (in) Top Bottom Elevation
no. N E Casing Ground Dia.(in) Depth ID OD depth depth Depth (MLLW) Date Depth Elevation Date Method
Cable
E1 229375.7 1229072.3 21.20 18.60 16.00 36.4 10.00 10.75 6.6 31.5 10.36 8.2 10/16/01 31.35 -12.75 10/1/01 tool
Cable
E2 229389.7 1229136.6 21.10 18.66 16.00 40.5 10.00 10.75 6.7 35.5 10.52 8.1 10/16/01 35.50 -16.84 10/1/01 tool
Cable
E3 229318.5 1229064.0 20.00 18.57 16.00 38.8 10.00 10.75 7.7 31.0 10.79 7.8 10/16/01 28.80 -10.23 10/9/01 tool
Cable
E4 229332.0 1229124.0 20.80 18.84 16.00 36.5 10.00 10.75 7.2 31.6 11.04 7.8 10/16/01 31.50 -12.66 10/8/01 tool
Cable
E5 229287.8 1229171.7 20.70 18.30 16.00 35.0 10.00 10.75 6.8 29.6 10.47 7.8 10/16/01 30.00 -11.70 10/2/01 tool
Cable
E6 229348.1 1229191.8 20.00 18.50 16.00 43.0 10.00 10.75 7.7 38.1 10.05 8.5 10/16/01 38.00 -19.50 10/4/01 tool
Notes:
1. All dimensions are in feet unless otherwise noted.
Static water levels were measured just prior to the E4 constant-discharge
2. test.
Table 2
Well E4 Pumping Test Summary
Test Start date Q (gpm) Duration (min) s max (ft) SC (gpm/ft) Efficiency (%)
Step 1 10/15/01 3.01 125 3.5 0.85 39.7
2 6.13 121 9.3 0.66 30.7
3 8.10 30 18.3 0.44 22.5
4 6.73 90 13.9 0.48 20.6
Constant Discharge 10/16/01 2.90 4320 6.01 0.48
Q = discharge
smax = maximum drawdown in the pumping well
SC = specific capacity
Table 3
Precipitation Data for Wyckoff Superfund Site
24-Hour
Day
(Midnight-Midnight)
Test Period (October
2001) Precipitation
(Inches)
12 0.04
Pre-test
background 13 0.04
14 0.03
Step test 15 0.00
16 0.08
Constant
discharge test 17 0.00
18 0.60
19 0.05
Recovery 20 0.00
21 0.20
22 0.05
Post-test 23 0.20
recovery 24 0.10
25 0.05
Table 4
Results of Pumping Test Analyses
Bounded Neuman Analyses NUFT Model Fits

s Background
Well Location Q r b max Correction Kr Kr/Kz S Sy Fit quality Kr Kr/Kz Sy Fit quality
(gpm) (ft) (ft) (ft) (ft/day) (ft/d) () (ft/ft) (ft/ft) A B (ft/d) () (ft/ft) A B
20.9
E1 wall 67.69 9 0.98 -0.059 42.4 52.0 0.0135 0.0182 fair good 26.0 4.0 0.13 fair
24.9
E2 wall 59.06 8 1.08 0.000 29.7 27.9 0.0118 0.0325 fair good 26.0 3.0 0.13 fair
18.0
E3 interior 61.50 1 0.88 0.000 66.8 388.7 0.0055 0.0073 fair good 26.0 7.0 0.13 good
20.4
E4 interior 2.904 0.00 6 6.01 -0.046
19.5
E5 interior 65.03 3 0.59 -0.033
27.9 1.36
E6 wall 69.69 5 5 -0.032
Average -0.028 46.3 156.2 0.0103 0.019 26.0 4.7 0.13
Q = discharge
r = distance from pumping well to observation well
b = saturated aquifer thickness Notes: 1. E4 (pumping well) data were not analyzed.
smax = maximum drawdown 2. E5 and E6 data were strongly affected by tides and were not analyzed.
Kr = horizontal hydraulic conductivity
vertical hydraulic
Kz = conductivity
S = storativity
Sy = specific yield
FIGURES
Figure 1. Pumping Test Well Layout
Fit Results: Kr= 42.4 ft/d S= 0.013
Pumping Well: E4
Observation Well: E1 Kz= 0.82 ft/d Sy= 0.018
SIGMA for type curves =0.0001 BETA= 0.2

10 ts 10
1E-1 1E+0 1E+1 1E+2 1E+3 1E+4 1E+5 1E+6 1E+7
1E+2
0.001
1
2.00E+01 1 100 1000 10000 0.004 1
0.01
1E+1 0.03
0.06
0.1
0.2
0.4 1 10 100
0.6
0.8
sD 1E+0 1.
1.5
2.
2.5
3. Type Curves
0.001 Test Data (A fit)
4.
0.2 5. Test Data (B fit)
1E-1
6. Match Points
7. A Scales (ft vs min)
B Scales (ft vs min)
0.03
1E-2
1E-5 1E-4 1E-3 1E-2 1E+0 1E+1 1E+2 1E+3
ty 1E-1
1E-1
Figure 2. Well E1 Neuman Analysis

Pumping Well: E4
SIGMA for type curves =0.0001 BETA= 0.2

ts
10 10
1E-1 1E+0 1E+1 1E+2 1E+3 1E+4 1E+5 1E+6 1E+7
1E+2
0.001
1.67E+01
1 1 100 1000 10000 0.004 1
0.01
1E+1 0.03
0.06
0.1
0.2
0.41 10 100
0.6
sD 1E+0 0.8
1.
1.5
2. Type Curves
0.001
3.
1E-1 0.2 Test Data (B fit)
4. Match Points
6. B Scales (ft vs min)
0.03 7.
1E-2
1E-5 1E-4 1E-3 1E-2 1E-1
1E-1 1E+0 1E+1 1E+2 1E+3
ty

Pumping Well: E4
10type curves =0.0001
SIGMA for BETA= 0.03 10
ts
1E-1 1E+0 1E+1 1E+2 1E+3 1E+4 1E+5 1E+6 1E+7
1E+2
2.70E+01
1 1 100 1000 10000 1
0.001
0.004
1E+1 0.01
0.03
0.06
0.1
1 10 100
0.2
0.4
sD 1E+0
0.6
0.8
1.
0.001 1.5
Type Curves
2.
1E-1 0.2 3. Test Data (B fit)
4. Match Points

6. B Scales (ft vs min)
0.03
7.
1E-2
1E-5 1E-4 1E-3 1E-2 ty 1E-1

1E-1 1E+0 1E+1 1E+2 1E+3

PUMPING TEST DATA (log-log plot)
Pumping Well: E4 Observation Well: E1
10.000
1.000
s (ft)
0.100
0.010
Test data
Model
0.001
1.00 10.00 100.00 1000.00 10000.00 100000.00
t (min)
Figure 5. NUFT data fit for well E1

10.000
1.000
s (ft)
0.100
0.010
Test data
Model
0.001
0.10 1.00 10.00 100.00 1000.00 10000.00
t (min)

10.000
1.000
s (ft)
0.100
0.010
Test data
Model
0.001
0.10 1.00 10.00 100.00 1000.00 10000.00
t (min)

Appendix D Microcosm Study
2340 Stock Creek Blvd.
Rockford TN 37853-3044
Phone (865) 573-8188
Fax: (865) 573-8133
Email: [email protected]
Microbial Analysis Report
Client: Marta Richards Phone: (513) 569-7692

U.S. EPA
Fax:
MI Identifier: 9bat Date Rec.: 10/30/01 Report Date: 12/18/01
Analysis Requested: PLFA
Project: Eagle Harbor
Comments:
NOTICE: This report is intended only for the addressee shown above and may contain confidential or privileged information. If
the recipient of this material is not the intended recipient or if you have received this in error, please notify Microbial Insights, Inc.
immediately. Thank you for your cooperation.
2340 Stock Creek Blvd.
Rockford TN 37853-3044
Phone (865) 573-8188
Fax: (865) 573-8133
Email: [email protected]
Microbial Analysis Report

