REMEDIATION SYSTEM EVALUATION
HIGGINS FARM SUPERFUND SITE
Princeton, New Jersey
Report of the Remediation System Evaluation,
Site Visit Conducted at the Higgins Farm Superfund Site
21-22 August 2003
Prepared by US Army Corps of Engineers
Hazardous, Toxic, and Radioactive Waste Center of Expertise
For the US Environmental Protection Agency
US Army Corps or Engineers
US Environmental Protection Agency
-------
This page is intentionally left blank.
-------
Office of Solid Waste EPA 542-R-04-034
and Emergency Response May 2004
(5102G) www.epa.gov/tio
clu-in.org/optimization
Remediation System Evaluation
Higgins Farm Superfund Site
Princeton, New Jersey
-------
This page is intentionally left blank.
-------
Executive Summary
The Higgins Farm Superfund Site is located in a rural portion of Franklin Township, Somerset County, New Jersey,
approximately 4 miles northeast of Princeton, New Jersey. The site occupies approximately 75 acres southeast of
New Jersey State Highway 518 and is largely used as pasture.
Though used as a farm and pasture land, the Higgins Farm Site was at times also used for the disposal of hazardous
wastes, including hazardous substances. In December 1985, the Franklin Township Health Department reported to
the New Jersey Department of Environmental Protection (NJDEP) that elevated levels of chlorobenzene existed in a
potable well adjacent to the site. NJDEP and EPA later provided alternate drinking water supplies for homeowners
with potentially impacted wells as part of the first Record of Decision (ROD) dated September 24, 1990. NJDEP
also investigated and discovered multiple drum burial and disposal areas throughout the site.
Mr. Higgins attempted to remove drums from a portion of the property. During excavation activities, some of the
containers, including drums were punctured and their contents spilled onto the ground. Liquids were pumped from
the excavation pit to a holding tank and visibly contaminated soils were placed in roll-off containers.
NJDEP sampled residential wells near the site and discovered that three often residential wells and groundwater
were contaminated with volatile organic compounds (VOCs). Later NJDEP personnel inspected Higgins Farm and
collected soil samples from the site, including the excavation pit area. Analysis of these samples indicated the
presence of volatile organic compounds, pesticides, metals, dioxins and furans in the soils at the site. The next year
NJDEP requested the EPA assume the lead role in the cleanup. EPA has performed a number of activities between
1987 and 1992.
The Remedial Investigation (RI)/Feasibility Study (FS) was prepared in 1992 and a second ROD to address site
ground water contamination was signed in September 1992. The selected alternative included the installation of
extraction wells and construction of a treatment plant that would include metals treatment by precipitation and ion
exchange and removal of volatile organic compounds by air stripping. A design was prepared in 1994 and
construction began in 1995. The groundwater extraction system and treatment plant began operation in May 1998.
Ground water conditions have been characterized at two general depths; shallow (less than 20 feet depth), and deep
(20- 200 feet depth). Shallow monitoring wells are constructed with casing and five to ten feet of well screen.
Deep monitoring wells installed during the RI have a surface casing into competent bedrock and are completed as
open holes to total depth. As such, the deep monitoring wells represent conditions over very long vertical intervals.
Ground water flow is primarily from the north-central part of the site to the south and southeast. Hydraulic
gradients are typically around 0.015 ft/ft with steeper gradients near the extraction system and on the western part of
the site. At times, the gradients in the central part of the site drop to as low as 0.005 ft/ft. Hydraulic conductivities
are highly variable, particularly in the fractured rock ranging from 2E-3 to 3E-7 cm/sec. Higher conductivities were
typically identified in the upper 70 feet of the deep wells.
The present operations staff has been doing a good job keeping the plant fully operational and in compliance with
permit equivalency. The staff has been improving operations by routinely upgrading and adjusting equipment inside
and outside the plant. Some examples include modifying the extraction well pump controls, increasing pipe and
pump capacity within the plant to increase the hydraulic capacity. The influent contaminant concentrations have
also decreased over the operational life of the plant. The reduction of metals concentrations to below discharge
-------
levels except for iron and manganese, and significant reduction in the VOC concentrations have made many of the
current processes inefficient. More effective process operations were evaluated and recommendations made to
replace the current metals removal system with a manganese greensand filter, and reduce the blower requirements
for the air stripping system.
The Remediation System Evaluation (RSE) identified 9 of the 20 extraction wells as having met cleanup standards
and recommends that pumping from those 9 wells be discontinued. Conversely, to enhance plume capture one
additional well is proposed between existing wells 11 and 12. The RSE evaluated monitoring needs for the site, and
proposed many suggestions to reduce sampling frequency and reporting based on the consistency of the last four
sampling episodes.
