REMEDIATION SYSTEM EVALUATION
FORMER OCCIDENTAL FACILITY
TACOMA, WASHINGTON
Report of the Remediation System Evaluation,
Site Visit Conducted at the Former Occidental Facility, Tacoma, Washington
July 8, 2003
-------
This page is intentionally left blank.
-------
Office of Solid Waste EPA 542-F-04-026
and Emergency Response August 2004
(5102G) www.epa.gov/tio
clu-in.org/optimization
Remediation System Evaluation
Former Occidental Facility
Tacoma, Washington
-------
This page is intentionally left blank.
-------
NOTICE AND DISCLAIMER
The U.S. Environmental Protection Agency (U.S. EPA) funded the work described herein and
preparation of this document by GeoTrans, Inc. under General Service Administration contract
GS06T02BND0723 to S&K Technologies Inc., Bremerton, Washington and under EPA contract
68-C-02-092 to Dynamac Corporation, Ada, Oklahoma.
This report has undergone review by the EPA site managers and EPA headquarters staff. For more
infomation about this project, contact: Mike Fitzpatrick (703-308-8411 orfitzpatrick.mike@epa.gov) or
Kathy Yager (617-918-8362 or yager.kathleen@epa.gov).
-------
EXECUTIVE SUMMARY
A Remediation System Evaluation (RSE) involves a team of expert hydrogeologists and engineers,
independent of the site, conducting a third-party evaluation of site operations. It is a broad evaluation
that considers the goals of the remedy, site conceptual model, above-ground and subsurface performance,
effectiveness in protecting human health and the environment, and site exit strategy. The evaluation
includes reviewing site documents, visiting the site for up to 1.5 days, and compiling a report that
includes recommendations to improve the system. Recommendations, including estimates of resulting
net cost impacts, are provided in the following four categories:
improvements in remedy effectiveness in protecting human health and the environment
reductions in operation and maintenance costs
technical improvements
gaining site closeout
The recommendations are intended to help the site team (the responsible party and the regulators)
identify opportunities for improvements. In many cases, further analysis of a recommendation, beyond
that provided in this report, might be needed prior to implementation of the recommendation. Note that
the recommendations are based on an independent evaluation by the RSE team, and represent the
opinions of the RSE team. These recommendations do not constitute requirements for future action, but
rather are provided for the consideration of all site stakeholders. This RSE report pertains to conditions
that existed at the time of the RSE site visit, and any site activities that have occurred subsequent to the
RSE site visit are not reflected in this RSE report (unless otherwise noted).
The former Occidental Chemical Corporation (OCC) property ("site") is approximately 37 acres in extent
and is located at 605 Alexander Avenue in Tacoma, Washington along the Hylebos Waterway. Between
1947 and 1973, OCC operated a chlorinated solvents plant. OCC transferred the facility to Pioneer
Americas, Inc. in 1997 but retained the environmental liability. A portion of the site was also leased to
the Navy in the 1940s and 1950s as part of their ship building and dismantling operations. The facility is
currently used as a terminal for shipping and storage of chemicals, but chemical production operations
are idle. Remediation oversight was provided by the EPA RCRA program through 1998 when oversight
was transferred to the Washington State Department of Ecology ("Ecology"). At the time of the RSE site
visit, Ecology and OCC were negotiating a scope of work to be incorporated into a Model Toxics Control
Act (MTCA) agreed order, which will become part of a new RCRA permit. In addition to the oversight
provided by Ecology, the EPA RCRA program remains involved, particularly in regard to determining
the attainment of the RCRA Environmental Indicators (Els). The EPA Superfund Program also has a
vested interest in the site. It is overseeing remediation to address contaminated sediments within the
Mouth of the Hylebos Waterway Problem Area of the Commencement Bay/Nearshore Tideflats
Superfund Site. This includes two non-time critical removal actions along with the remedial action of
dredging the Hylebos channel to the native sediment interface.
Site contamination predominantly consists of volatile organic compounds (VOCs) and elevated pH in the
ground water. The primary VOCs are tetrachloroethylene (PCE), trichloroethylene (TCE), and
breakdown products including cis-l,2-dichloroethylene (cis-l,2-DCE) and vinyl chloride. A pump and
treat (P&T) system has operated at the site since 1996, extracting contaminated ground water, treating it,
reinjecting a portion of the treated water to the subsurface, and discharging the remaining treated water to
the Hylebos Waterway. The RSE at this site focuses on this P&T system but considers other aspects of
the site as they relate to this system.
-------
The observations and recommendations contained in this report are not intended to imply a deficiency in
the work of either the system designers or operators but are offered as constructive suggestions in the
best interest of the EPA, the public, and the facility. These recommendations have the obvious benefit of
being formulated based upon operational data unavailable to the original designers.
Recommendations are provided in all four categories: effectiveness, cost reduction, technical
improvement, and site closeout. The recommendations to improve effectiveness include the following:
Plume capture should be enhanced by abandoning the current injection system and increasing
extraction throughout the site, where practicable.
Evaluation of plume capture through the use of potentiometric surface maps developed from site-
wide water level measurements and tidal corrections should be discontinued and replaced with
more cost-effective and reliable approaches.
Evaluation of plume capture should be improved by conducting regular water budget analyses in
which the amount of water recharging to the site from precipitation, underlying aquifers, and
other sources is compared with the amount of ground water extracted.
Evaluation of plume capture and ground water flow should be improved by installing additional
monitoring wells to establish well pairs that surround the VOC plume. During each gauging
event, the water levels from the two monitoring wells in a well pair should be gauged
simultaneously to avoid the need to correct for tidal influence. In addition, the measurements
should be made at least four evenly spaced times per day. This will allow the site team to
determine the direction of ground water flow at each well pair location at various parts of the
tidal cycle. This approach should provide more reliable information regarding capture and
should eliminate the need to collect 72-hour continuous water levels at 28 locations, which has
required substantial effort and cost and has not added substantially to the understanding of
capture at the site.
Additional monitoring wells are needed to delineate the VOC plume to the north approximately
100 feet below ground surface. The RSE team estimates that approximately four wells to this
depth will be needed.
Institutional controls should be considered to notify current and future property owners of the
contamination beneath their properties. In addition, a preliminary look at on-site and off-site
building information is merited to determine the potential for vapor intrusion.
Recommendations for cost reduction include the following:
Adding a second air stripper in series will provide more thorough and reliable mass removal,
which will eliminate the need for GAC, reduce the sludge handling requirements, and reduce the
process monitoring frequency. Implementing this recommendation may require a capital expense
of $150,000, but annual savings of over $160,000 per year are likely.
The ground water monitoring program can be revised to more clearly focus on plume capture.
The current program addresses both capture and progress toward cleanup; however, cleanup at
this site will likely take several decades and frequent ground water monitoring in the plume
interior does not provide beneficial information for the expense. Implementing this
11
-------
recommendation may provide annual savings of approximately $75,000 per year in sampling
costs and may reduce analytical costs associated with ground water monitoring by up to 40%.
The remaining recommendations pertain to technical improvement and site closeout. Notable
recommendations include improving the performance reports and continuing with P&T as the remedy.
The RSE team provides its reasons for continuing with P&T and increasing extraction rather
implementing alternate remediation technologies.
A table summarizing the recommendations, including estimated costs and/or savings associated with
those recommendations, is presented in Section 7.0 of this report.
in
-------
PREFACE
This report was prepared as part of a pilot project conducted by the U.S. EPA Office of Solid Waste and
Office of Superfund Remediation and Technology Innovation (OSRTI). The objective of this project is
to conduct Remediation System Evaluations (RSEs) of pump and treat systems under the Resource
Conservation and Recovery Act. The following organizations are implementing this project.
Organization
Key Contact
Contact Information
U.S. EPA Office of Solid Waste
(U.S. EPA OSW)
Mike Fitzpatrick
5303W
USEPA Headquarters
Ariel Rios Building 1200
Pennsylvania Avenue, N. W.
Washington, DC 20460
phone: 703-308-8411
fitzpatrick.mike@epa.gov
U.S. EPA Office of Superfund
Remediation and Technology
Innovation
(U.S. EPA OSRTI)
Kathy Yager
11 Technology Drive (ECA/OEME)
North Chelmsford, MA 01863
phone: 617-918-8362
fax: 617-918-8427
yager.kathleen@epa.gov
U.S. EPA Office of Superfund
Remediation and Technology
Innovation
(U.S. EPA OSRTI)
Ellen Rubin
EPA Headquarters 5102G
Ariel Rios Building
1200 Pennsylvania Avenue, N. W.
Washington, DC 20460
phone: 703-603-0141
rubin.ellen@epa.gov
Dynamac Corporation
(Contractor to U.S. EPA)
Daniel F. Pope
Dynamac Corporation
3601 Oakridge Boulevard
Ada, OK 74820
phone: 580-436-5740
fax: 580-436-6496
dpope@dynamac.com
GeoTrans, Inc.
(Contractor to Dynamac)
Doug Sutton
GeoTrans, Inc.
2 Paragon Way
Freehold, NJ 07728
phone: 732-409-0344
fax: 732-409-3020
dsutton@geotransinc.com
IV
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY i
PREFACE iv
TABLE OF CONTENTS v
1.0 INTRODUCTION 1
.1 PURPOSE 1
.2 TEAM COMPOSITION 2
.3 DOCUMENTS REVIEWED 2
.4 PERSONS CONTACTED 3
.5 SITE LOCATION, HISTORY, AND CHARACTERISTICS 3
1.5.1 LOCATION 3
1.5.2 POTENTIAL SOURCES 4
1.5.3 HYDROGEOLOGIC SETTING 5
1.5.4 POTENTIAL RECEPTORS 5
1.5.5 DESCRIPTION OF GROUND WATERPLUME 6
2.0 SYSTEM DESCRIPTION 7
2.1 SYSTEM OVERVIEW 7
2.2 EXTRACTION AND INJECTION SYSTEMS 7
2.3 TREATMENT SYSTEM 8
2.4 MONITORING PROGRAM 9
3.0 SYSTEM OBJECTIVES, PERFORMANCE AND CLOSURE CRITERIA 10
3.1 CURRENT SYSTEM OBJECTIVES AND CLOSURE CRITERIA 10
3.2 TREATMENT PLANT DISCHARGE CRITERIA 11
4.0 FINDINGS AND OBSERVATIONS FROM THE RSE SITE VISIT 13
4.1 FINDINGS 13
4.2 SUBSURFACE PERFORMANCE AND RESPONSE 13
4.2.1 WATER LEVELS 13
4.2.2 CAPTURE ZONES 14
4.2.3 CONTAMINANT LEVELS 15
4.3 COMPONENT PERFORMANCE 16
4.3.1 EXTRACTION SYSTEM WELLS AND VACUUM PUMP 16
4.3.2 AIR STRIPPER 17
4.3.3 GAC 17
4.3.4 FILTER PRESS AND SLUDGE DRYER 17
4.3.5 CATALYTIC OXIDIZER 18
4.3.6 INJECTION SYSTEM 18
4.3.7 CONTROL SYSTEM 18
4.4 COMPONENTS OR PROCESSES THAT ACCOUNT FOR MAJORITY OF MONTHLY COSTS 18
4.4.1 UTILITIES 19
4.4.2 NON-UTILITY CONSUMABLES 19
4.4.3 LABOR 19
4.4.4 CHEMICAL ANALYSIS 19
4.5 RECURRING PROBLEMS OR ISSUES 19
4.6 REGULATORY COMPLIANCE 20
4.7 TREATMENT PROCESS EXCURSIONS AND UPSETS, ACCIDENTAL CONTAMINANT/REAGENT RELEASES .... 20
4.8 SAFETY RECORD 20
-------
5.0 EFFECTIVENESS OF THE SYSTEM TO PROTECT HUMAN HEALTH AND THE ENVIRONMENT . 21
5.1 GROUND WATER 21
5.2 SURFACE WATER 21
5.3 AIR 21
5.4 SOILS 21
5.5 WETLANDS AND SEDIMENTS 22
6.0 RECOMMENDATIONS 23
6.1 RECOMMENDATIONS TO IMPROVE EFFECTIVENESS 23
6.1.1 ELIMINATE INJECTION SYSTEM AND USE EXTRACTION TO ENHANCE CAPTURE 24
6.1.2 DISCONTINUE CURRENT APPROACH OF USING POTENTIOMETRIC SURFACE MAPS TO EVALUATE
CAPTURE 26
6.1.3 CONDUCT AND REPORT A WATER BUDGET ANALYSIS 27
6.1.4 INSTALL ADDITIONAL MONITORING WELLS AND USE PAIRS OF MONITORING WELLS TO
DEMONSTRATE CAPTURE WITH INWARD GRADIENTS 27
6.1.5 DELINEATE THE VOC PLUME TO THE NORTH AT THE 100-FOOT DEPTH INTERVAL 29
6.1.6 CONSIDER INSTITUTIONAL CONTROLS AND A PRELIMINARY VAPOR EVALUATION 29
6.2 RECOMMENDATIONS TO REDUCE COSTS 30
6.2.1 ADD A SECOND AIR STRIPPER IN SERIES AND ELIMINATE THE USE OF GAC 30
6.2.2 MODIFY GROUND WATER MONITORING PROGRAM 31
6.3 MODIFICATIONS INTENDED FOR TECHNICAL IMPROVEMENT 32
6.3.1 IMPROVE QUARTERLY AND ANNUAL PERFORMANCE REPORTS 32
6.3.2 CONSIDER ABANDONING AND NOT REPLACING THE SLUDGE DRYER WHEN IT FAILS 33
6.4 CONSIDERATIONS FOR GAINING SITE CLOSE OUT 33
6.4.1 CONSIDERATIONS FOR AN EXIT STRATEGY 33
6.4.2 ADDRESS PLUME AND SOURCE CONTROL PRIOR TO INSTALLING A SEDIMENT CAP 34
7.0 SUMMARY 36
List of Tables
Table 7-1.
List of Figures
Figure 1-1.
Figure 1-2.
Figure 1-3.
Figure 1-4.
Figure 1-5.
Figure 1-6.
Figure 4-1.