Executive Summary
The microbial communities from eighteen soil samples from a microcosm study were characterized by
phospholipid fatty acid content (PLFA Analysis). Results from this analysis revealed the following:
Generally, biomass estimates were higher in the vadose zone samples than in the saturated zone
samples.
PLFA profiles showed that there were noticeable differences between the microbial communities in the
vadose and saturated zones.
Ratios of fatty acid biomarkers that provide indication of activity showed that overall turnover rates
appeared to be noticeably slower in the vadose zone samples, as compared to the saturated zone
samples.
Ratios of fatty acids biomarkers that indicate a metabolic response to environmental conditions showed
that the Gram-negative bacteria in the saturated zone were showing a much greater response to
environmentally induced stress conditions (i.e. toxicity, starvation) than in the vadose zone samples.
2
Microbial Insights, Inc. Project: Eagle Harbor
Overview of Approach:
Phospholipid Fatty Acid (PLFA) Analysis
The analysis of microbial membrane lipids, specifically phospholipid fatty acids (PLFA), is an effective tool for
monitoring microbial responses to their environment. Lipids are essential cellular components of the membrane
of all cells and play a role as storage materials. The PLFA profiles simultaneously contain general information
about the phylogenetic identity and physiological status of microbes. The microbial membrane reflects the
nature of both the intracellular components and the extracellular environmental conditions. Thus, PLFA analysis
tells us what types of microbes are present in a system and how they are reacting to environmental factors (e.g.,
pollution or disturbance). PLFA analysis is based on the extraction and separation of lipid classes, followed by
quantitative analysis using gas chromatography/mass spectrometry (GC/MS). The individual fatty acids differ in
chemical composition depending on the organism and environmental conditions. PLFA analysis provides
quantitative insight into three important attributes of microbial communities: viable biomass, community
structure, and metabolic activity.
Procedures:
PLFA analysis
Lipids were recovered using a modified Bligh and Dyer method (3). Extractions were performed using one-phase chloroform-methanol-
buffer extractant. Lipids were recovered, dissolved in chloroform, and fractionated on disposable silicic acid columns into neutral-, glyco-,
and polar-lipid fractions. The polar lipid fraction was transesterified with mild alkali to recover the PLFA as methyl esters in hexane. PLFA
were analyzed by gas chromatography with peak confirmation performed by electron impact mass spectrometry (GC/MS). PLFA
nomenclature follows the pattern of A:BC. The A position identifies the total number of carbon atoms in the fatty acid. Position B is the
number of double bonds from the aliphatic () end of the molecule. Position C designates the carbon atom from the aliphatic end before
the double bond. This is followed by a c for cis or a t for trans configuration. The prefix i and a stand for iso and anteiso branching.
Mid-chain branching is noted by me, and cyclopropyl fatty acids are designated as cy (4). Example: 18:17c is 18 carbons long with
one double bond occurring at the 7th carbon atom from the end, and the hydrogen molecules attached to the doubly bonded carbon
molecules are in the cis conformation.
Results and Discussion:
Biomass Content
Phospholipid fatty acids are found in the membranes of all living cells but decompose quickly upon cell death
because cellular enzymes hydrolyze the phosphate group within minutes to hours of cell death (3, 5). Thus,
measuring the total amount of PLFA content provides a quantitative measure of the viable microbial biomass
present.
Generally, biomass estimates (as determined by the total concentration of PLFA) showed that the vadose zone
samples contained higher biomass levels than in the saturated zone samples. Biomass levels averaged 2,461
pmoles of PLFA/g dry wt. (1,312) in the vadose zone as compared to 1,628 pmoles PLFA/g dry wt. (1,004) in
the saturated zone samples.
Within the saturated zone samples, biomass concentrations were at least 2 times higher in the P1-A and P1-B
samples.
3
5,000
4,500
Pmoles PLFA/g dry wt.
4,000
3,500
3,000 Eukaryotic
2,500 Bacterial
2,000
1,500
1,000
500
0
h1
h2
h1
h2
h1
ate 2
ate -A
-B
P1
7
e1
e2
e1
e2
e1
e2
h
T7
1-
1-
77
77
a tc
a tc
a tc
a tc
a tc
a tc
Liv
Liv
Liv
Liv
Liv
Liv
dP
dP
se
se
dT
dT
3B
3B
4B
4B
6B
6B
do
3
ate
do
VP
VP
VP
VP
VP
VP
ate
Va
SP
SP
SP
SP
SP
SP
Va
tur
tur
tur
tur
Sa
Sa
Sa
Sa
Sample
Figure 1. Biomass content is presented as the total amount of phospholipid fatty acids (PLFA) extracted from a given sample. Bacterial
biomass is calculated based upon PLFA attributed specifically to bacteria, whereas eukaryotic biomass is based on PLFA associated with
higher organisms.
Community Structure
The PLFA patterns derived from environmental samples provide a quantitative profile of the microbial
population, which accurately mirrors differences in community composition among samples. Specific groups of
microbes contain different fatty acid profiles making it possible to distinguish between them (1, 2, 4, 6). Table 1
describes the six major structural groups employed.
Table 1. Description of PLFA Structural Groups.
PLFA Structural Group General classification

Monoenoic (Monos) Found in Gram-negative bacteria, which can be fast growing, utilize many carbon sources,
and adapt quickly to a variety of environments.
Terminally Branched Saturated (TerBrSats) Representative of Gram-positive bacteria, but also are found in the cell membranes of
some Gram-negative bacteria.
Branched Monoenoic (BrMonos) Commonly found in the cell membranes of obligate anaerobes, such as sulfate- or iron-
reducing bacteria
Mid-Chain Branched Saturated (MidBrSats) Common in actinomycetes, sulfate-reducing bacteria, and certain Gram-positive bacteria.
Normal Saturated (Nsats) Ubiquitous in both prokaryotic and eukaryotic organisms, though dominant fatty acids
within this group will vary among organisms.
Polyenoic Found in organisms such as fungi, protozoa, algae, higher plants, and animals.
Comparison of the PLFA profiles from these samples showed that there were noticeable differences between
samples collected from the vadose and saturated zone (see Figure 2). Cluster analysis of the PLFA profiles
clearly shows distinct branching for samples collected from each zone (Figure 3). Within each cluster, replicate
samples from the various sampling locations (P3, P4, etc.) cluster together indicating that each replicate were
similar in community composition.
4
100%
80%
% of total PLFA
60%
40%
20%
0%
h2
h1
h2
-A
-B
P1
7
e1
e2
e1
e2
e1
e2
tch
tch
tch
T7
1-
1-
77
77
a tc
a tc
a tc
Liv
Liv
Liv
Liv
Liv
Liv
dP
dP
se
Ba
Ba
Ba
se
dT
dT
4B
6B
6B
do
3
ate
ate
do
4
VP
VP
VP
VP
VP
VP
ate
ate
Va
SP
SP
SP
SP
SP
SP
Va
tur
tur
tur
tur
Sa
Sa
Sa
Sa
Gram+/Anaerobic Gram - (TerBrSats) Gram - (Monos)
Anaerobic Metal Reducers (BrMonos) SRB/Actinomycetes (MidBrSats)
Genera (Nsats) Eukaryotes (polyenoics)
Figure 2. A comparison of the relative percentages of total PLFA structural groups in the samples analyzed. Structural groups are assigned
according to PLFA chemical structure, which is related to fatty acid biosynthesis. See Table 1 for detailed descriptions of structural groups.
Samples collected from both the saturated and vadose zones were primarily composed of Gram negative
bacteria (indicated by percentage of monoenoic PLFA). High proportions of Gram-negative bacteria are of
particular interest in contaminated ecosystems due to their ability to utilize a wide range of carbon sources and
adapt quickly to environmental conditions. Proportions of Gram-negative bacteria were higher in the saturated
zone samples (~716%) than in the vadose zone (~503%).
Principal components analysis (PCA) indicated that all of the vadose zone samples formed a tight cluster
together indicating few differences within their microbial communities. This clustering was strongly influenced
by the proportion of cy19:0, cy17:0, 18:1w9c and 10me16:0. The Gram-negative biomarkers cy17:0 and cy19:0
are produced when the turnover rates of bacteria decrease (discussed in further detail in the metabolic status
section). Further insight into this trend will be evaluated after the final event, as high proportions of cyclopropyl
fatty acids have also been found in certain anaerobes (members of thiobacillus contained unusually high
proportions of cy17:0 and cy19:0). The biomarker 18:19c, is a precursor for eukaryotic organisms but is also
found in Gram negative bacteria. Due to its strong correlation with 18:26 (prominent in fungi) it is considered
to be from eukaryotic origin (correlation 0.89). The mid-chain branched biomarker 10me16:0 is common in
anaerobic metal reducing bacteria (Desulfobacter-type) and likely indicate increased proportion of this type of
bacteria in the vadose zone samples.
Saturated zone samples formed a loose cluster together, which was mainly influenced by the Gram-negative
biomarkers (18:17c, 16:17c, 18:17t, and 16:17t) and the normal saturated biomarker 16:0 (found in higher
proportions in bacteria). High proportions of trans fatty acids (18:17t and 16:17t) within several of the
saturated zone samples are indicative of a metabolic response due to environmentally stress full conditions
such as toxicity or starvation (again further details provided in the metabolic status section).
5
Community Similiarity Tree Diagram