A site exit strategy should be identified that sets goals for reduction in concentrations at specific locations based on
the goals for human health and environmental protection. These goals may address achieving specific
concentrations at specific locations, taking specific additional remedial actions if concentrations do not decline, or
discontinuing extraction from certain wells or specific treatment processes if concentrations fall below a certain
level. Example site exit strategy components are offered in section 6.4.2 as a starting point for discussions.
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY i
TABLE OF CONTENTS iii
1.0 INTRODUCTION 1
1.1 PURPOSE 1
1.2 TEAM COMPOSITION 1
1.3 DOCUMENTS REVIEWED 2
1.4 PERSONS CONTACTED 2
1.5 SITE LOCATION, HISTORY, AND CHARACTERISTICS 2
1.5.1 SITE LOCATION 2
1.5.2 SITE HISTORY 2
1.5.3 SITE CHARACTERISTICS 3
1.5.3.1 HYDROSTRATIGRAPHY 3
1.5.3.2 SITE GROUND WATER CONDITIONS 4
1.5.3.3 SITE CONTAMINATION 5
1.5.3.4 SITE AND NEARBY LAND USE 5
2.0 SYSTEM DESCRIPTION 6
2.1 SYSTEM OVERVIEW 6
2.2 EXTRACTION SYSTEM 6
2.3 TREATMENT SYSTEM 7
2.4 MONITORING SYSTEM 7
3.0 SYSTEM OBJECTIVES, PERFORMANCE AND CLOSURE CRITERIA 10
3.1 CURRENT SYSTEM OBJECTIVES AND CLOSURE CRITERIA 10
3.2 TREATMENT PLANT OPERATION GOALS 10
3.3 ACTION LEVELS 10
4.0 FINDINGS AND OBSERVATIONS FROM THE RSE SITE VISIT 13
4.1 GROUND WATER EXTRACTION SYSTEM 13
4.1.1 WELL CONDITION 13
4.1.2 PLUME CAPTURE AND REMEDIATION 14
4.1.3 NATURAL ATTENUATION CONDITIONS 15
4.2 TREATMENT EQUIPMENT 16
4.2.1 EQUALIZATION TANKS 16
4.2.2 SOFTENING/METALS REMOVAL PACKAGE 17
4.2.3 FILTER SYSTEM 17
4.2.4 ION EXCHANGE UNITS 18
4.2.5 AIR STRIPPING SYSTEM 18
4.2.6 pH ADJUSTMENT AND WATER SYSTEM 18
4.2.7 CHEMICAL FEED 19
4.2.8 SLUDGE STORAGE AND DEWATERING SYSTEM 19
4.3 MONITORING SYSTEM AND PROGRAM 19
4.4 OTHER OBSERVATIONS 20
4.4.1 WATER DISCHARGE ALTERNATIVES 20
4.4.2 SANITARY FACILITIES 21
4.4.3 OTHER SUGGESTIONS FROM OPERATORS 21
4.5 COMPONENTS OR PROCESSES THAT ACCOUNT FOR MAJORITY OF COSTS 21
4.6 RECURRING PROBLEMS OR ISSUES 22
in
-------
4.6.1 LIMITED HYDRAULIC CAPACITY 22
4.6.2 SLUDGE BUILD-Up IN FLOCCULATION TANK 22
4.6.3 POOR QUALITY BACKWASH WATER FOR ION EXCHANGE UNITS 23
4.6.4 SULFURIC ACID PIPING LEAKS 23
4.6.5 CAUSTIC PIPE PLUGGING 23
4.6.6 EXTRACTION WELL SYSTEM CONTROLS 23
4.6.7 WELL FOULING AND PIPE FAILURE 24
4.7 REGULATORY COMPLIANCE 24
4.8 TREATMENT PROCESS EXCURSIONS OR UPSETS, ACCIDENTAL RELEASES 24
4.8.1 TREATMENT PROCESS EXCURSIONS OR UPSETS 24
4.8.2 ACCIDENTAL CONTAMINANT/REAGENT RELEASES 24
4.9 SAFETY RECORD 25
5.0 EFFECTIVENESS OF THE SYSTEM TO PROTECT HUMAN HEALTH AND THE
ENVIRONMENT 26
5.1 GROUND WATER 26
5.2 SURFACE WATER/WETLANDS AND SEDIMENTS 26
5.3 AIR 26
6.0 RECOMMENDATIONS 27
6.1 RECOMMENDATIONS TO ENSURE EFFECTIVENESS 27
6.1.1 PLUME DEFINITION AND POTENTIAL IMP ACTS TO GROUND WATER USERS 27
6.1.2 SURFACE WATER SAMPLING 27
6.1.3 ON SITE PLUME CAPTURE 27
6.2 RECOMMENDATIONS TO REDUCE COSTS 28
6.2.1 INVESTIGATE BYPASSING THE GROUND WATER TREATMENT PLANT 28
6.2.2 EXTRACTION SYSTEM REVISIONS 28
6.2.3 REDUCE THE SIZE OF AIR STRIPPING SYSTEM 29
6.2.4 EVALUATE ALTERNATIVE MANGANESE REMOVAL TECHNOLOGIES 29
6.2.5 CHANGES IN MONITORING PROGRAM 30