Cost Summary Table
Site Location Map
The Former OCC Facility and Current Well Locations
Plant East/West Geologic Cross-Section
Extent of Contamination at 25 Feet Below Ground Surface - August 2002
Extent of Contamination at 50 Feet Below Ground Surface - August 2002
Extent of Contamination at 100 Feet Below Ground Surface - August 2002
Example Potentiometric Surface Map for the 25-foot Interval (January 2002) Submitted as Part of
an Annual Performance Evaluation Report
VI
-------
1.0 INTRODUCTION
1.1 PURPOSE
During fiscal years 2000, 2001, and 2002 Remediation System Evaluations (RSEs) were conducted at 24
Fund-lead pump and treat (P&T) sites (i.e., those sites with pump and treat systems funded and managed
by Superfund and the States). Due to the opportunities for system optimization that arose from those
RSEs, EPA OSRTI and OSW are performing a pilot study of conducting RSEs at RCRA sites. During
fiscal year 2003, RSEs at up to five RCRA sites are planned in an effort to evaluate the effectiveness of
this optimization tool for this class of sites. GeoTrans, Inc. is conducting these evaluations, and
representatives from EPA OSW and OSRTI, attend most of the RSEs as observers.
The Remediation System Evaluation (RSE) process was developed by the U.S. Army Corps of Engineers
(USAGE) and is documented on the following website:
http://www.envi ronmental.usace. army, mil/1 ibrary/guide/rsechk/rsechk. html
An RSE involves a team of expert hydrogeologists and engineers, independent of the site, conducting a
third-party evaluation of site operations. It is a broad evaluation that considers the goals of the remedy,
site conceptual model, above-ground and subsurface performance, effectiveness in protecting human
health and the environment, and site exit strategy. The evaluation includes reviewing site documents,
visiting the site for 1 to 1.5 days, and compiling a report that includes recommendations to improve the
system. Recommendations with cost and cost savings estimates are provided in the following four
categories:
improvements in remedy effectiveness in protecting human health and the environment
reductions in operation and maintenance costs
technical improvements
gaining site closeout
The recommendations are intended to help the site team (the responsible party and the regulators)
identify opportunities for improvements. In many cases, further analysis of a recommendation, beyond
that provided in this report, might be needed prior to implementation of the recommendation. Note that
the recommendations are based on an independent evaluation by the RSE team, and represent the
opinions of the RSE team. These recommendations do not constitute requirements for future action, but
rather are provided for the consideration of all site stakeholders. This RSE report pertains to conditions
that existed at the time of the RSE site visit, and any site activities that have occurred subsequent to the
RSE site visit are not reflected in this RSE report (unless otherwise noted).
The former Occidental Tacoma, Washington facility was selected by EPA OSW based on progress made
toward Environmental Indicators and input from EPA Region 10. This report provides a brief
background on the site and current operations, a summary of the observations made during a site visit,
and recommendations for changes and additional studies. The cost impacts of the recommendations are
also discussed.
-------
1.2
TEAM COMPOSITION
The team conducting the RSE consisted of the following individuals:
Rob Greenwald, Hydrogeologist, GeoTrans, Inc.
Peter Rich, Civil and Environmental Engineer, GeoTrans, Inc.
Doug Sutton, Water Resources Engineer, GeoTrans, Inc.
1.3
DOCUMENTS REVIEWED
Author
CRA
CRA
CRA
CRA
CRA
CRA
CRA
CRA
CRA
CRA
CRA
CRA
OCC Tacoma, Inc., and
Pioneer Chlor Alkali
Company, Inc.
CRA
CRA
CRA
CRA
Date
3/1989
2/24/1993
11/29/1993
10/28/1994
3/23/1995
7/21/1995
8/1995
12/11/1995
2/26/1996
3/24/1997
5/6/1997
11/5/1996
5/20/1998
12/21/2000
12/21/2000
3/16/2001
11/29/2001
Title
RCRA Facility Investigation I, Volume 1
Estimation of Chemical Migration Rate
Engineering Report, Groundwater Remediation
Preliminary Extraction System Layout and
Operating Design
Preliminary Design Model Development Report
Conceptual Design, Proposed Groundwater
Extraction and Injection Systems
Final Report, Extraction System Pilot Testing
Program
Injection Testing
PRI Source Identification Program Report
pH Data Summary
Correction of Water Levels for Density Effects
Final Construction and System Startup Report
Renewal Application for Groundwater Treatment
Facility Corrective Action
Annual Performance Evaluation Report, October
1999 through September 2000
Bi-Annual Hydraulic Conductivity Assessment -
2000
January 2001 Well Redevelopment Report
Annual Performance Evaluation Report, October
2000 through September 2001
-------
Author
CRA
CRA
CRA
CRA
CRA
CRA
Tetra Tech EM Inc.
Date
4/1/2002
7/5/2002
12/20/2002
2/12/2003
4/4/2003
4/17/2003
6/25/2003
Title
January 2002 Monitoring Event and System
Performance Evaluation Report
May 2002 Monitoring Event and System
Performance Evaluation Report
Annual Performance Evaluation Report, October
2001 through September 2002
Draft Rapid Assessment Report, Embankment Area
Rapid Groundwater pH Assessment
February 2003 Monitoring Evaluation Event and
System Performance Evaluation Report
Post-Treatment Work Plan, Area 5 106 Removal
Site Characterization Report, Former Occidental
Chemical Corporation, Tacoma, Washington
1.4
PERSONS CONTACTED
The following individuals associated with the site were present for the visit:
Howard Orlean, EPA Region 10 (RCRA)
Jonathan Williams, EPA Region 10 (Superfund)
Leon Wilhelm, Washington State Department of Ecology
Stan Leja, Washington State Department of Ecology
Maury Wassmann, Glenn Springs/Occidental
Al Meek, Glenn Springs/Occidental
Jeff Brown, Pioneer
Sam Chamberlain, Pioneer
Larry McGaughey, Geomatrix
Carol Barren, Conestoga Rovers and Associates (CRA)
Frank Rovers, Conestoga Rovers and Associates (CRA)
1.5
1.5.1
SITE LOCATION, HISTORY, AND CHARACTERISTICS
LOCATION
The former Occidental Chemical Corporation (OCC) property ("site") is approximately 37 acres in extent
and is located at 605 Alexander Avenue in Tacoma, Washington along the Hylebos Waterway. Between
1947 and 1973, OCC operated a chlorinated solvents plant. OCC transferred the site to Pioneer
Americas, Inc. in 1997 but retained the environmental liability. Figure 1-1 depicts the site and the
surrounding area, including the Hylebos and Blair Waterways. For simplicity and consistency, this
report, including the figures, adopts a "plant north" convention used in other site documents. Based on
-------
this "plant north", the site is bordered to the east by the Hylebos Waterway, which is part of
Commencement Bay. The site is bordered to the north by Port of Tacoma property, to the south by Port
of Tacoma and Department of the Navy property and to the west by Alexander Avenue. In addition, a
portion of the site was also leased to the Navy in the 1940s and 1950s as part of their ship building and
dismantling operations. Across Alexander Avenue is additional Port of Tacoma property and the Blair
Waterway. The facility is currently used as a terminal for shipping and storage of chemicals, and
chemical production operations are idle. Figure 1-2 depicts the infrastructure at the site, the well
networks, and other select items.
Remediation oversight was provided by the EPA RCRA program through 1998 when oversight was
transferred to the Washington State Department of Ecology ("Ecology"). Ecology and OCC are currently
negotiating a scope of work to be incorporated into a Model Toxics Control Act (MTCA) agreed order,
which will become part of a new RCRA permit. In addition to the oversight provided by Ecology, the
EPA RCRA program remains involved, particularly in regard to determining the attainment of the RCRA
Environmental Indicators (Els). The EPA Superfund Program also has a vested interest in the site. It is
overseeing remediation to address contaminated sediments within the Mouth of the Hylebos Waterway
Problem Area of the Commencement Bay/Nearshore Tideflats Superfund Site. This includes two non-
time critical removal actions (one for the Embankment area and the other for Area 5106) along with the
remedial action of dredging the Hylebos channel to the native sediment interface. Approximately 36,000
cubic yards of contaminated subtidal sediment from Area 5106 was hydraulically dredged and thermally
treated before disposal in a confined disposal facility between October 2002 and March 2003.
Additional work, however, will be required to meet sediment quality objectives and contaminant
concentrations are sufficiently high that a standard sand cap would become contaminated over time and
would not prevent the flow of contaminated ground water into the Hylebos Waterway. Additional
characterization and remedy selection efforts are underway but are not discussed in this RSE report. The
EPA Superfund program is working with Ecology to ensure that a remedy for the site is developed that
will remediate contaminated sediments, prevent sediment recontamination, and prevent discharges of
ground water into the Hylebos Waterway above applicable regulatory thresholds.
Site contamination predominantly consists of volatile organic compounds (VOCs) and elevated pH in the
ground water. The primary VOCs are tetrachloroethylene (PCE), trichloroethylene (TCE), and
breakdown products including cis-l,2-dichloroethylene (cis-l,2-DCE) and vinyl chloride. A pump and
treat (P&T) system has operated at the site since 1996, extracting contaminated ground water, treating it,
reinjecting a portion of the treated water to the subsurface, and discharging the remaining treated water to
the Hylebos Waterway. The elevated pH (8.5 to 13+) is widespread across the site and discharges to the
Hylebos Waterway. In addition to discharging to surface water above regulatory limits, the elevated pH,
particularly above 9.5, limits the effectiveness of injection and extraction systems due to fouling. The
RSE at this site addresses this P&T system but also considers other aspects of the site as they relate to
this system.
1.5.2 POTENTIAL SOURCES
Based on a review of facility operations and the extent of elevated pH, the June 2003 Site
Characterization Report suggests 14 potential locations for sources of the elevated pH. These locations
include waste management units (WMUs) C and H, which are adjacent to the Hylebos Waterway. In all
instances, it is presumed that caustic material (e.g., spent lime from processing activities or unspent lime)
was released into the environment during processing, transport, storage, or disposal activities. These
caustic materials have a high density compared to water and will tend to migrate vertically downward
over time, which offers a potential explanation for the "pH blob" found at a depth of over 100 feet.
-------
The same Site Characterization Report identifies 12 potential source locations for the VOC
contamination. At each location, the release of chlorinated solvents may have occurred. The report
suggests that the most likely sources are the former settling pond areas, where effluent from the PCE and
TCE processing was released. These areas are all located in the general area between Dock 1 and
Alexander Avenue. Although soils were removed from one of these locations (WMU H), remaining total
chlorinated organic compound concentrations in soil were as high as 150 mg/kg. Another area, near
Dock 2, may provide an additional source of VOC contamination; however, VOC concentrations in
shallow groundwater are relatively low in this area, suggesting that the other sources are likely
responsible for the majority of the VOC plume.
VOCs in the form of DNAPL are likely present. Although DNAPL has not been directly observed in
multiple locations, the concentrations of some individual VOCs in ground water exceed 1% of their
solubility. Furthermore, substantial contamination exists at depth, and the presence of DNAPL provides
a reasonable explanation for this deep contamination.
1.5.3 HYDROGEOLOGIC SETTING
The June 2003 Site Characterization Report describes the shallow subsurface as a thick sequence (about
150 feet) of fluvial deltaic deposits, underlain by the less permeable Vashon Till unit and overlain by fill.
The fluvial deposits consist mainly of sand and silty sand with discontinuous, thin layers of clayey silt,
silt, and silty sand. The lower portion of this aquifer also contains some gravel zones. The overlying fill
is likely dredge material and is discontinuous and variable across the site. The Vashon Till likely
provides a lower confining unit to the fluvial deposits. The horizontal hydraulic conductivity of the
fluvial deposits is estimated to range between 10"4 and 10"2 cm/s (approximately 0.3 to 30 feet per day).
Ground water recharge to the shallow aquifer comes from two sources: infiltration from precipitation and
flow from the Vashon Till that is recharged by precipitation in the highlands to the north. Surface
recharge is likely limited by asphalt and concrete over much of the site and the neighboring properties,
and is estimated at 2 to 8 inches per year in the 1996 Corrective Action Plan. In the absence of pumping,
ground water flow in the upper 50 feet of the aquifer is dominated by this limited surface recharge, which
creates a ground water divide parallel to Alexander Avenue. Ground water east of the divide discharges
to the Hylebos Waterway, and ground water west of the divide discharges to the Blair Waterway.
Ground water flow in the 50 to 100-foot depths is dominated by regional flow and generally flows west.
Although two distinct ground water flow patterns exist at the site in the fluvial deposits, it is generally
understood that there is no continuous confining unit separating the areas with the different flow patterns.
Discontinuous clay layers are likely present however. A cross-sectional conceptual diagram of ground
water flow is presented in Figure 1-3.
Water levels are influenced by both tidal and density effects and have been corrected for in historical site
documents. In general, the tide ranges from -4 to 14 feet NGVD and ground surface at the site is
approximately 20 feet NGVD. The June 2003 Site Characterization Report calculates an approximate
ground water velocity of 0.4 to 4 feet per day. During pumping vertical gradients suggest an upward
component of ground water flow, particularly between the 100 and 50-foot depths.
1.5.4 POTENTIAL RECEPTORS
Ground water in the area surrounding the site is not used for drinking water. The primary receptors are
the Blair and Hylebos Waterways, with most emphasis on the Hylebos Waterway because of the
proximity to the site, the observed contamination seeps (discussed in Section 1.5.5), and the ongoing
Superfund remedy for the embankment. The primary potential human exposure is through the
consumption offish caught in the Hylebos Waterway.
-------
1.5.5 DESCRIPTION OF GROUND WATER PLUME
Ground water contamination at the site is extensive both horizontally and vertically, with contamination
extending from the surface to over 100 feet in depth and VOC contamination extending beyond the
horizontal boundaries of the 37-acre OCC property.
Most of the site monitoring wells are screened in three intervals that are approximately 25, 50, or 100
feet below ground surface within the fluvial deposits. Ground water contamination is generally tracked
in each of these intervals but a number of wells are also completed at greater depths. Figures 1-4 through
1-6 show the extent of VOC contamination and the areas with pH over 10 for the 25, 50, and 100-foot
depth intervals, respectively. The elevated pH at the 100-foot interval corresponds to the "pH blob"
identified in Figure 1-3.
As is evident from the figures, the VOC contamination is adjacent to, and likely discharges to, the
Hylebos Waterway to the east. The VOC contamination also extends onto the Port of Tacoma property
to the north and west. At the 25 and 50-foot intervals, the VOC contamination appears delineated;
however, at the 100-foot depth interval, the VOC plume extends to the north and is not fully delineated.
Many locations in all intervals have individual and/or total VOC concentrations greater than 10,000 ug/L.
In general, however, the greatest VOC contamination appears to be at the 100-foot depth interval with
concentrations indicative of dense non-aqueous phase liquid (DNAPL).
The elevated pH in the 25 and 50-foot intervals appears more limited in extent than the VOC
contamination. Investigation during the 2003 Rapid Groundwater pH Assessment included pH sampling
at the 75-foot interval. These results at the 75-foot interval (not shown in this report) suggest that the
elevated pH at the 100-foot interval is distinct from the elevated pH plumes at shallower depths.