Ward`s method
1-Pearson r
VP3_1
VP3_2
VP4_1 Vadose Zone
VP4_2
VP6_1
VP6_2
VP1
VT77
SP3_1
SP4_1
SP4_2
SP6_1 Saturated Zone
SP6_2
SP3_2
SP1A
SP1B
ST77A
ST77B
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Linkage Distance
Figure 3. Hierarchical Cluster Analysis (HCA) of the PLFA profiles. Similarities between pairs of samples in a data set are calculated and
compared. When distances between samples are relatively small, this implies that the samples are similar. The primary purpose of HCA is to
present data in a manner that emphasizes natural groupings.
6
0.08
0.06 VP3-2 V
VP6-2 P4-2
0.06151
VP6-1
VP3-1
VP1 VT77
0.04
0.02
Factor 2
0.00 SP3-2
-0.02
ST77B SP3-1
-0.04
SP1A SP1B
ST77A
SP4-1
-0.06
S6-2 SP-4-2 SP6-1
-0.08
0.184 0.186 0.188 0.19 0.192 0.194 0.196
Factor 1
0.4
cy19:0
0.3
10me16:0
18:1w9c
0.2 i17:1w7c
16:1w5c
i15:0
cy17:0
a15:0
'18:0
0.1
Factor 2
-0.1 '16:0
18:1w7t
-0.2
-0.3 18:1w7c
16:1w7c
16:1w7t
-0.4
0 0.1 0.2 0.3 0.4 0.5 0.6
Factor 1
Figure 4. Principal components analysis (PCA) is built on the assumption that variation implies information in the same way that HCA is built
on the assumption that short multivariate distance implies similarity. Principal components analysis projects the multivariate data (PLFA
profiles) onto a reduced number of dimensions (principal components) thereby simplifying the data so relationships between sample sets can
be observed easily. Thus, PCA analysis can show which microbial communities are similar by visually overlaying the top graph onto the
bottom graph.
Metabolic Activity
Lipid composition of microorganisms is a product of metabolic pathways and thus reflects phenotypic responses
of the organisms to their environment. Knowledge of specific lipid biosynthetic pathways can provide insight into
the metabolic activity of the microbial community because certain fatty acids provide indications of turnover rate
and physiological responses to environmental conditions. In Gram negative bacteria there are two main
biosynthetic pathways: one that preferentially synthesize 16:17c (herein referred to as Group A Gram negative
bacteria) and another that preferentially synthesize 18:17c (Group B). Ratios of cy/7c and 7t/7c can be
7
used as quantitative indicators of how a portion or all of the Gram-negative community is responding to
environmental factors (toxicity, starvation, etc.) and/or engineered treatment. Specifically, Gram-negative
bacteria form cyclopropyl fatty acids (f.a.) (cy17:0 & cy19:0) preferentially over monoenoic f.a. (16:17c and
18:17c) as the turnover rate decreases.
Ratios of cy /7c showed that overall turnover rates appeared to be noticeably slower in the vadose zone
samples, as compared to the saturated zone samples. However, as mention previously some anaerobic
bacteria contain high proportions of cyclopropyl fatty acids. Further insight into this trend will be evaluated after
the final event of the microcosm study.
1.60
1.40 group A (cy17:0/16:1w7c)
1.20
Ratio cy/w7c
group B (cy19:0/18:1w7c)
1.00
0.80
0.60
0.40
0.20
0.00
h1
h2
h1
h2
h1
h2
-A
-B
P1
7
e1
e2
e1
e2
e1
e2
T7
1-
1-
77
77
a tc
a tc
a tc
a tc
a tc
a tc
Liv
Liv
Liv
Liv
Liv
Liv
dP
dP
se
se
dT
dT
3B
3B
4B
4B
6B
6B
do
3
ate
ate
do
VP
VP
VP
VP
VP
VP
ate
ate
Va
SP
SP
SP
SP
SP
SP
Va
tur
tur
tur
tur
Sa
Sa
Sa
Sa
Sample
Figure 3. Growth rate of the Gram-negative community as assessed by the ratio of cyclopropyl f. a. to 7c f. a. Specifically, 16:17c and
18:17c fatty acids are converted to cyclopropyl fatty acids (cy17:0 & cy19:0) as microbial growth slows (i.e., a high ratio indicates decreased
turnover rate).
Gram-negative bacteria also generate trans fatty acids to minimize the permeability of their cellular membranes
as an adaptation to less favorable environments (5). Ratios of trans to cis fatty acids were markedly higher in
the saturated zone samples than calculated in the vadose zone samples. This observation suggests that the
Gram-negative bacteria (particularly the Group A Gram negatives) were showing a much greater response to
conditions of environmentally induced stress (i.e. toxicity, starvation) in the saturated zone samples.
0.70
0.60 group A (16:1w7t/16:1w7c)
Ratio w7t/w7c
0.50 group B (18:1w7t/18:1w7c)