6.2.6 REDUCE OVERSIGHT BY USACE 30
6.2.7 REPLACE TRAILER WASTE HOLDING TANK PUMP OUT CONNECTION 31
6.3 RECOMMENDATIONS FOR TECHNICAL IMPROVEMENT 31
6.3.1 SUGGESTED CHANGES IN WELL MAINTENANCE METHODS 31
6.3.2 PIPING TESTING 31
6.3.3 CONTROL SYSTEM MODIFICATIONS 31
6.3.4 OTHER ISSUES 32
6.4 MODIFICATIONS INTENDED TO GAIN SITE CLOSEOUT 32
6.4.1 EVALUATION OF THE ADEQUACY OF PAST SOURCE REMOVAL 32
6.4.2 SITE EXIT STRATEGY DEVELOPMENT 32
6.5 SUGGESTED APPROACH TO IMPLEMENTATION OF RECOMMENDATIONS 33
6.6 RECOMMENDATIONS FOR SITE MONITORING AND MONITORING REPORTS 33
6.6.1 MONTHLY REPORTS 33
6.6.2 MONITORING REPORTS 34
7.0 SUMMARY 35
IV
-------
LIST OF TABLES
TABLE 1. MONITORING ZONES IN SITE WESTBAY WELLS 8
TABLE 2. SITE CLEANUP AND TREATMENT PLANT DISCHARGE CRITERIA 9
TABLE 3. SITE AND PLANT DISCHARGE CRITERIA 11
TABLE 4. AVERAGE FLOW FROM EXTRACTION WELLS, SPRING 2003 13
TABLE 5. MASS REMOVAL BY RECOVERY WELL 15
TABLE 6. SELECTED DISSOLVED OXYGEN MEASUREMENTS 2003 16
TABLE 7. COST SUMMARY TABLE FOR INDIVIDUAL RECOMMENDATIONS 36
-------
1.0 INTRODUCTION
1.1 PURPOSE
In the OSWER Directive No. 9200.0-33, Transmittal of Final FYOO - FY01 Superfund Reforms Strategy,
dated July 7, 2000, the Office of Solid Waste and Emergency Response outlined a commitment to optimize
Fund-lead pump and treat systems. To fulfill this commitment, the U.S. Environmental Protection Agency
(EPA) Office of Superfund Remediation and Technology Innovation (OSRTI) and EPA Region 2 identified
a need for an evaluation of the Higgins Farm Superfund Site, New Jersey.
As a matter of background, in fiscal year (FY) 2001, a nationwide effort identified all Fund-lead pump-and-
treat systems in the EPA Regions, collected and reported baseline cost and performance data, and evaluated
a total of 20 systems. The site evaluations were conducted by EPA contractors, GeoTrans, Inc. and the
United States Army Corps of Engineers (USAGE), using a process called a Remediation System Evaluation
(RSE), which was developed by USAGE and is documented on the following website:
http://www.environmental.usace.armv.mil/librarv/guide/rsechk/rsechk.html
In FY 2002 and 2003, additional RSEs were commissioned to address sites either recommended by a
Region or selected by OSRTI.
A RSE involves a team of expert hydrogeologists and engineers, independent of the site team, conducting a
third-party evaluation of site operations. It is a broad evaluation that considers the goals of the remedy, site
conceptual model, above-ground and subsurface performance, and site exit strategy. The evaluation
includes reviewing site documents, visiting the site for up to 1.5 days, and compiling a report that includes
recommendations to improve the system. Recommendations with cost and cost savings estimates are
provided in the following four categories:
• improvements in remedy effectiveness
• reductions in operation and maintenance costs
• technical improvements
• gaining site closeout
The recommendations are intended to help the site team identify opportunities for improvements. In many
cases, further analysis of a recommendation, beyond that provided in this report, is required prior to
implementation of the recommendation.