Ground water discharges to the Hylebos Waterway, and two distinct types of seeps have been identified:
the flowing seeps and the milky seeps. The pH of the flowing seeps is approximately 8.5 and the pH of
the milky seeps ranges from 8 to 11.2. The flowing seeps (also referred to as the bank seeps) have
similar chemistry to the Waterway and likely represent bank storage associated with the tidal cycle. The
milky seeps, however, are identified by the presence of suspended white solids or encrusted white solids
at the discharge point. The solids are likely the result of high pH ground water discharging to the
waterway causing calcium and magnesium to precipitate. The milky seeps are also contaminated with
VOCs. During the rapid pH assessment, PCE and TCE were detected in the pore water at concentrations
of 200 ug/L and 110 ug/L, respectively. In addition, nearby ground water (as determined from direct-
push samples) had PCE concentrations of 420 ug/L and 99 ug/L. The sediments along the embankment
are also contaminated with PCE and TCE. Sediment samples during the rapid pH assessment indicated
PCE and TCE concentrations of up to 6,100 mg/kg and 3,190 mg/kg, respectively, that remain after
dredging.
Subsequent to the RSE site visit, EPA reported to the RSE team that the properties to the south (i.e., the
Mariana Property and the U.S. Navy and Port of Tacoma Properties) also have shallow ground water
contaminated by PCE and high pH. The site team has reportedly been reviewing data from these
properties to better characterize the OCC site.
-------
2.0 SYSTEM DESCRIPTION
2.1
SYSTEM OVERVIEW
The original P&T design was completed in 1993. The pumping system was to have 35 extraction wells
in three branches with pneumatic submersible pumps. The treatment portion of the system was to have a
dual train system for separate treatment of relatively neutral and high pH groundwater. The treatment
was to include (pH adjustment for high pH water) solids removal, steam stripping, liquid GAC polishing
and vapor destruction by catalytic oxidizer. Sludge would be dewatered with a belt press and sludge
dryer.
The treatment plant was built following the original design and began operation in March 1994 with one
well and one collection trench. The extraction well, located in a high pH area, had significant fouling
problems, and the collection trench had little hydraulic influence at depth. Due to the problem extracting
water with a high pH (i.e., greater than about 9.5 s.u.), the pumping system was revised to include
injection near the Hylebos Waterway and extraction throughout the other areas of the site where the pH
was less than 9.5. The injection was to provide a hydraulic barrier to the east (intended to prevent
discharge to the Hylebos Waterway) and the extraction was to provide mass removal and a hydraulic
barrier to the west.
Because high pH ground water would not be extracted, the pH adjustment portions of the treatment
system have been bypassed. In addition, the steam stripper has been replaced by an air stripper (in 1998)
and the belt filter has been replaced by a plate and frame press. The P&T system in this form began
operation in May 1996.
2.2
EXTRACTION AND INJECTION SYSTEMS
There are currently four extraction branches (A, B, C, and D) with a total of 25 extraction wells. Of
those 25 wells, 22 are screened from about 40 to 70 ft bgs with drop tubes to a depth of about 27 feet bgs,
and three (A-2A, D-4, and D-5) are screened at approximately 100 feet. Two of these three deeper wells
(A-2A and D-5) were added between December 2001 and January 2002, with screened intervals at
approximately 100 feet deep. The following table presents the approximate distribution of extraction
among the four extraction branches in August 2002 as determined from flow totalizers on the individual
extraction networks.
Extraction Branch
A
B
C
D
Total
Approximate Extraction Rate
52gpm
17gpm
24gpm
12gpm
105 gpm
-------
A small portion of the treated water is reinjected by gravity through two injection branches (E and F)
located along the Hylebos Waterway, with a total of 21 injection wells. The remainder of the extracted
water is discharged to the Hylebos Waterway under a NPDES permit. Figure 1-2 depicts the locations of
each extraction and injection well along with the NPDES outfall.
The extraction rate is approximately 105 to 110 gpm and has previously been over 140 gpm. After
treatment, less than 5 gpm (and likely less than 2 gpm) is reinjected. The injection system is operated to
maintain a level of 2.5 ft MSL or greater in the wells. A flow meter, actuated valve and manual valve
and transducer are included at each injection well. The actuated valve is opened automatically based on
the water level to allow flow to the well. The injection rate has been decreasing overtime as a result of
fouling. For example, the initial injection rate in 1996 was approximately 20 gpm. In addition, most of
the water is reinjected through injection branch E.
2.3 TREATMENT SYSTEM
The treatment system is currently treating about 110 gpm. It has hydraulic capacity for about 200 gpm
and possibly higher with pump replacements. The treatment system includes the following components:
20 HP vacuum assisted centrifugal pump for ground water extraction
20,000 gallon decanter tank (no DNAPL has been collected)
lamella settler with sludge pump
8,500 pumping tank with 10 HP pump
100,000 gallon holding tank with 30 HP pump
Hazleton Maxi-Strip air stripper with (5) 7.5HP circulation pumps and a -15 HP effluent pump
two 20,000 pound GAC vessels arranged in series
injection holding tank
catalytic oxidizer with scrubber and 100-foot stack (for treating -700 cfm air from the air
stripper and a small amount of additional system air)
DURCO plate and frame filter press for dewatering sludge
sludge drier
Allen-Bradley control logics system with a PC interface
Solids from the settler and GAC backwash are pumped to a holding tank then through the filter press and
sludge drier. The solids from the settler account for approximately 5% of this volume and the GAC
backwash accounts for approximately 95% of this volume. Tank levels and flow rates can be monitored
through the control system, and an autodialer is present to alert operators to alarm conditions in off-
hours.
Prior to the addition of A-2A and D-5, the influent concentration of total contaminants of concern
(COCs) was between 15,000 and 20,000 ug/L, which translates to an approximate COC mass removal
rate of 18 to 24 pounds of per day. After the addition of A-2A and D-5, the influent concentration of
COCs has ranged from 25,000 ug/L to 40,000 ug/L, which translates to a COC mass removal rate of well
over 30 pounds per day. This increase in influent concentrations caused an NPDES exceedance for
chloroform and 1,1,2-TCA in early 2002 because the system could not remove sufficient contaminant
mass. The GAC changeout frequency has been increased to address the added influent mass, and no
further NPDES exceedances have occurred. Pumping from these two wells is also throttled to prevent
the extraction of more contaminant mass than the treatment system can handle.
-------
2.4
MONITORING PROGRAM
The system effluent is sampled and analyzed weekly for VOCs and monthly for select inorganics per
NPDES requirements. Additionally the system influent, stripper influent, stripper effluent, and first
GAC unit effluent are sampled and analyzed weekly for VOCs.
Approximately 38 monitoring wells are sampled on a quarterly frequency and 37 additional wells are
sampled semi-annually. The monitoring wells and sample frequency from the February 2003 Monitoring
Event and System Evaluation Report are summarized in the following table.
Quarterly Sampling
25-foot*
14-25R
15-25R
16-25
12-25
25-25
5-25
8-23
35-25
36-25
40A-25
43-25
50-foot*
14-50R
15-50R
12-45
18-50R
23-50
25-50
5-50
56-50
57-50
8-54
35-50
36-50
40-50
43-50
100-foot*
5-100
15-120
12-160
8-99R
36-100R
40-100R
40A-100
six additional new wells are proposed for quarterly
monitoring
Semi-Annual Sampling
25-foot*
Tl-25
19-25
21-25R
22-25R
4-25R
53-25
50-foot*
Tl-50
16-50
41-50
19-50R
21-48
22-50
22-70
A-5
B-3
C-3
C-7
4-45R
11-45
32-50R
34-50
53-50
100-foot*
Tl-100
1-100R
41-100
41-138
A-2A
D-4
4-83R
4-115R
4-175R
11-100
11-183
35-100
53-100
two additional new wells are proposed with semi-
annual sampling
* Approximate depth of screened interval.
In addition to water quality sampling and analysis, the monitoring program includes regular
measurements of water level. Water levels are collected quarterly from all injection wells, seven
extraction wells, 65 monitoring wells, and the Hylebos Waterway. In addition, on a quarterly basis,
water levels are measured continuously for 72 hours in 28 of the above wells to provide a correction for
tidal influence. All water levels are converted to fresh water equivalent heads to account for the spatial
variation in density. The extraction wells are also sounded on a regular basis and potentiometric surface
maps are generated on a quarterly basis primarily using water levels from injection wells, extraction
wells, and a limited number of monitoring wells.
The monitoring program also includes a bi-annual hydraulic conductivity assessments for the monitoring
wells included in the monitoring program and redevelopment of those wells where the hydraulic
conductivity has decreased by more than 50% from its baseline.
-------
3.0 SYSTEM OBJECTIVES, PERFORMANCE AND CLOSURE
CRITERIA
3.1
CURRENT SYSTEM OBJECTIVES AND CLOSURE CRITERIA
At the time of the RSE site visit, the site remedy was operating under an expired RCRA permit while an
Agreed Order and new permit were being negotiated. The expired permit specified four requirements
that will likely be included in some form in the negotiated order:
creation, through the operation of the injection wells, a continuous hydraulic barrier to the flow
of site ground water into the Hylebos Waterway
interception, by the extraction system, of ground water flowing from the injection system toward
the Blair Waterway
containment and recovery of the ground water containing chemicals of concern west of
Alexander Avenue
adequate progress toward achieving ground water cleanup
Adequate progress toward achieving ground water cleanup is defined in the permit as continuing to meet
cleanup standards in wells where the standards are already met and observing a declining trend in wells
where the standards are not met.
Although this permit has expired due to the transfer of oversight from EPA to Ecology, the same
requirements listed above apply and will likely be included in the new Agreed Order under the Model
Toxics Control Act (MTCA) and Ecology oversight. One set of cleanup standards has applied under the
RCRA permit, and another set of standards will apply under the MTCA. The following table summarizes
the both sets of standards for a subset of the site COCs. The appropriate cleanup standards for the site
have not been determined and are a current topic of discussion for the site team.
Parameter
Methylene chloride
trans- 1 ,2-Dichloroethylene
Trichloroethylene
1 , 1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Carbon tetrachloride
1 , 1 -Dichloroethy lene
Chloroform
1 , 1 ,2-Trichloroethane
Vinyl chloride
RCRA Permit Standard
(ug/L)
5.0
5.0
5.0
5.0
7.0
5.0
5.0
6.0
6.0
10.0
Proposed MTCA Standard*
(ug/L)
960
32,800
55.6
6.48
5**
5**
5**
283
25.3
5**
*Model Toxics Control Act, Method B Surface Water Cleanup Levels, February 2001
**Practical quantitation limit, SW-846Method 8260.
10
-------
In addition to the above objectives, the site has the following objectives under the EPA RCRA Program
and the EPA Superfund Program.
demonstrate that there are no human exposures to site-related contamination (human exposures
RCRA Environmental Indicator)
demonstrate the migration of contaminated ground water is under control (ground water RCRA
Environmental Indicator)
prevent remediated sediment or a sediment cap for the Hylebos Waterway embankment from
becoming recontaminated by VOCs or adversely affected by solids fouling due to the elevated
pH (implied objective for Commencement Bay Superfund Site)
Because the only expected human exposure is through consumption of fish from the Hylebos Waterway,
human exposures can be eliminated if complete containment can be achieved and demonstrated.
However, because contamination and contaminant source material likely remains between the current
injection line and waterway, the complete containment cannot likely be provided by injection. As a
result, it appears that potential human exposure and contamination of the cap could occur if injection
continues. It is noted that as a nominal value, the Superfund Program indicates that a PCE concentration
of greater than 20 ug/L discharging to the waterway would be sufficient to cause sediment contamination
in excess of the applicable standard (57 mg/kg) and that this PCE concentration could continue to
discharge to the waterway, even if the injection system operates as intended.
It should be noted that subsequent to the RSE site visit, EPA noted that the site team (including OCC)
agreed that the sediment planned cap for the embankment area would not chemically isolate or
adequately attenuate VOC contaminated ground water. The site team is discussing how to approach
sediment remediation.
3.2
TREATMENT PLANT DISCHARGE CRITERIA
The discharge criteria for the treatment plant effluent are stipulated by an NPDES permit that currently
applies to both reinjected water and water discharged to the Hylebos Waterway. The discharge limits are
provided in the following table:
Parameter
Vinyl chloride
1 , 1 -Dichloroethy lene
Methylene chloride
trans- 1 ,2-Dichloroethylene
cis-l,2-Dichloroethylene
Chloroform
Carbon tetrachloride
Trichloroethylene
1 , 1 ,2-Trichloroethane
Tetrachloroethylene
1 , 1 ,2,2-Tetrachloroethane
Concentration (ug/L)
NA
5
185
NA
NA
75
10
440
5
50
180
11
-------
The RSE team notes that the NPDES discharge limits for some parameters, such as TCE and PCE, are
higher than the cleanup standards detailed in the former RCRA permit and the proposed MTCA cleanup
standards. In addition, the discharge standard for PCE (50 ug/L) is greater than the 20 ug/L that
Superfund says can contaminate a sediment cap. Under the current system, this represents an
inconsistency given that injection of water with concentrations above cleanup standards would be
allowed. Fortunately, the responsible party voluntarily treats extracted water to concentrations that are
lower than both the cleanup standards and the discharge standards.
Other limits for treatment plant effluent are established for inorganics, including pH in the range of 6 to
9. Flow is required to be less than 24,500,000 gallons per day (over 10,000 gpm), which is orders of
magnitude higher than is practicable to extract from ground water and treat at this site.
12
-------
4.0 FINDINGS AND OBSERVATIONS FROM THE RSE SITE VISIT
4.1 FINDINGS
The RSE team observed a site with complex hydrogeology and substantial subsurface contamination.
The observations provided below are not intended to imply a deficiency in the work of the system
designers, system operators, or site managers but are offered as constructive suggestions in the best
interest of the EPA and the public. These observations obviously have the benefit of being formulated
based upon operational data unavailable to the original designers. Furthermore, it is likely that site
conditions and general knowledge of ground water remediation have changed over time.
4.2 SUBSURFACE PERFORMANCE AND RESPONSE
4.2.1 WATER LEVELS
Potentiometric surface maps are generated and reported for each of the three depth intervals (25, 50, and
100 feet bgs) on a quarterly basis. For the 25 and 50-foot intervals, the maps are generated using water
levels from monitoring wells and operating injection and extraction wells. However, the water levels
from operating injection and extraction wells are not likely indicative of aquifer conditions and using
these maps to draw conclusions about plume capture is unreliable. Figure 4-1 provides an example of a
potentiometric surface map generated from the January 2002 water level data. The RSE team notes the
following inconsistencies from this particular potentiometric surface map and that similar inconsistencies
are present in other submitted potentiometric surface maps.