0.40
0.30
0.20
0.10
0.00
1
-B
-A
-B
P1
7
e1
e2
e1
e2
e1
e2
tch
tch
tch
tch
tch
tch
T7
1-
P1
77
77
Liv
Liv
Liv
Liv
Liv
Liv
dP
se
Ba
Ba
Ba
Ba
Ba
Ba
se
dT
dT
d
do
3
ate
ate
do
6
VP
VP
VP
VP
VP
VP
ate
ate
Va
SP
SP
SP
SP
SP
SP
Va
tur
tur
tur
tur
Sa
Sa
Sa
Sa
Sample
Figure 4. Adaptation of the Gram-negative community to changes in the environment is determined by the ratio of 7t/7c fatty acids. Ratios
(16:17t/16:17c and 18:17t/18:17c) greater than 0.1 have been shown to indicate an adaptation to a toxic or stressful environment,
resulting in decreased membrane permeability.
8
Table 2. Summary of PLFA results.
Biomass (pmoles PLFA/g dry wt. of sample) Community Structure (% of total PLFA) Metabolic Activity Physiological Response
Sample Name Total Biomass Cell equivalent Bacterial Eukaryotic Ratio Gram+/ Gram - Anaerobic SRB/ Genera Eukaryotes Group A Gram Group B Gram Group A Gram Group B Gram
value1 biomass biomass bacteria/ anaerobic (Monos) metal Actinomycetes (Nsats) (polyenoics) (cy17:0/16:17c) (cy19:0/18:1w7c) (16:17t/16:17c) (18:17t/18:17c)
eukarya Gram - reducers (MidBrSats)
(TerBrSats) (BrMonos)
VP3 Live1 1,212 2.42E+07 1,017 195 5 9.3 45.6 3.5 6.9 18.5 16.1 0.31 0.46 0.03 0.05
VP3 Live2 1,572 3.14E+07 1,303 269 5 8.2 46.6 3.4 6.8 18.0 17.1 0.34 0.47 0.04 0.04
VP4 Live1 2,986 5.97E+07 2,665 321 8 9.7 49.1 3.2 7.4 19.9 10.8 0.66 0.78 0.05 0.01
VP4 Live2 2,155 4.31E+07 1,923 233 8 9.0 49.7 3.3 7.6 19.7 10.8 0.67 0.74 0.05 0.00
VP6 Live1 4,110 8.22E+07 3,824 286 13 10.2 50.5 2.5 8.1 21.7 7.0 1.39 1.03 0.10 0.08
VP6 Live2 4,370 8.74E+07 4,059 310 13 10.1 50.1 2.7 7.8 22.2 7.1 1.51 1.06 0.10 0.08
Vadose P1 739 1.48E+07 685 53 13 6.1 56.2 2.5 10.7 17.2 7.2 0.88 0.75 0.05 0.00
Vadose T77 2,547 5.09E+07 2,411 137 18 6.2 52.9 1.8 12.1 21.6 5.4 1.41 0.84 0.12 0.07
SP3 Batch1 1,130 2.26E+07 1,077 53 20 3.7 64.8 0.6 2.6 23.5 4.7 0.14 0.10 0.60 0.15
SP3 Batch2 943 1.89E+07 876 68 13 5.2 57.2 2.0 4.7 23.8 7.2 0.22 0.13 0.28 0.13
SP4 Batch1 1,481 2.96E+07 1,448 33 44 1.6 72.5 0.5 1.7 21.5 2.2 0.10 0.04 0.54 0.14
SP4 Batch2 1,529 3.06E+07 1,512 17 90 1.4 74.2 0.0 0.4 22.9 1.1 0.10 0.04 0.46 0.12
SP6 Batch1 1,468 2.94E+07 1,450 17 83 1.2 72.7 0.0 0.2 24.6 1.2 0.18 0.05 0.43 0.10
SP6 Batch2 1,394 2.79E+07 1,382 11 122 1.2 72.7 0.0 0.7 24.6 0.8 0.16 0.04 0.34 0.09
Saturated P1-A 3,479 6.96E+07 3,435 44 78 1.2 77.5 0.3 0.7 19.0 1.3 0.20 0.06 0.09 0.02
Saturated P1-B 3,412 6.82E+07 3,361 52 65 1.6 76.5 0.6 0.7 19.2 1.5 0.20 0.06 0.08 0.02
Saturated T77-A 814 1.63E+07 783 31 26 2.0 72.9 0.0 0.5 20.8 3.8 0.22 0.10 0.18 0.05
Saturated T77-B 635 1.27E+07 569 66 9 1.4 68.7 0.4 0.5 18.6 10.4 0.34 0.10 0.18 0.05
1 12 8
The cell equivalent value is calculated from experiments with typical bacteria isolated from soil and water. This value is based on 2.0 x 10 cells per gram dry weight of cells and 10 picomoles of
phospholipid/gram dry weight of cells. The number of cells/gram of dry weight may vary and is dependent on the environmental conditions from which the microorganisms were recovered.
Quality Assurance Section
Sample Arrival and Holding Times:
Eighteen samples were received between 10/30/01 and , accompanied by a chain of custody form. All arrival
conditions and required holding times were acceptable according to SOP #SREC.
Sample Analysis and QA/QC Parameters:
Samples were analyzed under the U.S. EPA Good Laboratory Practice Standards: Toxic Substances Control
Act (40 CFR part 790). All samples were processed according to standard operating procedures.
Notes: No QC or analytical problems were encountered.
Calibrations and Solvent Checks:
All laboratory equipment and instruments used throughout the analyses were calibrated and operated within
acceptable ranges. The instruments were calibrated according to Standard Operating Procedures (EQ4). All
solvents used in these analyses were tested for purity.
Data Validation:
All data analyses were performed correctly. All calculations and transcriptions of raw and final data were
verified.
References:
1. Dowling, N. J. E., F. Widdel, and D. C. White. 1986. Phospholipid ester-linked fatty acid biomarkers
of acetate-oxidizing sulfate reducers and other sulfide forming bacteria. Journal of General Microbiology
132:1815-1825.
2. Edlund, A., P. D. Nichols, R. Roffey, and D. C. White. 1985. Extractable and lipopolysaccharide fatty
acid and hydroxy acid profiles from Desulfovibrio species. Journal of Lipid Research 26:982-988.
3. White, D. C., W. M. Davis, J. S. Nickels, J. D. King, and R. J. Bobbie. 1979. Determination of the
sedimentary microbial biomass by extractable lipid phosphate. Oecologia 40:51-62.
4. White, D. C., H. C. Pinkart, and D. B. Ringelberg. 1997. Biomass measurements: Biochemical

approaches, p. 91-101. In C. J. Hurst, G. R. Knudsen, M. J. McInerney, L. D. Stetzenbach, and M. V.
Walter (ed.), Manual of Environmental Microbiology. ASM Press, Washington.
5. White, D. C., and D. B. Ringelberg. 1995. Utility of signature lipid biomarker analysis in determining in
situ viable biomass, community structure, and nutritional / physiological status of the deep subsurface
microbiota. In P. S. Amy and D. L. Halderman (ed.), The microbiology of the terrestrial subsurface.
CRC Press, Boca Raton.
6. White, D. C., J. O. Stair, and D. B. Ringelberg. 1996. Quantitative comparisons of in situ microbial
biodiversity by signature biomarker analysis. Journal of Industrial Microbiology 17:185-196.
10
Appendix E Instrumentation Design and Construction
Appendix E - Instrumentation Design and Construction
E.1.0 Subsurface Monitoring Design
E.1.1 Temperature Monitoring

The USACE decided to operate two different subsurface temperature-monitoring
systems at the Wyckoff site. One was composed of a series of thermocouples; the other
was a DTS system using fiber optic technology. This decision was based on the fact that
since the Wyckoff project was a pilot test site, it would be a good idea to test a promising
new technology, DTS. The DTS fiber optic lines were installed in 18 wells total (seven
in extraction wells and 11 in instrumentation (Geoprobe) borings), primarily in the SE
and SW corners of the project. Both DTS and thermocouples were installed in
instrumentation boring T7, located on the north side of the project, just inside the sheet
pile wall, so that values of the two systems could be directly compared. See Figure 6.2-2
for well locations and Table E1.1-1 for a complete list of instrument types and numbers
in each well or boring.
DTS. The DTS purchased was from Sensa, Inc. (Sensa), of Bakersfield,
California. Sensa began installing the DTS in about 1998 in oil wells to monitor
temperatures during steam injection and oil recovery operations. The Wyckoff site was
their first experience installing the DTS equipment at a Superfund site. The DTS is
measured by sending a pulse of laser light down the fiber optic line. Molecular vibration,
which is directly related to temperature, creates weak reflected signals that are detected in
an opto-electronic surface readout unit (controller) and converted to values of
temperature at one meter intervals along the fiber. The system components include a
OD stainless steel control line, flexible fiber optic line, controller, and a portable
computer running the operating software. Specially coated optical fiber is pumped into
the control line. The controller is capable of monitoring six control lines simultaneously.
Two controls lines were used on the site, each line strapped to the electrical supports
under the steam lines and branching off along the ground surface and installed in nine
wells or borings. The control lines were connected to the flexible fiber leads in a junction
box in the vicinity of extraction well E5. These flexible leads then ran to the north end of
the instrumentation trailer where the controller was located. The leads ran underground
between borings T73 and T78 to allow for the passage of vehicles into the site.
Thermocouples. The thermocouples used for the subsurface monitoring of
temperatures were type E. Type K would have been acceptable, but type E was selected
because it produces a higher voltage per degree output (better resolution) and is more
repeatable in thermal cycling over the temperature range. The two metals used in type E
E-1
are Constantan and Chromel. The other two standard types, J and T, were not considered
because they use iron and copper conductors respectively and are thus more susceptible
to corrosion. There was some evidence that at other monitoring projects thermocouples
had a high rate of failure. For this reason, care was taken in selecting the appropriate
insulation material for the thermocouples. The typical insulation used in thermocouple
construction is Magnesium Oxide (MgO). MgO is hydroscopic (absorbent), therefore not
an ideal insulation material to use in subsurface installations. The insulation used in the
construction of the Wyckoff thermocouples was Teflon, with Teflon jacketing, and a
stainless steel overbraid to protect the leads from physical damage. They were purchased
from Conax Buffalo Technologies.
E.1.2 Pressure Monitoring

The transducer selected for monitoring subsurface pressures was a vibrating wire
transducer manufactured by Geokon, model number 4500ALX-25X. This is the only
subsurface transducer that could be found to stand up to the high temperatures. Absolute
pressures were read (corrected for water temperature), then corrected for standard
barometric pressure to arrive at an approximate pressure or water level depth.
E.2.0 Above Ground Monitoring Design
E.2.1 Pump Stroke Counters