Monitoring wells located between operating extraction wells have substantially higher water
levels indicating that the water levels in the extraction wells are not indicative of the surrounding
aquifer. For example, monitoring well 53-25 has a water level of 1.25 feet and extraction wells
D2 and D3, which straddle 53-25 have water levels of-4.58 and -4.40 feet, respectively. To
account for this, the potentiometric surface map includes a contour representing an "island" of
elevated water between two contours representing substantially lower water levels. Rather, true
conditions in the aquifer are likely represented by the monitoring well, and the extraction wells
represent small "islands" of reduced water level.
Monitoring wells T5 and T6 are located near injection well F5. The potentiometric surface map
contours are drawn based on the water levels from F5 and the surrounding injection wells. The
resulting gradient is from east to west between the injection wells and the extraction wells.
However, the injection well water levels are likely biased due to fouling and are not indicative of
aquifer conditions. T5 and T6 have water levels of 2.61 and 4.46 feet, respectively, which would
indicate a gradient from west to east.
Using data from monitoring wells (and not injection or extraction wells) would provide a different
picture of ground water flow across the site. The following examples are provided:
13
-------
Referring to monitoring wells Tl, 4R, and 53, there is a flow divide near 4R with water east of
the divide flowing east and water west of the divide flowing west. Such a flow divide may also
be present near T6, however, insufficient water level data is available to make this conclusion.
Referring to 25A and 25, ground water apparently flows from east to west away from extraction
wells C5 and C6. This could suggest that the influence from C5 and C6 is relatively small and
providing ineffective capture. An additional piezometer between 25A and 25 might indicate that
there is a flow divide between the two wells, suggesting that capture might be sufficient.
In general, the water levels on the southern end of the site appear higher than those on the
northern end of the site. Monitoring wells 8 and 25A have levels of over 3 feet while 40A, 40,
12, 12A, and 53 have levels of under 3 feet. This suggests that there may be a northward
component of ground water flow. This northward component could be due to natural ground
water flow patterns but could also be due to preferential extraction from branches A, B, and D
compared to extraction at branch C. This is consistent with reports during the RSE site visit that
the C branch has been dewatered frequently from pumping. It is also consistent with extraction
records that show that in 2002 the A branch is responsible for the majority of the extraction.
Water level data from existing monitoring wells generally support some of the conclusions made from
the existing potentiometric surface map. The following examples are provided using the January 2002
data.
Referring to 40A and 40 ground water apparently flows from north to south toward extraction at
A-l,A-2,andA-3.
Referring to 12 and 12A, ground water apparently flows from north to south toward extraction at
B-4.
Referring to 44 and 35, ground water apparently flows from west to east toward extraction at B3.
These three examples support inward gradients toward the extraction wells for this January 2002 data set
for the 25-foot interval. Other data sets appear in general agreement, but there may be a few exceptions.
In conclusion, however, the above examples clearly indicate that the potentiometric surface maps for the
25 and 50-foot intervals (both of which rely on operating extraction and injection wells for data) are
likely not representative but that capture in some areas is likely sufficient.
4.2.2 CAPTURE ZONES
Given that the potentiometric surface maps are unreliable, using these maps for interpreting plume
capture is also unreliable. In addition, insufficient data are available for fully evaluating plume capture.
There is, however, ample evidence that a degree of capture is provided in the 25-foot depth interval (and
likely the 50-foot interval). As stated above, there appears to be a flow divide near 4R and potentially
near T6. Therefore, contamination to the west of these wells is likely captured and prevented from
flowing east into the waterway. On the other hand, it is reasonable to conclude that contaminated ground
water to the east of this apparent divide continues to flow into the waterway.
For capture in other directions, the water level data from well pairs and water quality data from sentinel
wells provide useful information. Three of the well pairs noted in the previous section (40A/40, 12/12A,
and 44/35) indicated capture to the north and west. Well pair 25A and 25 indicates a potential lack of
capture, but the absence of contaminants above cleanup standards at 25 supports capture (provided that
14
-------
the concentrations do not increase over time). Other ground water quality data also support capture at
the 25-foot depth interval, including monitoring wells 40A-25, 12-25, 42-25, and 44-25. However,
monitoring wells 43-25 and 36-25 have detectable VOCs concentrations that are higher than previous
data, suggesting the potential for migration in these areas. The above analyses only apply to the 25-foot
depth interval during January 2002, but similar analyses and likely similar conclusions can be made
regarding the 50-foot depth interval and other sampling events
Capture at the 100-foot depth interval is inconclusive due to a lack of data, and a thorough capture
evaluation at this depth is also complicated by the lack of plume delineation. In particular, the
concentrations at monitoring wells 41-138, 40A-100, and 40-100 all have VOCs well above relevant
cleanup standards, and there are no monitoring wells or extraction wells downgradient of these locations
to provide delineation. The substantial decrease in VOC concentrations in 41-138 could be interpreted as
an indication of capture, but the decrease started well before the addition of A-2A. Therefore, the
decrease is likely due to dilution from the injection of clean water at E-l and the movement of
contaminated water further from the site. The increases in VOC concentrations at 40A-100 and 40-100
might indicate that these wells are within the capture zones of nearby extraction wells and that
contaminated ground water is passing through these monitoring wells on route to extraction at A-2A. It
could also potentially represent uncaptured contamination. Moreover, substantial contamination may
exist beyond 40A-100 and 40-100 given that these two wells have substantial concentrations (38,000
ug/L of total contaminants of concern at 40-100 and 9,800 ug/L of total contaminants of concern at 40A-
100 in August 2002) and no monitoring points are available to the north.
4.2.3 CONTAMINANT LEVELS
Elevated pH
The areas with elevated pH have not changed substantially since the operation of the treatment system
began. One benefit of this lack of change, is that the elevated pH is not migrating toward the extraction
system. If the extraction wells begin extracting water with elevated pH, substantial solids fouling of both
the extraction wells and the treatment system will likely occur. The relatively slow migration or lack of
migration of the elevated pH may result from the natural buffering capacity of the soil.
VOC Plume
Trends in contaminant concentrations have been observed in a number of site wells. Some wells have
increased in concentration while others have decreased in concentration. Some wells have shown large
variations in concentrations between 1996 and 2002 with no discernible trend, perhaps indicating the
influence of nearby injection or extraction.
Select Perimeter Wells
Some of the perimeter wells that have decreases in concentrations include 19-25, 35-25, and 57-50. In
general, these wells provide evidence of some control over the plume to the west and the south. Other
perimeter wells that show decreases in concentration include the Tl series, 14-50, and 41-138. The
decreases in these wells are likely due to dilution from injecting treated water. While this is a desired
effect for Tl, it is not likely desirable at 14-50, which is adjacent to the waterway, or 41-138, which has
no nearby extraction wells to capture the contaminated ground water that might be spreading. Some
perimeter wells are also increasing in concentration, including 40-100, 40A-100, 15-25R, and 36-50.
These increases may be of concern if there is no nearby extraction (e.g., 36-50) or if the wells are on the
waterway side of the injection system (e.g., 15-25R).
15
-------
Select Interior Wells
A number of the interior wells have shown decreasing trends including many of the wells in the 4 series,
11-45, 11-100, 21-25R, 32-50, 34-50, 53-25, and 53-50. Although decreases are evident, contamination
in many of these wells remains elevated, with some concentrations exceeding 10,000 ug/L of total COCs.
Other interior wells have either fluctuated without a discernible trend or have remained largely
unchanged since pumping began.
Treatment Plant Influent
The monitoring of the influent to the treatment plant indicates that substantial contaminant mass is being
removed from the subsurface. On average between October 1997 and September 2002 approximately
750 pounds of contaminants have been removed from the subsurface per month. Although new
extraction wells A-2A and D-5 provide extra potential to remove contaminant mass, the current treatment
system is not designed to cost-effectively remove this much additional mass.
Summary of Contaminant Levels
In general, the ground water quality monitoring provides a mixed interpretation for capture. In some
areas, capture appears adequate (e.g., some areas to the west), in other locations capture might not be
adequate (e.g., to the north at the 100-foot depth interval and along the waterway), and in still other areas
insufficient data are available for a thorough capture evaluation (e.g., some areas to the southwest).
Ground water quality monitoring and process monitoring demonstrate that substantial contaminant mass
is being removed, and additional mass can be removed with additional extraction if the treatment system
is upgraded. Decreasing concentrations have been evident in some locations along the perimeter of the
plume and in the interior of the plume, but substantial contamination remains. For example, at
monitoring well 4-25R, the concentration of contaminants of concern has decreased from 120,000 ug/L
but has stabilized at over 50,000 ug/L. At 11-100, the concentration of contaminants of concern has
decreased from 450,000 ug/L but has stabilized at over 250,000 ug/L (including TCE at over 100,000
ug/L). With these persistent, elevated concentrations, it is likely that DNAPL is present as a continuing
source of dissolved ground water contamination and that remediation might occur indefinitely.
Therefore, demonstrating meaningful progress toward complete cleanup is difficult.
4.3 COMPONENT PERFORMANCE
4.3.1 EXTRACTION SYSTEM WELLS AND VACUUM PUMP
The extraction system continues to extract approximately 105-110 gpm, and based on this information
alone, there is no indication of a problem with the extraction system. However, the extraction system is a
network of vacuum extraction wells manifolded to one 20 HP vacuum pump. If fouling of one well
occurs the flow rate could be made up by other wells resulting in reduced extraction in some areas and
additional extraction in others. For a containment remedy with relatively evenly spaced wells, localized
reductions in extraction could result in a reduction in plume capture. Of particular concern is extraction
branch C, which has reportedly been running dry over the previous months prior to the RSE site visit. In
addition, flow rates in extraction branch C have declined from 55 gpm in 1999 (approximately 40% of
the total system extraction) to 24 gpm in August 2002 (approximately 23% of the total system
extraction).
16
-------
The October 2001 through September 2002 Annual Performance Evaluation Report also provides
evidence of decreasing well efficiency. For example, the report explains, "The horizontal gradient across
the cone of depression in May and August 2002 was approximately 12 feet, from an approximate
elevation of 0 feet NGVD at the perimeter of the plume and 3 feet NGVD along the injection system to
an elevation of approximately -11 feet NGVD at the center of the cone of depression. This gradient has
increased by approximately one foot in each of the least two reporting periods." This increased gradient
is based on increased drawdowns in extraction wells. This increased drawdown is likely due to a
decrease in well efficiency due to fouling and is representative of the water in the operating extraction
well but is not likely representative of the surrounding aquifer.
4.3.2 AIR STRIPPER
The air stripper provides the primary mechanism for removing VOCs from the extracted ground water,
and its efficiency is lower than the original design efficiency despite cleaning on a monthly or bimonthly
basis. The two problematic constituents are chloroform and 1,1,2-TCA. The stripper efficiency for
chloroform has generally been well below the design efficiency. For example, in June 2003 the
efficiency for chloroform was 97.65% compared to a design efficiency of 99.18%. With the design
influent concentration of chloroform over 1,000 ug/L, a further increase in the chloroform influent
concentration or a further decrease in the efficiency will lead to increased reliance on GAC to meet the
NPDES requirements. The 1,1,2-TCA is generally undetectable in the plant influent due to the high
concentrations of the other constituents that require samples to be diluted. Historical data have indicated
that the influent 1,1,2-TCA concentration is less than 50 ug/L, but the recent increases of VOCs in the
influent due to extraction from D-5 suggests that the influent concentration of 1,1,2-TCA may be up to
100 ug/L. Rather than approximately 5 ug/L, the stripper effluent for 1,1,2-TCA is now approximately
15 ug/L. The stripper now rarely meets the discharge criteria of 5 ug/L and the treatment plant is reliant
on the GAC to meet discharge standards.
4.3.3 GAC
Approximately 17,000 pounds of GAC is replaced every two months, primarily due to the breakthrough
of 1,1,2-TCA, which has a relatively low isotherm for adsorption to GAC. Breakthrough of 1,1,2-TCA
through the GAC is likely also enhanced by channeling through the GAC caused by solids in the process
water. Backwashing of the GAC is required approximately once every five days and is conducted
manually by connecting the units to hydrants with hoses. The solids from the GAC backwash represent
the majority of the solids that require dewatering and disposal.
4.3.4 FILTER PRESS AND SLUDGE DRYER
Solids removal and associated sludge handling are necessary at this site especially when extraction wells
closer to the high pH (and highest VOC) areas are pumped. However, the amount of solids removed
(about 200 Ibs/day of sludge) is significantly less than the original design value of 13 tons/day as dry
solids. The current system includes a plate and frame filter press that replaced a belt press (an effective
optimization effort by the site team). The original sludge drier is still used even though it would not be
warranted unless it were already in place. Since it requires relatively little energy it is reasonable to
continue using it. Removal of the unit should be considered if any major maintenance to it is required.
The plate and frame press use should continue because the alternative of collecting the sludge in a
thickener tank and having it processed by a mobile service is not feasible due to the large volume of
dilute sludge. Maintaining the filter press also allows the flexibility of handling additional solids if
extraction wells are added.
17
-------
4.3.5
CATALYTIC OXIDIZER
The catalytic oxidizer requires regular maintenance and is likely nearing the end of its useful life,
primarily due to corrosion. It typically receives approximately 700 scfm of flow and a contaminant mass
of approximately 25 pounds per day. It is the leading cause of system down time, which is generally less
than 5% per year.
4.3.6
INJECTION SYSTEM
As stated in Section 2.2 of this report, the injection wells are fouled, which has substantially reduced the
amount of water that can be reinjected. Rather than the original 20 gpm injection rate, the current
injection rate is approximately 2 gpm. The majority of the injection occurs along the E injection branch.
4.3.7
CONTROL SYSTEM
The control system is an Allen-Bradley control logics system with a PC interface. The operator can
track a number of individual plant parameters including the extraction rate from each extraction branch
and the injection rate of each injection well. Data can be downloaded and reviewed but trends are
generally not analyzed. Most plant functions are automated, but items such as GAC backwash and solids
handling are manual. The system is equipped with an autodialer that calls an operator in case any of the
10 system alarms sound. The PC records the alarms to facilitate the diagnosis of problems.
4.4
COMPONENTS OR PROCESSES THAT ACCOUNT FOR MAJORITY OF
MONTHLY COSTS
Annual O&M costs have historically been approximately $1.5 million for the last several years, but
automation upgrades have required a significant effort and are part of those costs. The automation
upgrades successfully have reduced the required system staffing from five to two full-time operators, and
the site suggests future costs will be approximately $1 million. A breakdown of the O&M costs provided
during the RSE site visit is provided in the table below. Additional costs may apply.