The pump stroke counters supplied by the contractor were analog stroke counters
which also provided a contact closure output, capable of being read by the Geomation
3300 ADAS.
E.2.2 Vapor Extraction Flows at E4

The contractor supplied a 4-20mA output differential flow meter to monitor the
vapor extraction flow rate from E4, capable of being read by the Geomation 3300 ADAS.
This flow meter was not installed until after operations began. It was activated on
December 27, 2002, but never monitored because the vapor extraction system was not
operational.
E.2.3 Total Organic Carbon of Extracted Liquid

A TOC instrument was purchased by the USACE to monitor and measure the
contaminant concentration in the extraction liquid flow stream. This instrument also
provided a 4-20mA output and was read by the Geomation 3300 ADAS.
E-2
E.3.0 Automation Systems Design
The ADAS that was selected is the Geomation System 3300 manufactured by
Geomation, Inc. of Golden, Colorado. The system was selected because of its capability
of reading all of the instrument types at the project in all weather conditions, and that it
had built-in transient (lightning) protection. The system is a secure supervisory and data
acquisition (SCADA) system which is a computer system designed to gather and analyze
real-time data. SCADA systems are used to monitor and control a plant or equipment in
industries such as telecommunications, water and waste control, energy, oil and gas
refining, and transportation. A SCADA system is centrally driven, meaning that all data
requests are initiated from a central computer.
The hardware components of the System 3300 include remote terminal units
(RTUs), input/ouput (IO) module assemblies, and a PC running the SCADA software. At
the Wyckoff site six RTUs were used, five attached on the steam line supports in various
locations monitoring all the field instruments and one in the boiler room monitoring
readings from the TOC instrument. The RTUs ran off alternating current (AC) power
distributed along the steam line and were connected by wireline to a PC located in the
instrumentation trailer adjacent to the DTS PC. The appropriate number of modules were
housed in a rainproof enclosure mounted on a 4x4 wooden post adjacent to each well
or boring. The module is the device that contains the electronics to read each type of
instrument. Four types of modules were used; thermocouple, 4-20mA, pulse counter, and
vibrating wire. As many as two thermocouples could be attached to each thermocouple
module, one 4-20mA transmitter to each 4-20mA module, two stroke counters to each
pulse counter module, and one vibrating wire instrument with thermistor to each
vibrating wire module. A single cable carrying power and communications was run from
each RTU to a number of module enclosures. The all-weather cable was laid on the
surface of the gravel pad.
The SCADA software used to run the Geomation 3300 ADAS was Intellution
iFix. Intellution iFix is an operating software normally used for data acquisition, process
visualization, and supervisory control of plant operations (wastewater, manufacturing,
utilities, etc). The driver used to communicate with the Geomation 3300 system was
Modbus. The iFix software is extremely powerful and therefore complex. It was overkill
for this application but was recommended by Geomation as the appropriate software to
run the 3300 system.
E.4.0 Data Management Processes
A data management and review process was designed for the Wyckoff project.
The goal of the process was to make all data available in plot format on a commercial
E-3
web site so that all parties concerned (USACE, USEPA, contractor, and consultants)
could review the data on a daily basis and use the data to make decisions on the operation
of the system. The following tasks had to be accomplished to meet this goal.
1. Convert the DTS data stored in bin files and load into a master database
2. Query the Geomation data stored in an iFix database and load it into the master
database with the DTS data
3. Query the master database on a daily basis to create a series of ASCII files used
for plotting and for loading into a GMS package
4. Create temperature sections and slices of the pilot site with the GMS software
5. Plot data time history and temperature profile data
6. Store the plotted data on the commercial web site
7. Store the contractors plots and data on the commercial web site
8. Incorporate a review and comment procedure (message board) accessed via a link
on the web site
9. Automate the procedure as much as possible to keep labor to less than one hour
per day
Items one through three were accomplished with a single visual basic program
which ran automatically via Windows scheduler. Grapher form Golden Software was
used to create the time history and profile plots. This process was also automated so that
the numerous plots could be created in just a couple of minutes. All data plots were
transferred to the extranet web via ftp protocol. The message board was called Simple
Message Board and was hosted on the same commercial web site at no additional cost.
E.5.0 Instrumentation Construction
E.5.1 Thermocouples and Pressure Transducers

The thermocouples and pressure transducers were attached to a 3/8 in diameter
fiberglass rod and strapped to injection and extraction wells using fiberglass channels as
spacers to keep the instruments separated from the stainless steel casing screen.
The initial seven vibrating wire pressure transducers installed in the
instrumentation borings were grouted in the formation using silica fume grout. Grouting
in vibrating wire piezometers has been a favored method used by various installers for
many years and has proven to work very well. The piezometers grouted in with the silica
flour grout, however, did not function properly. This was attributed to either intrusion of
E-4
the grout through the porous filter stone, locking the diaphragm in place or the grout was
too tight to allow for the transfer of pressure to the sensor.
Nine additional transducers were therefore installed to replace the failed
instruments. Five were located adjacent to the original piezometer borings inside the
steam line loop, since they were easily accessible. The four outside the loop were not
replaced due to access difficulties; instead an additional three were installed inside the
loop in various locations to improve the understanding of groundwater pressures in the
E4 quadrant and one was installed near instrumentation string T46 to provide better up-
gradient data on the east side of the pilot test area. The new transducers were installed
with a sand filter around the sensor then grouted up with the silica flour grout.
E.5.2 DTS
The installation of the DTS was conducted in two phases. The first phase was the
installation of the in stainless steel tubing in each of the wells or borings. The tubing
was delivered to the site in pre-designated lengths, uncoiled and strapped to 3/8 in
diameter fiberglass rods. The fiberglass rods helped to keep the tubing straight and
allowed for easier transport around the site. The rods were then strapped to the extraction
well casing, using fiberglass channel spacers to keep the tubing separated from the
stainless steel casing. A representative from Sensa was on site during the drilling of the
first two extraction wells to assist with the installation. The tubing was also installed in
eleven instrumentation borings drilled with a Geoprobe rig. The tubing strapped to the
fiberglass rods was installed inside the drill casing, once a final depth had been achieved,
and the boring was then grouted up.
The second phase of the installation was completed by two Sensa personnel after
all subsurface tubing had been installed. In this phase the Sensa personnel made in
stainless steel tubing surface connections to all 18 wells, pumped fiber into each of the
two control lines, installed the splice enclosure, spliced the tubing to the flexible fiber
cable, installed the controller and portable computer (pc) in the instrumentation trailer,
connected the flexible fiber to the controller and the controller to the pc, and programmed
the operating software on the pc. This installation phase took one week.
E.5.3 Geomation 3300 ADAS

The Geomation hardware was installed in six man-weeks by in-house labor. A
representative from Geomation was on site for one week to assist with the software
programming. The communication cable from the RTUs to the PC was run underground
alongside the flexible DTS fiber optic cable between instrumentation borings T73 and
T78.
E-5
E.5.4 USACE Involvement
Subsurface and automated instrumentation. USACE designed, procured,
installed, and operated the instrumentation system with assistance from Sensa
(two weeks) for the installation of the DTS and from Geomation (one week) for
assistance in programming the iFix software. The monitoring system was
operational at the start of the pilot test.
E.6.0 Recommendations for Future Projects
E.6.1 Automated Monitoring System
Geomation Operating Software (Intellution Fix). Due to the complexity of the