Item Description
Labor (plant operator)
Labor (sampling, well gauging, reporting, etc.)
Labor (project management and contract maintenance)
Utilities: Electricity
Utilities: Natural gas
Non-utility consumables: GAC
Non-utility consumables: chemicals
Discharge fees and waste disposal
Total Estimated Cost
Estimated Cost
$150,000/yr
$250,000/yr
$175,000/yr
$60,000/yr
$60,000/yr
$80,000/yr
$6,000/yr
$10,000/yr
$791,000*
% of Total
Costs
19%
32%
22%
8%
8%
10%
<1%
1%
100%
* This value does not include the cost for chemical analysis, which may be $50,000 to $100,000 per year. The chemical analysis
cost provided during the RSE site visit was subsequently determined to be inaccurate, and an accurate or representative value
was not provided.
18
-------
4.4.1 UTILITIES
The treatment plant, including the vacuum pump for the extraction system, utilizes approximately 112.5
HP of pumps and blowers. At approximately $0.06 per kWh, this accounts for the majority of the
electricity cost. The natural gas at the plant is primarily used for the catalytic oxidizer and sludge dryer.
4.4.2 NON-UTILITY CONSUMABLES AND DISPOSAL COSTS
The GAC is the primary consumables expense at approximately $80,000 per year. Approximately
17,000 pounds of GAC is replaced every two months, and this frequency is increasing given the added
mass extracted from D-5. The only chemical added to the treatment plant is a defoamer that is added to
the air stripper. Hydrochloric acid from the catalytic oxidizer scrubber is recycled through the plant to
neutralize the process water. For disposal, approximately 200 pounds of sludge is produced per day,
which is dewatered with the filter press and sludge drier and then disposed of as non-hazardous waste.
4.4.3 LABOR
The two system operators that are employees of Glenn Springs/Occidental each work 40 hours per week
such that the plant is manned seven days per week. This means that there is overlap between the
operators three days per week (Wednesday, Thursday, and Friday). This double coverage allows the
operators to conduct a number of two-person tasks, including cleaning the filter press and GAC
backwashes. Other operator tasks include taking water levels, transferring solids from the settling tank,
dewatering sludge, permitting, taking PID readings from the catalytic oxidizer stack, and routine
maintenance. The system operators spend significant time on backwashing the GAC vessels about once
every five days due to solids fouling and emptying the filter press about once per week.
Ground water sampling consists of four events per year: one large event, one medium event, and two
smaller events. Staff from Vancouver, British Columbia and Stockton, California conduct the sampling.
Approximately three people are required for three weeks for the large event, three people for two to three
weeks for the medium event, and two people for two weeks for the smaller events. The site team is
currently discussing the option of establishing a local office, which could reduce travel costs associated
with sampling. This translates to approximately 1,225 man-hours per year. Additional hours are
required for reporting and the bi-annual well performance testing.
4.4.4 CHEMICAL ANALYSIS
As indicated in the footnote to the table, the chemical analysis cost provided during the RSE site visit
was subsequently determined to be inaccurate. Therefore, chemical analysis costs are not included in the
annual O&M cost provided in the above table.
4.5 RECURRING PROBLEMS OR ISSUES
Although not stated in the site reports, well fouling is likely one of the most significant issues for the
remedy. Well fouling reduces the amount of water that can be injected and therefore adversely affects
the attempt to establish a hydraulic barrier along the Hylebos Waterway. Well fouling also likely results
in preferential extraction from some extraction wells compared to others and reduced extraction overall.
This may compromise capture in some areas.
19
-------
The tidal influence on the water levels is also problematic. Unlike many sites, collecting water levels at
this site is more expensive than water quality sampling. This is primarily associated with the requirement
to collect continuous water levels over 72 hours at many locations, to calculate corrected site-wide water
levels from that data, and to frequently recalibrate those probes.
The exceptionally high total VOC concentrations in the extracted water from D-5 also provides a
problem given the current efficiency of the air stripper. Without an increase in the efficiency, extraction
from D-5 will be purposely limited to avoid unreasonable expense associated with GAC replacement.
The catalytic oxidizer also requires increasing maintenance due to corrosion and has been the primary
cause of system downtime.
4.6 REGULATORY COMPLIANCE
The treatment plant has had one exceedance of its NPDES discharge criteria for chloroform and 1,1,2-
TCA, and this was shortly after D-5 was brought on line. The site team has prevented further
exceedances of the NPDES permit by throttling back D-5 and increasing the frequency of GAC
changeout.
4.7 TREATMENT PROCESS EXCURSIONS AND UPSETS, ACCIDENTAL
CONTAMINANT/REAGENT RELEASES
No specific incidents were reported during the RSE site visit.
4.8 SAFETY RECORD
No specific incidents were reported during the RSE site visit.
20
-------
5.0 EFFECTIVENESS OF THE SYSTEM TO PROTECT HUMAN
HEALTH AND THE ENVIRONMENT
5.1 GROUND WATER
The only receptor for the site is the Hylebos Waterway. Contaminated ground water continues to
discharge to the waterway as evidenced by the milky seeps and likely through other areas (primarily near
WMU H). Preventing this discharge is a difficult task. Extraction near the waterway has been piloted
but not implemented full time because of the high pH and the related rapid fouling of the wells with
solids. Similarly, injection in this area also causes fouling, which reduces the amount of water that can
be injected and therefore hampers the ability to create a hydraulic barrier between the bulk of the
contamination and the waterway. Even if injection were feasible at greater volumes, contamination is
apparently present between the injection wells and the waterway, and injection would be expected to
push that contamination into the waterway.
5.2 SURFACE WATER
In addition to the discussion above, it is important to note that surface water represents the only known
human exposure pathway for the site. Native Americans are known to fish in the Hylebos Waterway, and
consumption of the fish could result in consumption of site-related contamination, particularly if
bioaccumulation occurs. No noticeable impacts have been found in surface water, but fish may feed on
organisms that live in the pore spaces or on the sediments, which have measurable VOC contamination.
Contaminated sediments are still present at the site, and ground water is discharging through sediment
pore space at concentrations above regulatory limits.
5.3 AIR
Although VOC concentrations at the site are high and the water table is relatively shallow, indoor air has
not been raised as a concern at the site. The majority of the facility is outside, however, some buildings
are present on site. The RSE team did not collect information regarding the construction of each building
at the site and cannot further evaluate the potential for indoor air exposures. Areas of particular concern
might include (but would not be limited to) those areas surrounding the following monitoring wells:
4R-25 (11,000 ug/L of PCE and 25,000 ug/L of TCE in August 2002)
21-25R (5,400 ug/L of PCE and 3,800 ug/L of TCE in August 2002)
53-25 (16,000 ug/L of PCE and 2,900 ug/L of TCE in August 2002)
5.4 SOILS
The majority of soils at the site are covered by concrete and asphalt and should not provide a route of
exposure, unless trenching or other subsurface activities are conducted. The RSE team did not fully
investigate the ground covering throughout the site.
21
-------
5.5 WETLANDS AND SEDIMENTS
Sediments as a receptor are discussed above in Section 5.2. Much of the contaminated sediments at the
site have been dredged, but some areas of contamination remain because not all areas were feasible to
dredge. A sediment cap was planned along the entire embankment; however, subsequent to the RSE site
visit, EPA noted that the site team (including OCC) agreed that the sediment planned cap for the
embankment area would not chemically isolate or adequately attenuate VOC contaminated ground water.
The site team is discussing how to approach sediment remediation.
22
-------
6.0 RECOMMENDATIONS
Cost estimates provided herein have levels of certainty comparable to those done for CERCLA
Feasibility Studies (-307+50%), and these cost estimates have been prepared in a manner consistent with
EPA 540-R-00-002, A Guide to Developing and Documenting Cost Estimates During the Feasibility
Study, July 2000.
6.1 RECOMMENDATIONS TO IMPROVE EFFECTIVENESS IN PROTECTING
HUMAN HEALTH AND THE ENVIRONMENT
The recommendations in this category primarily involve enhancing plume capture and improving the
evaluation of plume capture. Enhancing plume capture to the east would reduce the discharge of
contaminated water to the Hylebos Waterway, and complete plume capture in the other directions would
prevent further migration of the plume toward the Blair Waterway. These recommendations are based on
the following opinions of the RSE team.
The current capture zone analyses are not reliable, primarily due to bias caused by the extensive
use of water levels at operating injection and extraction wells and relative lack of water levels
from monitoring wells or piezometers.
With the substantial decrease observed in water injected through the injection wells, it is likely
that the intended hydraulic barrier along the Hylebos Waterway is not effective and that further
attempts to create a hydraulic barrier through injection would similarly not be effective.
Capture to the north and west in the 25 and 50-foot interval is likely sufficient but the current
capture zone evaluations do not credibly present this position.
There is a degree of capture to the north in the 100-foot interval, but it is difficult to determine if
it is sufficient due to a lack of water quality data in this area that precludes complete delineation.
In addition, it is the RSE team's opinion that source removal actions along the waterway would likely be
unable to completely remove the source. Based on sampling from monitoring well 14-50, contamination
is present at more than 50 feet below ground surface and more than 25 feet below the water table during
low tide. Based on sampling from monitoring well 1-100, DNAPL may be present at 100 feet below
ground surface and more than 75 feet below the water table during low tide. The combination of the
depth of the contamination, the extent of overlying saturated soil, the horizontal extent of the
contamination, the steep embankment, the existing infrastructure, and the variable tide would make
complete source removal extremely difficult and likely impracticable. The RSE is concerned that even
after substantial source removal efforts, source material would still remain. Therefore, the RSE team
generally favors a remedial approach that focuses on containment and best efforts to establish an inward
gradient. The site team may, however, determine that source removal in spot locations may be
beneficial. If the site team opts for source removal in spot locations, the RSE team would suggest
piloting in-situ approaches rather than excavation that extends many feet below the water line.
23
-------
6.1.1 ELIMINATE INJECTION SYSTEM AND USE EXTRACTION TO ENHANCE CAPTURE
Along the site boundary with the Hylebos Waterway, only two monitoring well pairs (non-injection
wells) are present to indicate hydraulic gradients. These are the 4R/T1 pair and the T5/T6 pair. At the
25-foot interval, both of these pairs routinely suggest a gradient toward the Hylebos Waterway,
suggesting the hydraulic barrier is ineffective. At the 50-foot interval the 4R/T1 pair also suggests a
hydraulic gradient toward the waterway, but the T5/T6 pair suggests an inward gradient that favors
capture. At the 100-foot interval, both well pairs show flow toward the waterway. Although injection
may have originally been sufficient to create an effective hydraulic barrier, well fouling and reduced flow
since 1996 have suggested that the current injection system is not adequate and that future injection
systems will likely suffer from the same problems.
The RSE team envisions the following four primary strategies for improving containment in this
direction, but only one of these appears practicable.
Rehabilitate the existing injection system and increase the injection rate (not recommended)
Rehabilitating the existing injection system will likely improve the injection rate; however, the wells are
likely fouled to such a degree that the improvement will be small and insufficient. If rehabilitation is
conducted, additional monitoring wells would be required to determine if a suitable hydraulic barrier is
established. The addition and analysis of data from these wells is discussed in Section 6.1.4. The
contaminant mass on the waterway side of the injection line would need to be remediated to prevent
further discharge, or be allowed to discharge as it has since 1996. The elevated pH and associated solids
fouling would likely hamper most remediation activities, including sparging, injection of nutrients, and
extraction.
Install another injection system, perhaps with the use of a trench (not recommended)
This approach would involve trenching in the saturated zone. It would disrupt above-ground facility
operations, and would have a limited influence at greater depths (as suggested in the extraction pilot test
in 1994). This lack of influence at depth is a significant limitation and one of the primary reasons the
RSE team does not recommend this approach. Furthermore, the same issues regarding additional
monitoring wells and remediation of contaminant mass on the waterway side of the injection line would
still apply. The trench itself would likely cost under $1 million.
Install sheet piling or some other impermeable barrier (not recommended)
This approach would have the same drawbacks as the injection trench with the added problem of needing
to extend the sheet piling to approximately 200 feet to the nearest confining unit. If it is feasible to drive
sheet pile to a depth of 200 feet given the surrounding infrastructure and waterway bank, the estimated
cost for this type of approach would likely be approximately $5 million. Without meeting a confining
unit, the potential would exist for water to discharge through the gaps between the sheet piling and the
confining unit. As with the injection system, there would be a portion of the contaminated aquifer that
would remain on the waterway side of the barrier. The RSE team believes that this approach offers little
benefit in meeting site objectives.
Increase extraction to enhance inward gradients and provide capture (recommended)
In the opinion of the RSE team, increased extraction likely provides the most effective and efficient
method for enhancing containment of the plume. The RSE team acknowledges that containment of the
24
-------
plume up to the waterway boundary may not be provided at all portions of the tidal cycle, but based on
the reviewed documents and the site visit, believes that this approach offers the greatest benefit to the site
team.
Increasing extraction will push the east/west flow divide further to the east.
If complete containment is not provided, increasing extraction represents an improvement in
reducing the discharge of contaminated water to the waterway.
Increasing extraction will increase the overall amount of mass removal from the subsurface.
Increasing extraction will make it easier to evaluate plume capture through the use of water
levels and water budgets.
Efforts should be taken to extract additional water from areas with a pH less than 9.5 (extraction in areas
with a higher pH is impracticable due to solids fouling) in the attempt to establish an inward gradient
throughout the site. The first step to accomplishing this increased extraction is to rehabilitate and/or
redevelop the existing extraction wells. Additional study would be needed to determine the capacity for
increasing extraction from existing extraction wells and to determine the optimal locations and extraction
rates for additional wells. If possible, some wells should likely be located near the Hylebos and others
should likely be located in other areas that require enhanced extraction, and extraction at all depth
intervals can be considered. Some wells, such as the southernmost wells in the C extraction branch,
could also be taken offline to arrange for more capacity in other more pertinent areas.
Increasing extraction will require modifications to the treatment system, but such modifications would
likely be required anyway given the inefficiency of the air stripper and the condition of the catalytic
oxidizer. These modifications are discussed in Section 6.2.1. Though the RSE team believes that
consistent pumping is preferred, the site team could consider varying the extraction rate with the tide,
increasing the rate when the tide is low and decreasing the rate when the tide is high. The large influent
tank should allow for this type of variation. As with the other approaches, additional monitoring points,
especially along the waterway, would likely be necessary to demonstrate capture.