iFix software, it is recommended that simpler options be considered on future
projects. Geomation currently has one additional possibility on the market and is
developing an easier interface software package that should be available for
purchase in the summer of 2004. Neither of these options was available at the
time of purchase of the equipment for this project.
Power Outages. The Geomation RTUs were AC powered and plugged into an
adjacent GFI receptacle. The GFI would occasionally trip due to moisture
intrusion. It would therefore be advantageous to have battery backup at the RTUs
for future similar installations. A rechargeable battery is a standard option
supplied by Geomation.
DTS Operating System. Communication between the controller and the PC
would occasionally lock up, requiring someone to close and restart the software
and/or reboot the controller. The graphics on the pc monitor would also
occasionally not display properly. The display problem was solved by
reconfiguring the video mode selection for pcAnywhere to compatibility mode.
The communication problem was never solved even after swapping out the laptop
twice. This problem appeared to be related to the Windows 98 operating system.
The DTS operating software has now been upgraded to run under Windows
2000/XP, but has not been tested by Seattle District personnel.
Additionally, by comparing the temperature in instrumentation boring T7 it was
noted that temperatures read by the DTS were about three C higher than the
temperatures read by the thermocouples. To verify this manual temperatures were taken
with a portable probe in extraction well E-5 before it was sealed. The readings verified
the 3 C offset. Sensa explained that the calibration of the system could be modified to
E-6
correct this, but it was never done. The offset was constant and independent of
temperature so it was easy to correct for.
DTS versus Thermocouples (Geomation 3300). The DTS worked very well
except for the communication problem discussed above. Assuming this problem
is solved with upgraded software, DTS is an excellent option for monitoring
temperatures in future projects. The thermocouples and Geomation 3300 system
also worked very well. Although the installation was more complex, it functioned
almost perfectly. A few thermocouple modules failed (6-10) but were easily
repaired or replaced. The main advantage of the system is that it is also capable
of reading other types of instruments (i.e., 4-20mA, pulse counters, vibrating
wire). Geomation 3300 systems also have the capability of communicating via
radio to meet more complex site requirements. A combination of the two systems
would be an ideal solution for future projects; DTS to monitor temperatures and
Geomation 3300 to monitor everything else.
Silica Grout. The ITTAP panel recommended that silica flour grout be used in
areas where steam would be injected. Some of the pressure transducers (7) failed
when installed with silica grout. The use of standard grout or a sand filter should
be investigated to see if either are acceptable alternatives.
Pump Stroke Counters. The pump stroke counters on the extraction wells
manufactured by Severn Trent did not function properly. Both the rotating dial
and the digital output failed routinely.
E.6.2 Data Management
Automated data collection. This part of the project worked extremely well. Little
lag time was noticed for those monitoring systems where data were generated
automatically and managed by the USACE. Generally the daily process was
completed in about an hour. The most time consuming task was the creation of
the sections and slices in GMS.
Non-automated data collection. There was a lag time between data collection and
posting to the web page. Data such as steam injection rates and well pumping
rates were not transferred to the webpage fast enough for daily decision making,
therefore, the data were compiled every morning and reported verbally to the
team during the daily operations call. When data became available, it was posted,
making the webpage more effective as an accessible data archive, but less
effective as a real-time source of data. The manual nature of this data gathering
E-7
was inefficient and costly. Better data collection tools need to be designed for
this part of the project.
Web page. Having data and communications logs posted to the project web page
was an effective method for team communication.
E-8
Table E 1.1-1 Instrument Types and Numbers in Each Well or Boring
Well Thermocouples DTS Pressure Transducer
I-1 3
I-2 3
I-3 3
I-4 3
I-5 3
I-6 3
I-7 3
I-8 3
I-9 3
I-10 3
I-11 3
I-12 3
I-13 3
I-14 3
I-15 3
I-16 3
E-1 11 1
E-2 12 1
E-3 10 1
E-4 11 1
E-5 10 1
E-6 13 1
E-7 9 1
T-1 8
T-2 8
T-3 10
T-4 8
T-5 8
T-6 8
T-7 9 11
T-8 9
T-9 9
T-10 9
T-11 8
T-12 8 1*
T-13 9
T-14 7 1*
T-15 8
T-16 8
T-17 8
T-18 7
T-19 7
T-20 7
T-21 8
T-22 7 1*
T-23 7
E-9
Well Thermocouples DTS Pressure Transducer
T-24 7
T-25 8 1*
T-26 8
T-27 8 1*
T-28 8
T-29 8 1*
T-30 8
T-31 8
T-35 8
T-36 8
T-37 7
T-38 8
T-39 8
T-40 8 1*
T-41 7
T-42 7
T-43 9
T-44 9
T-45 7
T-46 7 1*
T-47 9
T-48 8
T-49 9
T-50 9
T-51b 9
T-52 8
T-53 8
T-55 7
T-56 7
T-57 8
T-58 7
T-63 7 1*
T-72 7
T-73 7
T-74 7
T-75 7
T-76 8
T-77 8
T-78 2
T-79 2
T-80 2
T-81 9
* Vibrating wire replacement pressure transducer located in Geoprobe boring adjacent to original
boring.
E-10
Appendix F Operations Logs (CD ROM Only)
G.1 Introduction
The expected inter-phase differentiation of NAPL components was discussed in

Section 4.1. During preparation of the Design Analysis, it was understood that aliphatic
compounds would be the most volatile fraction of the creosote mixture, and these
compounds would be recovered primarily in the vapor phase. Highly soluble compounds
that are solid at ambient conditions, such as naphthalene and other LPAHs, would be
concentrated in the aqueous phase. Table G-1 shows predicted component-group
concentrations at 100 C and atmospheric pressure, in the oil, aqueous, and gas phases.
G.1.1 Precipitation from the Aqueous Phase

The NAPL concentrations in Table F-1 represent the average of samples taken
throughout the FPA prior to the pilot study. The data show the average composition of
the creosote mixture to be 83% LPAH (35% naphthalene), with an LPAH/aliphatic ratio
of 6.4. The theoretical LPAH/aliphatic ratio is only 2.1 for the gas phase in equilibrium
with NAPL, due to the higher volatility of the aliphatic group. The ratio is much higher,
24.9, for the aqueous phase in equilibrium with NAPL, due to the higher solubility of the
LPAH group. Since the SEE process should result in combined recovery of all three
phases, the recovered product was expected to be a liquid mixture similar to the original
NAPL in the subsurface.
Steam-injection was maintained at low rates during the 5-month pilot study due to
repeated equipment failures and concerns about the capacity of older equipment in the
treatment plant (i.e. the outfall discharge pipe). The result was a treatment process
resembling slow hot-water circulation. The average steam injection and pumping rates
were 13% and 25% of design rates, respectively. The vapor-extraction system was in
operation for a total of 1 month, with no more than 3 days continuous operation.
Groundwater velocities were probably too low to hydraulically mobilize significant
NAPL, and vapor flow was not sufficient for significant mobilization of aliphatic
hydrocarbons. Aqueous-phase recovery consisted primarily of LPAH compounds,
including naphthalene, which are solid at ambient conditions. For these reasons, the
recovered product was not a creosote-like liquid mixture, as anticipated in design; instead
it was primarily dissolved LPAH, which precipitated and crystallized in wells, pipes and
in the treatment system.
G-1
G.1.2 Precipitation from the Vapor Phase
LPAH encrustation, primarily naphthalene crystals, also occurred in the vapor-
extraction system, and was seen to precipitate from steam plumes discharging from
pump-exhaust ports and open wells. These observations are consistent with data shown
in Figure G-1 for predicted naphthalene concentrations in the gas phase. Naphthalene is
classified as a semi-volatile compound, and will partition into the gas phase from NAPL
according to Raoults Law, and from naphthalene-saturated water according to Henrys
Law. The data in Figure G-1 suggest that gas in the presence of water containing
dissolved creosote components can contain even higher naphthalene concentrations than
gas associated with creosote alone. The implication is that the problem of solids
precipitation in the vapor extraction system was exacerbated by operating the pilot study
as a hot-water-circulation process rather than a balanced SEE operation.
G.2 Importance of Vapor-Phase Transport
As described in Section 4, progressive downsizing of the vapor extraction system

occurred during system design changes. The design criteria called for well head vacuum
at 0.5 atm, cooling capacity of 30% of injected enthalpy, and a maximum non-
condensable gas flow of 1,350 acfm. The original design criteria could have been
optimistic, since design modeling indicated that the peak-extracted enthalpy could
approach 100% of the injected enthalpy. In contrast to the design criteria, the as-built
system was thought to be capable of 0.25-atm vacuum at the well heads, a cooling
capacity of 10% of injected enthalpy, and non-condensable gas flow at 450 acfm. It is
likely that the 0.25-atm well head vacuum would have been insufficient to extract any
vapor during pressure-cycle shut-in periods, based on model results. As a means to
reduce construction costs, the as-built vapor extraction capacity of the system was
significantly lower than the original design capacity. In retrospect, the reduction in
capacity should have been more thoughtfully implemented considering the original
system design basis and the projected vapor flow and transport.
The data shown in Table G-1 can be extended to pressures lower than one
atmosphere, and combined with the multiphase-extraction rates assumed during pilot
study design (Section 4.1) to predict ideal contaminant mass-removal rates for each phase
(Figure G-2). Figure G-2 illustrates the critical importance of vapor-phase transport for
thermal treatment of creosote: theoretical mass-removal rates for the vapor and NAPL
phases are equal at 0.25 atm, and vapor-phase removal is higher at vacuum levels above
0.25 atm. Predicted mass-removal in the aqueous phase is minor compared to the other
phases.
G-2
Both gas-phase and aqueous-phase removal shown in Figure G-2 are assumed to be at
equilibrium with the NAPL phase, and may be over-estimated relative to actual field
conditions. On the other hand, the projected NAPL-removal rate includes considerable
vapor-transported product that condenses in the cooling and treatment system, and in and
around the extraction wells. These data indicate that a primary objective for thermal
remediation of creosote should be aggressive steam flushing throughout the treatment
area, and that hot-water circulation without significant hydraulic mobilization of NAPL is
not an effective process.
Table G-1 Predicted Multiphase Component Concentrations
Concentration of Component Group (g/g)1

LPAH/aliphatic
Phase
ratio
LPAH HPAH Aliphatic
NAPL2 0.8277 0.042976 0.12900 6.42
Gas3 0.0535 0.000333 0.02591 2.06
Aqueous 0.0108 0.000002 0.00044 24.86
Notes:
1. All values are estimated for 100 deg. C and atmospheric pressure, with all phases
in equilibrium.
2. Mass fractions are based on an average of all NAPL samples taken to date.
3. Gas phase is assumed to be primarily air.
G-3
Figure G-1. Theoretical Naphthalene Concentrations in Gas Phase
aqueous - steam aqueous - air
oil - steam oil - air

Concentration in gas phase (g/g)
0.18
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.00
50000 60000 70000 80000 90000 100000

Pressure (Pa)
G-4
Figure G-2. Theoretical Contaminant Mass Removal Rates
NAPL gas (air)

gas (steam) aqueous
2000
Removal Rate (kg/d)
1000
0
0.00 0.10 0.20 0.30 0.40 0.50
Vacuum Pressure (atm)
G-5
G-6
Appendix H - Calculation Methods for Process Streams
Appendix H Calculation Methods for Process Streams
Note: Discrepancies exist between some fundamental assumptions (pilot area, pilot volume, soil
porosity) used in Appendix H and those used in the body of the text of this report, as well as in previous
design documents. There was not an opportunity to correct the differences prior to publication of this
report.
H.1 Mass Balance Estimation Methods
The water mass balance is calculated as follows:

Min,steam = Mout, liquid + Mout,vapor - Mnet extraction
The steam injection rate was estimated for each of the 16 injection wells
Min,steam = (min, steam x t)
where min, steam is the flow rate, and t is time. Each value of m was calculated
based on a valve setting, the pressure drop across the valve, and the temperature of the
steam. Average values for each day of injection were used.
Since the data from the steam regulator valves was not recorded for the entire
period of steam injection, two additional lines of data were used to document the actual
steam injection rates from early October to May:
1. The diesel usage was converted to an equivalent steam production rate, assuming
a boiler efficiency of 85%, and a diesel energy content of 140,000 BTU/gal.
2. The water used by the boiler was converted to equivalent steam injection rates,
assuming negligible blow-down volumes.
The mass removal in the liquid for is a simple summation of the measurements
from each of the seven extraction pumps:
Mout, liquid = (mliquid x t)
Where the values for mliquid were derived from a pump stroke counter installed for
each pump.
The mass removal in the form of vapor (steam, water vapor) ideally is calculated
by the liquid production rate in the condenser:
Mout,vapor = (mcondensate x t)
Where mcondensate is the flow rate of condensate. However, the flow meter that
should have quantified this stream was never installed. Instead, we estimated condensate
H-1
rates based on observed temperatures, and experience from other steam sites (Alameda
Point, Young-Rainey STAR Center Area A).
The net extraction was estimated based on the water balance equation presented in
the beginning of this section. Similarly, net water extraction rates were estimated by the
difference of the measured flows:
mnet extraction = mliquid + mvapor - min,steam
Due to the absence of a flow meter for the produced condensate, Mvapor and mvapor
had to be estimated.
H.2 Energy Balance Estimation Methods
Cumulative energy (E) is calculated as a summation of enthalpy fluxes (Q):