Adopting the enhanced extraction approach would require a modification to one of the remedy objectives
in the RCRA permit (i.e., the first remedy objective listed in Section 3.1 of this report). The objective
could potentially be reworded as follows:
extracting ground water at an average rate of 150 to 200 gpm from within the VOC plume
boundary in a concerted attempt to establish an inward gradient from the waterway to the site
as determined by hydraulic gradients measured from a series of monitoring well pairs located
along the waterway, oriented perpendicular to the site, and screened at approximately 25, 50,
and 100 feet below ground surface
The objective could be further clarified to state that extraction need not occur within an area with pH
over 9.5 and that the "monitoring wells" refers to wells where routine extraction and injection do not
occur. Further clarification could even include specific monitoring well pairs.
Enhancing extraction will require capital costs to rehabilitate/redevelop existing wells, to investigate
other potential areas and depth intervals for new extraction wells, installing the new extraction wells, and
connecting them to the current treatment system. The scope of work could vary and the cost for full
implementation is difficult to estimate. The RSE team therefore only provides an estimated cost of
$285,000 for the following assumptions:
25
-------
analyzing historical extraction well water levels and extraction branch pumping rates to
determine those wells in greatest need of redeveloping and/or rehabilitating up to 20 wells
($50,000)
analyzing plume maps and bi-annual slug test data to determine productive and practicable
locations within the VOC plume (and preferably near the waterway) to add extraction wells
($15,000)
adding two extraction wells for each of the three depth intervals for a total of 6 extraction wells
($120,000)
connecting these wells to the pumping system with an assumed piping/trenching length of 1,000
feet and double-contained piping ($100,000)
Furthermore, the site team could consider developing a three-dimensional ground water flow model to
help evaluate various pumping scenarios. The site is relatively complex. Developing and calibrating a
ground water flow model would likely require substantial effort and might cost up to $50,000. Use of
this model in conducting simulations would likely be an extra cost depending on how the model was
used. However, using the model to assist in locating extraction wells could potentially reduce the
number of wells installed and the associated costs. The estimated cost of $50,000 is not included in
Table 7-1.
6.1.2 DISCONTINUE CURRENT APPROACH OF USING POTENTIOMETRIC SURFACE MAPS TO
EVALUATE CAPTURE
Potentiometric surface maps are used to evaluate capture at this site, and this current approach has the
following two primary complications that make it both unreliable and very expensive:
The current capture zone analyses are not reliable, primarily due to bias caused by the extensive
use of water levels at operating injection and extraction wells and relative lack of water levels
from monitoring wells or piezometers.
Because a potentiometric surface map requires site-wide water level data that are not biased by
tidal influence, extensive effort is required to continuously measure water levels for a 72-hour
period at 28 locations and to use the resulting data to eliminate the tidal influence from the site-
wide data. The resulting potentiometric surface maps are not necessarily representative of net
ground water flow and capture throughout the tidal cycle.
The RSE team suggests abandoning the effort to evaluate capture based on potentiometric surface maps
and to use three alternative approaches to evaluate capture. These alternative approaches include the
following:
analysis of water quality data at sentinel wells, which is currently done at the site and reported in
the performance reports but is further discussed in Section 6.2.2
calculation of a water budget that compares ground water flow through the site with ground
water extraction (discussed in Section 6.1.3)
use of monitoring well or piezometer pairs to demonstrate inward gradients along the plume
boundaries (discussed in Section 6.1.4)
26
-------
Implementing this recommendation would involve substantial cost savings that may be as high as
$125,000 per year if approximately half of the current $250,000 of sampling and reporting cost is
devoted to collection and interpretation of site-wide and continuous water level measurements. This
recommendation, however, is provided in the context of protecting human health and the environment
because the RSE team believes that the current approach to evaluating capture is unreliable, and in many
cases, misleading.
6.1.3 CONDUCT AND REPORT A WATER BUDGET ANALYSIS
Although the site team referred to a water budget analysis during the RSE site visit, such an analysis does
not appear to be conducted regularly and is not reported in the quarterly or annual performance reports.
An analysis was conducted as part of the 1996 Corrective Action Plan, which suggested approximately
14 gpm would be needed for capture. The O&M data (e.g., the water level information discussed in
Section 4.2.1) suggests that this is likely low and that the analysis should be repeated given the improved
understanding of the site. A demonstration that more water is extracted from the plume than flows
through the area requiring capture would support plume capture. It is generally agreed that flow through
this site is provided both through infiltration of precipitation and through contributions from the
underlying aquifers. With the initiation of pumping, water is also recharged from the surrounding
horizontal areas (providing inward gradients). With injection, water is also recharged through the
injection wells.
The analysis should include a revised calculation of the amount of recharge provided through the surface
and from the underlying formations. This recharge should then be compared to the extraction rate. As a
rule of thumb, it would be preferable for the extraction rate to exceed the recharge rate by a factor of 1.5
to 2.0. The necessary information is already collected. Infiltration could be estimated from local weather
information and surface cover. Recharge from the underlying formations could be calculated with
Darcy's Law using estimated vertical gradients between the 100-foot and 50-foot intervals and an
appropriate hydraulic conductivity. The RSE team suggests using the lower estimate of the horizontal
hydraulic conductivity (approximately 10"4 cm/s). Simulations with the ground water flow model
mentioned in Section 6.1.1 could be used in place of this water budget analysis.
As stated by the site contractor, this approach is one of the most straightforward methods of evaluating
capture at this site. The RSE team estimates that these evaluations should not add significantly to the
current annual evaluation and reporting costs since the majority of the work is already done but not
discussed in the reports.
6.1.4 INSTALL ADDITIONAL MONITORING WELLS AND USE PAIRS OF MONITORING WELLS
TO DEMONSTRATE CAPTURE WITH INWARD GRADIENTS
Comparing water levels from monitoring well pairs (two nearby monitoring wells) helps determine the
direction of ground water flow in that area. This is particularly useful at the plume boundary because an
inward gradient supports capture in that area. By monitoring gradients at well pairs along the entire
plume boundary, the capture of the entire plume can be evaluated without generating potentiometric
surface maps. A number of such monitoring well pairs are located around the site and can be used. The
RSE team notes the following existing well pairs that may be useful for this approach:
eastern plume boundary: T1/T3, T6/T5
southern plume boundary: none
southwestern boundary: 8/45, 25/25A, 8/55, 36/56, 36/57
27
-------
western boundary: 35/44
northwestern boundary: 34/43, 12/12A
northern boundary: 40A/40
Based on these current well pairs there are significant gaps, primarily along the eastern boundary.
Approximately 15 piezometers (to be used for measuring water levels only and likely not for sampling
ground water quality) could be added as follows:
a piezometer cluster located west of (and to be paired with) the monitoring well 2 cluster with
screened intervals at 25, 50, and 100 feet bgs
two piezometer clusters located between E-7 and E-8 and perpendicular to the embankment with
screened intervals at 25, 50, and 100 feet bgs
two piezometer clusters located near E-3/E-3A and perpendicular to the embankment with
screened intervals at 25, 50, and 100 feet bgs
These additional piezometers provide a total of 14 well-pair locations with three depth intervals per well
pair (a total of 42 well pairs or 84 wells). Water levels in these well pairs could be measured and
compared in order to determine the flow direction of ground water at the plume boundaries. Because
using well pairs is a new approach for this site, the RSE team suggests a quarterly frequency for the first
year and a semi-annual frequency thereafter. A further decrease in the sampling frequency to annual
events could be implemented if data from the quarterly and semi-annual events support a further
reduction.
Because this approach is based on one water level relative to only one other nearby water level, the water
level measurements from a particular well pair can be taken nearly simultaneously. Because there is little
or no time difference between the two water level measurements from a single well pair, the
measurements do not need to be corrected for tidal influence. However, steps should be taken to
demonstrate inward gradients at each well pair throughout the tidal cycle. For each gauging event, the
water levels for a particular well pair should be measured at least four times evenly spaced over the
course of 24 hours (e.g., a measurement every 6 hours). This could involve three shifts of two to four
individuals each working one 8-hour day or two shifts of two to four individuals each working a 12-hour
day.
The water levels from many of the well pairs will likely show an inward gradient for all portions of the
tidal cycle, confirming capture. For the well pairs along the Hylebos Waterway, an inward gradient may
be maintained for portions of the tidal cycle (i.e., during increasing tide and high tide, and perhaps the
decreasing tide); however, it will likely be impossible to maintain an inward gradient over the entire tidal
cycle. In general, the hope is that the multiple measurements will indicate a net flow of ground water
directed inward. This information should not only be used to evaluate capture but should also be used to
potentially optimize the remedy in the future.
Over time, the site team may be able to correlate successful or adequate capture with certain extraction
rates. This correlation could be useful in helping the site team determine if capture is adequate. If the
extraction rate falls below the minimal rate where capture is adequate, then the site team will know that
increased extraction is necessary, perhaps through rehabilitation of existing wells or the addition of new
wells. Although this type of correlation is helpful, the RSE team does not suggest that the team rely on it
for evaluating capture. Overtime, hydrogeologic conditions may change due to regional changes in
infiltration or pumping, which would invalidate the capture/extraction rate correlation.
28
-------
The estimated capital cost for installing the new piezometers is approximately $90,000 based on an
estimated cost of $100 per foot drilled and should include associated work plans, health and safety plans,
and reporting. The RSE team notes, however, that the responsible party could better estimate this cost
based on the previous costs of installing monitoring wells at this site. The annual costs of implementing
this well-pair monitoring approach would be approximately $36,000 in the first year and $18,000 in the
following years. For each gauging event, this assumes approximately $3,000 per 8-hour day for three
technicians and equipment multiplied by three to account for the rotating shifts over the course of 24
hours. Compared to the estimated annual cost of $125,000 per year (assuming one half of the $250,000
sampling/reporting cost), this recommendation represents a net savings. Additional savings could be
realized if the site team is successful in using pressure transducers to automatically collect information
over the tidal cycle during each event. Historically, the site team has had difficulty with automated water
levels due to corrosion and fouling; however, there may be some locations where transducers can be used
during the event to eliminate the need for multiple shifts throughout a 24-hour period.
In Table 7-1, the first year cost is considered a capital cost of $36,000 and the other years are assigned an
annual O&M cost of $ 18,000. Recommendation 6.1.2 already accounts for the reduction of $ 125,000
associated with discontinuing the current well gauging approach; therefore, implementation of this
recommendation is considered a cost rather than a cost savings.
6.1.5 DELINEATE THE VOC PLUME TO THE NORTH AT THE 100-FOOT DEPTH INTERVAL
Ground water quality data from monitoring wells 40A-100 (7,100 ug/L vinyl chloride in August 2002)
and 41-138 (12,000 ug/L vinyl chloride in November 1996 and 63 ug/L vinyl chloride in August 2002)
suggest that contamination is not delineated to the north. As stated previously in this report, 40A-100 is
likely within the capture zone of A-2A and contamination 41-138 is likely diluted by injection from E-l.
The site team should discuss the merits of delineating beyond 40A-100 and 41-138 to determine the true
extent of contamination and the potential need to expand the extraction network to the north. The RSE
team estimates that a total of four additional monitoring wells at approximately 100 feet will be required
to delineate contamination in these areas.
The estimated cost for implementing this recommendation is approximately $50,000 based on an
estimated cost of approximately $100 per foot drilled. This estimated cost includes the cost for
developing a work/health and safety plans, conducting two rounds of sampling, and reporting the results.
The RSE team notes, however, that the responsible party can better estimate these costs based on
previous costs for monitoring well installation and sampling.
6.1.6 CONSIDER INSTITUTIONAL CONTROLS AND A PRELIMINARY VAPOR EVALUATION
To prevent potential future exposures to site-related contamination, the site team should consider
implementing institutional controls for the site and surrounding properties. The controls could likely
consist of notices attached to the relevant property deeds describing the contamination and the potential
routes of exposure. The purpose is to notify current and future property owners of the existing
contamination so that workers are not exposed to contaminated ground water, soil, or vapor.
The site team should also consider evaluating the potential for vapor intrusion into site buildings.
Significant VOC contamination exists at shallow depths, and some buildings on or off site may be
affected. The RSE team suggests that the site team identify all buildings overlying the plume and
determine the ventilation information, construction, and uses of those buildings. This collected
information should then be reviewed to determine the need for further vapor intrusion evaluations.
29
-------
The RSE team estimates that implementing the institutional controls and the preliminary vapor intrusion
work might require a one-time cost of approximately $30,000.
6.2
RECOMMENDATIONS TO REDUCE COSTS
6.2.1
ADD A SECOND AIR STRIPPER IN SERIES AND ELIMINATE THE USE OF GAC
The ability of the current system to remove mass is limited by the efficiency of the air stripper. The
relatively low efficiency requires frequent GAC changeouts and a reliance on the GAC to meet discharge
standards, especially the 1,1,2-TCA standard.
The existing air stripper VOC removal efficiency is lower than the design values causing the GAC
vessels to remove VOCs in order to meet the discharge requirements. This has resulted in an increased
cost for GAC changeouts and an increased reliance on manual GAC unit backwashes, which require
substantial operator time. The air stripper efficiency should be improved with a goal of bypassing the
GAC vessels mainly to eliminate change out costs. Bypassing the GAC vessels will also reduce operator
labor and reduce generated solids.
Vendor literature indicates that additional modules might be added to the existing system to improve
efficiency, but with increased extraction and likely an increased influent concentration, this will not
likely be sufficient. The RSE team suggests the use of two tray type strippers arranged in series. The
first unit would remove over 99% of the contaminant mass and send that mass to the catalytic oxidizer.
The second unit would have a very high air flow rate and would act as a polishing stage for the more
recalcitrant contaminants, such as 1,1,2-TCA.
Both air strippers should be rated for at least 200 gpm. The existing air stripper is likely sufficient to
serve as the first air stripper, but, for illustration purposes, the RSE team presents this dual air stripper
approach with two new Carbonair models. The RSE team is not necessarily recommending
Carbonair models. Other air stripper brands, including North East Environmental Products, could
provide similar results. According to the Carbonair software, the first stripper should have six trays and
an air to water ratio of 24:1 (i.e., 650 cfrn of air flow for 200 gpm of water flow). The second unit should
also have six trays but should have an increased air to water ratio of 67:1 (i.e., 1,800 cfrn of air flow for
200 gpm of water flow). The first unit should only require a 10 HP blower, and the second unit would
require a 20 HP blower. The following table provides the stripper influent concentrations from June
2003, the calculated effluent for each constituent from the first air stripper, the calculated effluent for
each constituent from the second air stripper, the mass removal rate for the second air stripper, and the
effluent criteria.