E = (Q x t)
An estimated energy balance will be maintained for the site.
Ein, steam = Eout + Estorage + Eloss
The energy fluxes are related for each time step as follows:
Qin, steam = Qout + Qstorage + Qloss
Where Q denotes enthalpy flux (in BTU/hr). Energy increments are estimated as
follows:
Qin, steam = min,steam x Hsteam-ambient
Where m is mass of steam. This calculation was done for each of the 16 injection
wells, and average daily values were used for the steam flow rates. The enthalpy of the
steam was calculated from steam tables, using a steam pressure of 20 psig, and an
ambient temperature of 15 oC:
Hsteam-ambient = (1,167 25) BTU/lb = 1,142 BTU/lb
The following energy fluxes were calculated for each of the seven extraction
wells:
Qliq = mwater x cp, water x (T T0)
Ideally, for the extracted fluid stream, the energy flux in vapor and steam should
be estimated based on treatment system data:
Qnon cond. gas = mair x cp, air x (T T0)
H-2
Qsteam out = mcondensate x Hsteam-ambient
Where m is mass, H is specific enthalpy (in BTU/lb), cp is heat capacity (in
BTU/lb/F), and T is temperature. Since very limited and unreliable vapor flow data is
available for periods other than November 26, 2002 through December 16, 2002, an
assumed average flow rate of 100 scfm was chosen based on the average rate observed
during this period for the vacuum system. After December 16, when the vapor extraction
system was turned off, zeroes were inserted. The calculation shows that the actual vapor
flow rate has a very small effect on and energy balance, due to the low heat capacity of
air. Similarly, the rate of condensate production was estimated based on observed
temperatures and experience from other sites.
The total energy removal from the pilot test volume was estimated as follows:
Qout, total = Qliq + Qnon cond. gas + Qsteam out + Qheat loss
The actual heat loss cannot be calculated using accurate measures. An estimate
can be made based on thermal profiles at the bottom and top of the treatment cell, and
along the sheet-pile wall, using the following calculations:
Qheat loss = A x KT x dT/dz
Where A is the surface area through which energy is conducted, KT is the thermal
conductivity of the subsurface material (saturated sand/clay near the aquitard, partially
saturated sand near the vapor cap), and dT/dz is the temperature gradient. No heat loss
was calculated for the southern boundary, since inflow of water due to the net extraction
would carry the conducted energy back into the pilot test area.
For the loss through the vapor cap, the temperature difference between the two
uppermost temperature sensors were used to calculate the gradient dT/dz. The area of the
heated zone was estimated at 27,500 ft2, which is slightly larger than the foot print of the
wells due to the steam migration south of the pilot test wells. For the heat loss through
the bottom of the site, the temperature gradient was estimated based on the bottom two
sensors in the monitoring locations, and the steam wells. The heat loss through the sheet-
pile wall was estimated based on average temperatures on the inside of the sheet pile wall
(using daily values for all the sensors), and assuming that near ambient temperatures
exists five ft away from these sensors. This is a relatively rough assumption, but no data
exists for an improved estimate.
Thermal conductivities of 2.5 W/mK were used for the heat loss calculations
through the bottom and sides, which is based on saturated soils. It was assumed that
saturated conditions exist along the boundaries of the site, due to steam condensing as it
cools. For the vapor cap heat loss, a value of 1.5 W/mK was used, corresponding to
partially saturated soil.
H-3
The temperatures achieved and measured using the temperature sensors were
compared to the temperatures estimated based on the calculated energy balance. The
stored energy is related to the pilot test heat capacity, and the measured average
temperature as follows:
Estorage = Cp,site x (Tavg - T0) + msteam x Hsteam-ambient
Where Cp,site is the overall heat capacity of the pilot test area, estimated from the
volume, saturation, and specific heat capacity of the soil and water:
Cp,site = Vsoil x cp, soil x Vwater x cp, water
The steam energy stored as a vapor at any given time is relatively small, and was
neglected in the calculations. For comparison with the measured temperatures, the
energy balance was used to estimate the average temperature (Tenergybal)of the pilot test
volume:
Tenergybal = T0 + Estorage/Cp,site = T0 + (Ein, steam - Eout - Eloss)/ Cp,site
Where T0 is set as the average background temperature (57 oF = 13.9 oC).
H.3 Pore Volume Calculation Methods
The number of pore volumes of steam injected has been used as a measure at
other sites, and during laboratory scale treatability studies. For this pilot test, the pilot
test pore volume was estimated as follows:
pvpilot test = * Vpilot test
Where is the average porosity, and V is the volume. For the injected volume, a
direct calculation from the steam flow rate measurements and the energy balance is made,
yielding the amount of water that entered the site as steam:
Vsteam = Min,steam * water = water * (min, steam x t)
Where water is the density of water.
The number of pore volumes injected is then derived from
pvsteam = Vsteam / pvpilot test
The vadose zone is included in this estimate, since the steam is intended to heat
both the saturated and vadose zones.
H-4
Appendix I Pilot Study Data Summary (CD-ROM ONLY)
Appendix J Installation Report for Well 02CD-MW01
CENWS-EC-TB-GE 26 September 2002
BAILEY/6682
MEMORANDUM FOR RECORD
SUBJECT: Recommendation for Additional Monitoring Well at the Wyckoff Superfund Site,
1. Recent acquisition of continuous water level data in extraction well E-6 has reinforced
interpretations and conclusions presented in a report titled Well E4 Pumping Test Summary by
Mick Easterly, dated 16 Nov 01. In that report strong tidal influence was noted for wells E-5 and
E-6. The report concluded that the aquitard between the upper and lower aquifers may be thin or
absent near E-6 and that vertical hydraulic control should be a goal during remediation
operations. Data collected recently from well E-6, in particular, have exhibited clear influence
from tides. The effect is not nearly as apparent for E-5.
2. Under current conditions, the upper aquifer beneath the site in general and the Pilot Area
specifically has been isolated from water level fluctuations in Eagle Harbor by the presence of
two sheet pile walls. Consequently, tidal effects should be minimal to nonexistent in the upper
aquifer. The presence of tidally influenced fluctuations in E-6 implies hydraulic connection with
the lower aquifer. The connection is apparently limited in areal extent, because other wells in
the Pilot Area exhibit minimal cyclic fluctuations.
3. The closest boring (for monitoring well 99CD-MW04) that penetrates through the glacial till
aquitard beneath the upper aquifer is approximately 80 feet to the northwest of E-6. The
aquitard at that location consists of roughly 20 feet of till with sand interbeds. The occurrence of
sand interbeds provides one mechanism for hydraulic connection between the upper and lower
aquifers, depending on the degree of interfingering of the interbeds. In the absence of direct
evidence at E-6, it is impossible to know the actual nature of the connection between the two
aquifers.
4. To evaluate existing conditions in the aquitard and lower aquifer near E-6, it is recommended
that one monitoring well be installed in the lower aquifer to a maximum depth of approximately
70 feet below ground surface. Due to imminent commencement of thermal remediation
activities in the Pilot Area, the well will have to be located outside the sheet pile wall. The
chosen location should be as close to the sheet pile wall as practical, taking into account safety
concerns resulting from the proximity to high pressure steam conveyance lines. It is further
recommended that continuous soil sampling (by split-spoon or other equivalent methods) start at
28 feet below ground surface, which should be 5 feet above the glacial till aquitard.
5. The action recommended in this memorandum is considered necessary to document the

presence or absence of site contaminants (including non-aqueous phase liquids) in the aquitard
and lower aquifer in light of observations that suggest some degree of hydraulic communication
between the upper and lower aquifers near E-6. One of the goals of the Pilot project is avoid
exacerbating contaminant conditions in the aquitard or lower aquifer. This can be accomplished
CENWS-EC-TB-GE
SUBJECT: Recommendation for Additional Monitoring Well at the Wyckoff Superfund Site,
only by fully understanding current conditions near E-6 and by periodically monitoring to ensure
that existing conditions are not made worse during Pilot operations.
6. If you have any questions about this memorandum, please feel free to contact the undersigned
at 206-764-6682.
Michael M. Bailey
Geologist

Pilot Summary Report v.3.0 October 2006

Uploaded by

Copyright:

Available Formats

Pilot Summary Report v.3.0 October 2006

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pilot Summary Report v.3.0 October 2006

Uploaded by

Copyright:

Available Formats

Wyckoff/Eagle Harbor Superfund Site

Soil and Groundwater Operable Units

Section Page No.

Bids for a combined construction and operations contract were considerably

There were equipment constraints that limited operations of the system.

In 1998, USEPA was asked to evaluate in-situ thermal technologies, including

The USEPA identified in-situ thermal technology (steam injection) as a potential

1. To assess the likelihood that full-scale thermal remediation will achieve

2. To provide information for implementation of the potential full-scale

The report is divided into the following sections:

1.1 STEAM-ENHANCED EXTRACTION TECHNOLOGY DESCRIPTION

Increased vaporization and extraction in the vapor phase.

Increased solubilization of contaminants with subsequent removal in the dissolved

In addition, in-situ destruction due to either chemical or biological mechanisms

A physical characteristic of steam injection processes is steam override due to

1.2 PROJECT OBJECTIVES

Performance Assessment Objectives

Demonstrate that thermal remediation technologies will remove substantially

Community and Environmental Impacts of Full-Scale Thermal Remediation

Evaluate operational approaches to thermal remediation that may impact the

Additionally, an objective to evaluate the effectiveness of biological treatment

1.3 PILOT STUDY PLANNING DOCUMENTS

Interim Record of Decision (Interim ROD). Groundwater Operable Unit.

Feasibility Study (FS). Soil and Groundwater Operable Units. CH2M

Focused Feasibility Study for Thermal Remediation Technologies (FFS).

Conceptual Design (CD). Soil and Groundwater Operable Units.

Record of Decision (ROD). Soil and Groundwater Operable Units.

Thermal Remediation Pilot Test Area Selection Memorandum.

Comprehensive Report. Wyckoff NAPL Field Exploration. Soil and

Thermal Remediation Shoreline Model Report. Groundwater and Soil

Final Design Analysis (DA). Thermal Remediation Pilot Study. USACE.

Final Design Analysis Amendment (DAA). Thermal Remediation Pilot

Preliminary Review Draft Baseline Characterization Report. Thermal

Draft Remedial Action Management Plan (RAMP). Thermal Remediation

1.4 INITIAL CONSTRUCTION

The solicitation to contractors was issued in the summer of 2001, however, an

1.5 PILOT STUDY DOWNSIZING

2.1 SITE HISTORY

Unconfined upper aquifer

Semi-confined lower aquifer

2.3 GROUNDWATER TREATMENT SUMMARY

In 1988 the Wyckoff Company was ordered by USEPA to install groundwater

The GWTP accommodates a maximum practical flow rate of 80 gpm. On

3.2 PILOT TEST AREA GEOLOGY AND HYDROGEOLOGY

3.3 CONTAMINANT DISTRIBUTION IN SOIL

3.4 MICROBIAL POPULATION EVALUATION

In order to evaluate microbial populations before steam injection, soil samples

In order to determine microbial community structure, which can be defined as the

3.5.1 Pilot Test Area Extraction Wells

Among PAHs, naphthalene is present at the highest concentration, as would be

3.5.2 Upper Aquifer Monitoring Wells

Upper aquifer monitoring was initiated to document conditions (for completeness)

3.5.3 Lower Aquifer Monitoring Wells

Sampling of lower aquifer monitoring wells was intended to document any

Groundwater samples were collected from upper aquifer extraction and

Three construction contractors were responsible for installation of the major

4.1 DESIGN PROCESS

o Conceptual Design. Soil and Groundwater Operable Units. 1999.

o Thermal Remediation Pilot Test Area Selection Memorandum and Amendment.