Parameter
Vinyl chloride
1 , 1 -Dichloroethy lene
Methylene Chloride
t- 1 ,2-Dichloroethylene
June 2003
Stripper
Influent
(ug/L)
3,032
100
600
500
1st Air
Stripper
Effluent
(ug/L)
1.6
0.2
105
15
2nd Air
Stripper
Effluent
(ug/L)
0.0
0.0
1.3
0.0
NPDES
Criteria
(ug/L)
NA
5
185
NA
2nd Air
Stripper
Mass
Removal
(Ibs/day)
0.00
0.00
0.25
0.04
30
-------
Parameter
c- 1 ,2-Dichloroethylene
Chloroform
Carbon tetrachloride
Trichloroethylene
1 , 1 ,2-Trichloroethane
Tetrachloroethylene
1 , 1 ,2,2-Tetrachloroethane
Totals
June 2003
Stripper
Influent
(ug/L)
17,280
3,020
100
11,908
100
1,304
100
38,044
1st Air
Stripper
Effluent
(ug/L)
927
182
0.3
140
40
9.5
76
1,495
2nd Air
Stripper
Effluent
(ug/L)
2
0.5
0.0
0.1
3.2
0.0
35
41.9
NPDES
Criteria
(ug/L)
NA
75
10
440
5
50
180
NA
2nd Air
Stripper
Mass
Removal
(Ibs/day)
2.22
0.44
0.00
0.34
0.09
0.02
0.10
3.50
As demonstrated in the table, these two air strippers would allow the treatment plant to reliably meet all
of its discharge criteria, even at the extremely high influent concentrations recorded in June 2003 and an
extraction rate of 200 gpm. A stripper with a higher air flow rate could also be used for the first unit
(such as one that provides a 34:1 air to water ratio with 900 cfm of air flow) provided that the current or
eventual replacement catalytic oxidizer can handle the additional flow. Based on the data provided in the
table, the off gas from this polishing stage would contain less than 3.5 pounds of total VOCs per day (of
which 2.2 pounds per day is cis-l,2-DCE). Under any national regulations, this emission rate should not
require treatment; however, local air regulations may differ and should be considered by the site team.
Using the current air stripper as the first unit, the ground water effluent concentration would likely be
much less than that in the table, and the emission rate from the second/polishing unit would likely be
under 1 pound per day.
With this approach, the current system for treating off gas can be used, and the tray strippers would likely
only require monthly or bimonthly cleaning. The GAC and the time consuming backwashing could be
eliminated, which would substantially reduce operator labor and offer minor savings associated with
sludge disposal. The logical choice is to use the current air stripper as the first unit in this two-air
stripper system, and to replace that unit only when necessary. Making this choice, the second air stripper
could be purchased for about $50,000. With installation, the capital investment would be approximately
$150,000. Eliminating GAC vessel changeouts would save about $80,000 per year. The electrical costs
would be increased by approximately $10,500 per year (20 HP at $0.06 per kWh) unless the current air
stripper effluent pump is downsized. The monitoring of the system influent and the influent and the
effluent from the first stripper could be reduced to monthly. The system effluent monitoring would need
to remain at weekly per the discharge permit. Therefore, the total number of VOC process samples
(including QA/QC) should be approximately 152 rather than 312, representing a savings of at least
$16,000. Operator labor could possibly be reduced to one full time employee, which would represent a
cost savings of approximately $75,000 per year.
6.2.2
MODIFY GROUND WATER MONITORING PROGRAM
Ground water cleanup at this site will take several decades; therefore, sampling within the plume on a
frequent basis provides extraneous information. The ground water monitoring program should be
tailored to evaluate capture on a quarterly or semi-annual basis. Monitoring for plume cleanup can be
conducted on an annual or biannual basis from a limited number of wells. If the site nears cleanup in the
31
-------
future, the number of wells and frequency can be increased to verify that cleanup is achieved throughout
the site.
Ground water quality monitoring for capture is generally effective when the well is outside of the capture
zone. When capture is complete, concentrations in wells outside of the capture zone decrease to
background concentrations if the well is contaminated, or remain at background concentrations if the
well is not contaminated. At this site, water quality at monitoring wells along the waterway will not help
evaluate capture because they would likely be in the capture zone.
For this reason, the RSE team recommends the following monitoring program, which is focused on
capture while cost-effectively monitoring the relatively slow progress toward site cleanup. This program
represents a decrease in the number of samples/analysis from approximately 250 per year to
approximately 100 per year with greater emphasis on evaluating capture and adequate monitoring of
progress toward cleanup. This monitoring program could be modified with time.
Semi-annual Sampling for Capture Evaluation
25-foot*
5-25
12-25
25-25
40-25
42-25
43-25
44-25
55-25
56-25
57-25
50-foot*
5-50
12-45
25-50
35-50
40A-50
43-50
55-50
56-50
57-50
100-foot*
5-100
12-160
35-100
31-100
56-100
57-100
Two
additional
wells that
delineate to
the north
(see Section
6.1.4)
Annual Sampling for Progress toward Cleanup
25-foot*
4R-25
14-25
53-25
50-foot*
Tl-50
4-45R
36-50
53-50
100-foot*
Tl-100
4-83R
53-100
11-100
15 to 20 additional wells selected by the site team
The cost savings from implementing this modification should be substantial, but it is difficult for the RSE
team to provide an estimate because the RSE team does not know the approximate breakdown in
sampling costs between the analytical sampling, well gauging, and bi-annual slug testing. The RSE team
also does not have an accurate value for the analytical costs. Assuming the ground water quality
sampling costs are approximately half of the reported $250,000 for all sampling and reporting,
implementing this scope of work should provide approximately $75,000 per year in savings (a reduction
of 150 samples out of 250). Likewise, analytical costs associated with long-term ground water quality
monitoring should be reduced by 40%.
6.3
MODIFICATIONS INTENDED FOR TECHNICAL IMPROVEMENT
6.3.1
IMPROVE QUARTERLY AND ANNUAL PERFORMANCE REPORTS
The quarterly and annual performance evaluation reports by CRA contain no information on treatment
system operation, performance, major maintenance or downtime. They also do not discuss the injection
system flow rate. The format of these reports should be revised to include this information. The reports
32
-------
should also be modified to discuss the containment/capture as detailed in Sections 6.1.2, 6.1.3, and 6.1.4.
The RSE team estimates that implementing this recommendation might cost an additional $10,000 per
year with an extra $5,000 for the first report, with the change in cost primarily due to the addition of the
treatment plant information.
6.3.2 CONSIDER ABANDONING AND NOT REPLACING THE SLUDGE DRYER WHEN IT FAILS
Solids removal and associated sludge handling are necessary at this site, especially when extraction wells
closer to the high pH (and highest VOC) areas are pumped. However, the volume of solids removed of
about 200 Ibs/day of sludge is significantly less than the original design value of 13 tons/day as dry
solids. The current system includes a plate and frame filter press that replaced a belt press (an effective
optimization effort by the site). The original sludge drier is still used even though it would not be
warranted unless it were already in place. Since it requires relatively little energy, it is reasonable to
continue using it. Removal of the unit should be considered if any major maintenance is required. The
plate and frame press use should continue because the alternative of collecting the sludge in a thickener
tank and having it processed by a mobile service is not feasible due to the large volume of dilute sludge.
Maintaining the filter press also allows the flexibility of handling additional solids if extraction wells are
added. The cost savings from implementing this recommendation is likely negligible.
6.4 CONSIDERATIONS FOR GAINING SITE CLOSE OUT
6.4.1 CONSIDERATIONS FOR AN EXIT STRATEGY
Although P&T will take decades (or possibly centuries) to reach clean-up goals, the P&T approach has
the potential to provide plume capture that cannot be duplicated by other technologies. In many P&T
systems, contaminant removal technologies such as sparging, bioaugmentation/biostimulation, chemical
oxidation, thermal technologies and excavation can be used for source removal to lessen the operating
life span of a P&T system. However, this site has a plume that occupies a very large volume of the
subsurface (over 20 acres and over 100 feet deep) with extremely high VOC concentrations and
substantial infrastructure at the surface. In addition, for large portion of the plume, the water table is
close to the surface preventing the use of soil vapor extraction that might accompany a sparging system.
It is doubtful that other technologies could be applied to the site and provide any meaningful remediation
(i.e., remediation that would significantly shorten the time to cleanup) for less than $50 million because
removing only a portion of the source material would not substantially reduce the duration of the remedy.
The limitations of a few common technologies are provided below.
Sparging/SVE or ground water recirculation: Fouling problems would likely occur and
insufficient room is available to extract the sparged vapors in the unsaturated zone. Even if
sparging were possible, about 400 sparge wells would likely be required at a 25 foot radius of
influence with air lines and vapor extraction lines, and this would likely only apply to the 25-foot
and possibly the 50-foot depth intervals. This installation is infeasible at an operating facility.
Bioremediation/Bioaugmentation: Distributing the amount of nutrients throughout the subsurface
would be impracticable based on the water injection problems as has been noticed with the
injection wells. Also, the concentrations indicate significant residual DNAPL mass which
would not degrade readily and would require a substantial amount of nutrient that is difficult to
inject.
33
-------
Chemical Oxidation: As with bioaugmentation and air sparging, distributing oxidizing chemicals
would require a substantial subsurface network that is infeasible at an operating facility.
Millions of pounds of chemicals would be necessary due to the substantial contaminant mass,
and injection would be hampered by fouling.
Thermal Technologies: Thermal heating technologies are extremely expensive due to energy use.
The RSE team has seen six-phase heating pilot tests with approximate costs of $150 per cubic
yard and higher. This unit cost would translate to over $100 million for only a portion of the
plume. In addition, an SVE system would be required to capture the VOC off gas from heating
the saturated zone. As mentioned previously, SVE would not be effective with the limited
unsaturated zone.
Rather than experimenting with these other technologies, the RSE team would support the installation
and operation of additional extraction wells to enhance capture and increase mass removal. The capital
costs would likely be much smaller than the above options, and the annual costs for P&T O&M will be
required for containment regardless. The RSE team notes that other technologies may develop over time
that may be applicable to this site. These technologies could be evaluated at that time.
6.4.2 ADDRESS PLUME AND SOURCE CONTROL PRIOR TO INSTALLING A SEDIMENT CAP
At the time of the RSE site visit, sediment removal along the Hybelos Waterway bank in the 5106 area
had been accomplished and a sediment cap is planned. During the site visit, EPA Superfund expressed
concern that the cap would be recontaminated by ground water if it is installed or adversely affected due
to solids fouling associated with the high pH. The RSE team shares this concern and recommends
delaying cap installation until the pumping rate is increased as suggested in Section 6.1.1 and the degree
of plume capture has been evaluated as described in Sections 6.1.2, 6.1.3, and 6.1.4. Subsequent to the
RSE site visit, EPA noted that the site team (including OCC) agreed that the cap would not chemically
isolate sediments or discharging ground water that are contaminated by VOCs.
If containment along the waterway cannot be provided through extraction alone, as described in Section
6.1.1, other technologies would be required to remove mass along the waterway such that the plume is
drawn back from the waterway and can be captured by extraction. This mass removal would be difficult
to achieve as demonstrated by the dredging activity where some hot spots could not be reached due to
infrastructure and the steep slope of the embankment. These same types of limitations would also likely
apply to other technologies that may be applied nearby. Although some mass might be removed, other
mass would likely persist in some areas. With respect to WMU H, further excavation would likely
provide little benefit, since the area was already excavated to the water table. The majority of mass has
likely already migrated downward into the saturated zone and removing only a portion would not be
beneficial if substantial mass is left in place.
Therefore, it appears impracticable to guarantee that a new sediment cap will not become contaminated
unless a constant inward gradient can be established. As improvements are made to the P&T system, the
Superfund Program can consider the degree of capture and the benefits of installing the cap or selecting
another remedial approach. The site team may determine that the benefits of covering the existing hot
spots may greatly outweigh the potential drawbacks of limited recontamination.
34
-------
6.5 SUGGESTED APPROACH TO IMPLEMENTATION
The RSE team would recommend proceeding with recommendations 6.1.2, 6.1.3, 6.1.4, 6.1.5, 6.2.1,
6.2.2, 6.3.1, and the analysis part of 6.1.1 immediately and concurrently. Evaluation of capture under
current conditions using the recommended approaches might better inform the site team of areas where
additional extraction is most needed. While capture under current conditions is evaluated, the treatment
system can be improved to accommodate the increased flow. Once the treatment system can better
accommodate the increased capacity, recommendation 6.1.1 can be fully implemented and capture can be
reevaluated on an ongoing basis as recommended.
At this point, which may be one or two years from the time of the RSE site visit, Superfund could better
evaluate the conditions for sediment remediation given the improved understanding of ground water flow
under enhanced extraction/capture (6.4.2). Recommendation 6.1.6 should be implemented prior to full-
scale operations at the Pioneer Americas plant, and recommendations 6.3.2 and 6.4.1 are for long term
consideration.
The effectiveness recommendations provided in this report focus on the VOC plume but consider the
elevated pH. As described in Section 6.1.1, the RSE team suggests approaching the VOC plume by
increasing extraction in an attempt to create an inward gradient that captures the plume. This inward
gradient might also capture the elevated pH. The RSE team recommends that extraction take place in
areas with a pH less than 9.5 to reduce problems with fouling. Over time, the elevated pH may migrate
toward the extraction wells, but historical data suggests that migration of the elevated pH is minimal, and
the natural buffering capacity of the soil should offer a degree of protection to these extraction wells. If
the site has to inject reagents to neutralize the elevated pH to protect the extraction wells, the mass of
reagents needed and the degree of fouling might be lower than attempting to neutralize the pH source
area.
35
-------
7.0 SUMMARY
The observations and recommendations contained in this report are not intended to imply a deficiency in
the work of either the system designers or operators but are offered as constructive suggestions in the
best interest of the EPA and the public. These recommendations have the obvious benefit of being
formulated based upon operational data unavailable to the original designers.
Recommendations are provided in all four categories: effectiveness, cost reduction, technical
improvement, and site closeout. Recommendations for effectiveness are primarily focused on enhancing
capture and improving the plume capture evaluations. Recommendations for cost reduction include
adding a second air stripper in series to eliminate the need for GAC and solids handling. They also
include modifications to the process monitoring and ground water monitoring programs. The technical
improvement recommendations include improving the performance report and avoiding replacement of
the sludge drier when it eventually fails. The recommendations for site closeout document the RSE
team's positions on the use of alternative technologies at this site and the implications of the RSE
recommendations on the remediation of sediments under the Superfund Program.
Table 7-1 summarizes the costs and cost savings associated with each recommendation in Sections 6.1
through 6.4. Both capital and annual costs are presented. Also presented is the expected change in life-
cycle costs over a 30-year period for each recommendation both with discounting (i.e., net present value)
and without it.
36
-------
Table 7-1. Cost Summary Table (Part 1 of 2)
Recommendation
6.1.1 Eliminate Inj action
System and Use Extraction to
Enhance Capture
6.1.2 Discontinue Current
Approach of Using
Potentiometric Surface Maps
to Evaluate Capture
6.1.3 Conduct and Report
a Water Budget Analysis
6.1.4 Install Additional
Monitoring Wells and Use
Pairs of Monitoring Wells to
Demonstrate Capture with
Inward Gradients
6.1.5 Delineate the VOC
Plume to the North at the
100-foot Depth Interval
6.1.6 Consider Institutional
Controls and a Preliminary
Vapor Evaluation
6.2.1 Add a Second Air
Stripper in Series and
Eliminate the Use of GAC
6.2.2 Modify Ground
Water Monitoring Program
6.3.1 Improve Quarterly
and Annual Performance
Reports
6.3.2 Consider
Abandoning and not
Replacing the Sludge Dryer
when it Fails
Reason
Effectiveness
Effectiveness
Effectiveness
Effectiveness
Effectiveness
Effectiveness
Cost
Reduction
Cost
Reduction
Technical
Improvement
Technical
Improvement
Estimated
Additional
Capital
Costs
($)
$285,000
negligible
negligible
$126,000
$50,000
$30,000
$150,000
$0
$5,000
$0
Estimated
Change in
Annual
Costs
($/yr)
negligible
($125,000)
negligible
$18,000
negligible
negligible
(>$ 160,000)
(>$75,000)
$10,000
unqualified
decrease
Estimated
Change
In Life-cycle
Costs
(S)1
$285,000
($3,750,000)
negligible
$666,000
$50,000
$30,000
(>$4,650,000)
(>$2,250,000)
$305,000
unqualified
decrease
Estimated
Change
In Life-cycle
Costs
($)2
$285,000
($2,018,000)
negligible
$417,000
$50,000
$30,000
(>$2,432,000)
(>$1,211,000)
$166,000
unqualified
decrease
37
-------
Table 7-1 Cost Summary Table (Part 2 of 2)
Recommendation
6.4.1 Considerations for an
Exit Strategy
6.4.2 Address Plume and
Source Control Prior to
Installing Sediment Cap
Reason
Site Closeout
Site Closeout
Estimated
Additional
Capital
Costs
($)
$0
$0
Estimated
Change in
Annual
Costs
($/yr)
$0
$0
Estimated
Change
In Life-cycle
Costs
(S)1
$0
$0
Estimated
Change
In Life-cycle
Costs
($)2
$0
$0
Costs in parentheses imply cost reductions.
1 assumes 30 years of operation with a discount rate of 0% (i.e., no discounting)
2 assumes 30 years of operation with a discount rate of 5% and no discounting in the first year
-------
FIGURES
-------
FIGURE 1-1. SITE LOCATION MAP.
PLANT
NORTH
TRUE
NORTH
FORMER OCC FACILITY
MARIANA PROPERTY
US NAVY & PORT OF
TACOMA PROPERTIES
(721 Alexander Avenue)
NOT TO SCALE
(Note: This figure is taken from Embankment Area Rapid Groundwater pH Assessment, Figure 2.1, CRA, 2003).
-------
FIGURE 1-2. THE FORMER OCC FACILITY AND CURRENT WELL LOCATIONS.
40AA
E-8
JWMU-C
°4
WMU-Hy
:'" L- 15
iHT^h
F-4A f 5 _6
T6 A
F-7
F-2
AT2
A
T4
AT3
^-1
A'2
D-41
PORT OF TACOMA
- 1
|D-5|
A-3
A-4
A-6
4RA
OCCIDENTAL
CHEMICAL
CORPORATION
'^
A21
D-3 TREATMENT
A53 PLANT
: I a
C-1 C-2
'C-3
f 12 f * 12A ^ T
^ ? 4 T t
25A
A
25
B-1
PORT OF TACOMA
A34 A31
36 A
B-2
B-3
.35
57
A
A42
J6
A19
c-4 c-5 c-6:-:'c-7
TRUE
NORTH
8 A
0 275
SCALE IN FEET
550
V
PLANT
NORTH
A
17
Oi,
23,
A45
48 A
51,
29 ;
: 52"
LEGEND
EXISTING MONITOR WELL
NEST LOCATION AND NUMBER
INJECTION WELL
EXTRACTION WELL
* "-FENCELINE
i i -RAILROAD
(Note: This figure is taken from Figure 2.1, Corrective Action Monitoring Program, CRA, December 2000).
-------
FIGURE 1-3. PLANT EAST/WEST GEOLOGIC CROSS-SECTION.
PLANT W
PLANT E
-APPROXIMATE
TIDAL MIXING ZONE
1
UJ
QD
O
O
m
O
I
>
X
O
ce
Q-
Q.
FLOWING SEEPS
MILKY SEEPS
- » ;.:V"v - :-',.'-'/-'--, "--.- '--' :.W:V v :--
#25 = 1.022*
25-FT ZONE
GROUNDWATER/
SURFACEWATER
INTERFACE
50-FT :ZONE
LEGEND
APPROXIMATE
Y GROUNDWATER
ELEVATION
APPROXIMATE
Y WATERWAY
ELEVATION
£ SEEP
ow DENSITY OF
H WATERWAY WATER
p50
t
i
DENSITY OF
GROUNDWATER
AND AQUIFER
ZONE
GROUNDWATER
FLOW DIRECTION
VERTICAL
GROUNDWATER
FLOW DIRECTION
MONITORING WELL
OR INJECTION
WELL
APPROXIMATE SCALE IN FEET
Hll SAND
EH SILTY SAND
Lc3F|LL
l"7^] GRAVEL
| | HIGH "pH BLOB"
LOW SALINITY
WATER
HIGH SALINITY
WATER
(Note: This figure is a re-creation of Figure 2 from the Site Characterization Report, Tetra Tech, June 2003).
-------
FIGURE 1-4. EXTENT OF CONTAMINATION AT 25 FEET BELOW GROUND SURFACE - AUGUST 2002
E-1 E-2
E-3
E-8
o
40AQ L'
^40
PORTOFTACOMA
I
D-5
A-3
A-4
A-5
f/0(
12*
A-6
12A
PB-4
H i 1... E. B 0 S A* A i c. n! W A *
: \A/MI i r WMtMtT-f^r^i I
-; WMU-C -i : n t ~ l
c~9* f-3\ fi4 F-4A *^T5 "* F 7 * "~"N r'\F~9" ^*
4 TC A^ ' *«J 1 \ / I ^
\ ID i i x / v,^ rf*..^
: : f F-2 T, ' V - : ^_-'
I I 1|3*« «4j2 xx % - i: ~xOO
! l 20A / - it:'- }^'^
, r *J r / : - i r . _^
^ T4 1 ; I : ; r
OCCIDENTAL ; : ji L
'VOA 4Rfl| CHEMICAL ! if^i
'\. r»ADDADATiAM : : /pvr/in]
v._ OUKrUKAMUN (n-t/fwy
DT \ - : .' . .;Al9 -
« \ : ; -.: -iv^.
1 \ ^ Uy. t : ^
w - ]Js,,,..s-.i^» - -H^,,»..^^.^.Ws,s-.%=|s| ,..;.. f
, 1 F:::H 21 ^ ,':,': * 5^
-Tiy - *IJD X^>i4tx /
/3^rSMENT 1 ' ?*i ': ' ^
A,, PLANT 1 i , '
t * . T
:*~ | A f
!"*-2 f 54 1 ~" '\ 1 1 1 /
I \ I * r. v iLJ
1 \' T 4 : : . JKk. i ,
* \ t f : . : RR ifess i
17* 1 * « H "'"""" ' ^^ ^CN* 1
i / f i*,..,%....ft.. , ..:.::: Xf>N !
f f J." ^ ^-v.^x . |
L" ' , «C"»2, « l^e-3C'4 C'SA ^6 H i ^7 C-B ^9^ ^
')1
"
/
/
/
/
^
-
S /-N
^
/
1
f
[
in ^
^
^'"S f |
\J \,J
v,,,y ^y
x'*"
1
1
f
..-....,"" ,
17
49*
\
\
\
1 s.
32.
'25
ALL COCs BELOW
APPLICABLE STANDARDS
iB-1
k56
43
~ ABOVE STANDARDS BUT O
U TOTAL COCs LESS THAN (VC/6.3)
100 ug/L
p, TOTAL COCs BETWEEN
w 100 ug/L AND 1,000 ug/L
gU TOTAL COCs BETWEEN
^ 1,000 ug/L AND 10,000 ug/L
31
B-2
(Trans/33)
O
42
B-3
O35
i7
TRUE
NORTH
(VC/6.3)
TOTAL COCs OVER 10,000 ug/L
COC THAT IS ABOVE STANDARDS AND
THE CONCENTRATION IN ug/L
O
44
V
PUMMT
NORTH
45,
18
23,
30
1/
LEGEND
EXISTING MONITOR WELL
NEST LOCATION AND NUMBER
INJECTION WELL
EXTRACTION WELL
FENCELINE
RAILROAD
INTERPRETED EXTENT OF WATER
WITH pH > 10"
o
275
550
SCALE IN FEET
(Note: Base features taken Figure 2.1, Corrective Action Monitoring Program, CRA, December 2000 and data taken from Annual Performance
Evaluation Report, October 2001 through September 2002, CRA, November 2002).
-------
FIGURE 1-5. EXTENT OF CONTAMINATION AT 50 FEET BELOW GROUND SURFACE - AUGUST 2002
CJ
(VC/6.3)
40
PORTOFTACOMA
A-2
(TCE/98)
t^
^
o!
taB-4
O
,_
43
ALL COCs BELOW
APPLICABLE STANDARDS
ABOVE STANDARDS BUT
TOTAL COCs LESS THAN
100 ug/L
TOTAL COCs BETWEEN
100 ug/L AND 1,000 ug/L
TOTAL COCs BETWEEN
1,000 ug/L AND 10,000 ug/L
TOTAL COCs OVER 10,000 ug/L
COC THAT IS ABOVE STANDARDS AND
THE CONCENTRATION IN ug/L
! WMU-C
510-
1W CK c
-?5A
O
25
3,
36
56
n
|5| "
O(VC/3.4)
±42 35
TRUE
NORTH
A
17
1 | T6 * X ^ ;^ ^ * j V-/ lx_/
^tfn I2 >v ; --"'NOO
14 \j OCCIDENTAL ; ~ _ |:
t rHFMirAi
4 ~ v CORPORATION . -U
\ " ' : r~\g
\ -.- LJ ..-
\ ^^"; I
j ---^NV ' ^^" ^ -i ' '-
V i , ,..j i i"**"*^ :
r« -K 43ROUNDWATER U^ .^""t^^
TREATMENT * : : i i 7
|53 PLANT I U" "
A j ; X
ij * j .;.-;. -V-- y ,- "*«y !""""
i ; '- TT i ; 55 ^x-
. ! : ! ! : ^ N'-«!
=^
r^c/
V-
?j;
f
10"
PUMsIT
NORTH
0
275
SCALE IN FEET
550
(Note: Base features taken Figure 2.1, Corrective Action Monitoring Program, CRA, December 2000 and data taken from Annual Performance
Evaluation Report, October 2001 through September 2002, CRA, November 2002).
-------
FIGURE 1-6. EXTENT OF CONTAMINATION AT 100 FEET BELOW GROUND SURFACE - AUGUST 2002
4V~^
(VC/92)
1-2
.14 r 15
j|j F-5.TS|Q.
PORTOFTACOMA "
D-5
A-3
V*r
/ » i
/ 1 1_
: \P
i \ Am
1 'p
«T* \
V°A
-f
i Ii
9 F^,
F-2,AT2
-10MT1 20A
A
x T4
\
_^___
«*,
^
j^-s ,||,||....|
F-4 r j« A «
F-
T6A
OCCIDENTAL
CHEMICAL
^ CORPORATIONx
^gp^
F-7 y
A-4
I
-A, X »f
| ;
1 /"4i;i : :
A-51 /' :
1 :
1 A-6 .
1
a*' f «B-4
T A T
f 1 o * *
.J2J U?*....:. f
"-
D-3
: ^~\5
i ^
-D-2
i: 7
1 "
32A
\\ -
.- «GROUN0S/VATER
TREAIMENT
54
..!frrtB-i^
2SA
24
25
p> ALL COCs BELOW
^ APPLICABLE STANDARDS
_ ABOVE STANDARDS BUT
U TOTAL COCs LESS THAN
100 ug/L
f-N TOTAL COCs BETWEEN
^ 100 ug/L AND 1,000 ug/L
ps TOTAL COCs BETWEEN
^ 1,000 ug/L AND 10,000 ug/L
34
36
B2
x->.
(vc/18)
56
C
44
(VC/6.3)
TOTAL COCs OVER 10,000 ug/L
COC THAT IS ABOVE STANDARDS AND
19
55
TRUE
NORTH
-fcUU-
V
x
PLANT
NORTH
THE CONCENTRATION IN ug/L
C10
«A
17
49J
A 3
LEGEND
EXISTING MONITOR WELL
NEST LOCATION AND NUMBER
INJECTION WELL
EXTRACTION WELL
*-*-*- FENCELINE
-t < i- RAILROAD
INTERPRETED EXTENT OE WATER
WITH pH > 10"
0 275
1^-
SCALE IN FEET
550
(Note: Base features taken Figure 2.1, Corrective Action Monitoring Program, CRA, December 2000 and data taken from Annual Performance
Evaluation Report, October 2001 through September 2002, CRA, November 2002).
-------
FIGURE 1-7. EXAMPLE POTENTIOMETRIC SURFACE MAP FOR THE 25-FOOT INTERVAL (JANUARY 2002) SUBMITTED AS PART OF
AN ANNUAL PERFORMANCE EVALUATION REPORT.
41,
3.00
(3.27}
* \
}
/E-3
(2.96)
'"/ ,
_
2.50
OCCIDENTAL
CHEMICAL
CORPORATION
0
^'
PLANT _£\
NORTH
275
SCALE IN FEET
550
A 44
(1.94)
A-BRANCH 51 GPM
B-BRANCH 77 GPM
C-BRANCH 29 GPM
D-BRANCH 13 GPM
110 GPM
' ' 50A 1
181*:. V
L f
tiBA
2.00'
(1.77)
LEGEND
EXISTING MONITOR WELL
NEST LOCATION AND NUMBER
INJECTION WELL
EXTRACTION WELL
FENCELINE
RAILROAD
INTERPRETED EXTENT OF WATER
WITH pH > 10"
GROUNDWATER CONTOUR
FRESH WATER EQUIVALENT
ELEVATION (FT NGVD)
(Note: Information for this figure is taken from the Annual Performance Evaluation Report, October 2001 through September 2002, CRA 2002).
------- |