OSRTI and EPA Region 2 cooperatively selected the Higgins Farm Superfund Site. This site has high
operation costs relative to the cost of an RSE and a long projected operating life. The RSE process
described above was applied to the Higgins Farm Site. The EPA Region 2 RPM, Ms. Pamela J. Baxter,
arranged for the site visit and facilitated the delivery of relevant documents to the RSE team. A site visit
was made on 21 -22 August 2003. This report provides a brief background on the site and current
operations, a summary of the observations made during the 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:
Dave Becker, Geologist, USAGE Hazardous, Toxic, and Radioactive Waste Center of Expertise
(HTRW CX)
-------
Lindsey Lien, Environmental Engineer, USAGE HTRW CX
In addition, Matt Charsky, from HQ USEPA, participated in the first day of the RSE.
1.3 DOCUMENTS REVIEWED
The following documents and data were reviewed as part of the RSE:
Record of Decision Higgins Farm - September 30, 1992
Remedial Investigation, CH2M Hill - June, 1992
Design Basis Report, CH2M Hill - December, 1993
Summary of Field Activities, CH2M Hill - November 1996
Operation and Maintenance Manual, Sevenson Environmental Services - June 3, 2002
Final Plans, CH2M Hill - March, 1995
Operation and Maintenance Reports Sevenson Environmental Services - December 2002, January
2003, February 2003, March 2003, and May, 2003
Quarterly monitoring well sampling results, Sevenson Environmental Services - June 2001, March
2002, May 2002, September 2002, and March 2003
Extraction well sampling results March 2002, June 2002, September 2002, and March 2003
Quarterly Groundwater Monitoring Reports, Sevenson Environmental Services - November 2002,
April 2003.
1.4 PERSONS CONTACTED
Pamela J. Baxter, Remedial Project Manager, EPA Region 2
Neal Ravensbergen, US Army Corps of Engineers, New York District
Jim Russell, Sevenson Environmental Services, Lead Operator
Barney Fite, Sevenson Environmental Services, Operator
Dawn Cermak, Sevenson Environmental Services, Project Manager
Jane TenEyck, New Jersey Department of Environmental Protection, Case Manager
1.5 SITE LOCATION, HISTORY, AND CHARACTERISTICS
1.5.1 SITE LOCATION
The Higgins Farm Superfund site is located in a rural portion of Franklin Township, Somerset County,
New Jersey, approximately 4 miles northeast of Princeton, New Jersey. The site occupies approximately
75 acres southeast of New Jersey State Highway 518 and is largely used as pasture. The site is located at
approximately 40° 24.3' N latitude and 74° 36.2'W longitude.
1.5.2 SITE HISTORY
The following site history was largely derived from the information contained in the 1992 Record of
Decision. Though used as a farm and pasture land, the Higgins Farm Site was at times also used for the
disposal of hazardous wastes, including hazardous substances. In December 1985, the Franklin Township
Health Department reported to the New Jersey Department of Environmental Protection(NJDEP) that
elevated levels of chlorobenzene existed in a potable well located at Route 518, Franklin Township,
Somerset County, New Jersey. NJDEP investigated and discovered the presence of a drum burial dump at
the site approximately forty yards from the contaminated well. Another disposal area, informally named
the NJDEP fenced area, was discovered in the south-central portion of the site.
In April, 1986, the site owner, Mr. Higgins, employed O.K. Materials, to remove drums from a portion of
the property, informally called the excavation pit area. Approximately fifty containers, including drums,
were excavated. During excavation activities, some containers were punctured and their contents spilled
-------
onto the ground as the drums were excavated. Liquids were pumped from the excavation pit to a holding
tank and visibly contaminated soils were placed in roll-off containers.
On April 26, 1986, NJDEP sampled ten residential wells in the vicinity of the site and discovered that three
wells were contaminated with volatile organic compounds (VOCs). Nine of the ten residential wells were
resampled by NJDEP in August 1986. The analysis confirmed the presence of volatile organic
contamination in the ground water. On May 8, 1986, NJDEP personnel inspected Higgins Farm and
collected soil samples from the site, including the excavation pit area. Analysis of these samples indicated
the presence of volatile organic compounds, pesticides, metals, dioxins and furans in the soils at the site.
In March 1987, NJDEP formally requested that EPA assume the lead role in mitigating the Higgins Farm
Site. On April 8, 1987, EPA initiated activities to stabilize the site and to control the release of hazardous
substances into the environment. The following actions were undertaken: the construction of a barn to
house contaminated material, including over-packed drums and roll-off containers; the excavation pit was
drained, lined and backfilled; the pumped liquids were treated and stored in holding tanks; and the
excavation pit area was fenced to prevent access by unauthorized persons. EPA conducted additional
excavation of 94 drums in 1992 in the southwestern portion of the site.
In March 1987, EPA responded to the presence of contamination in drinking water wells neighboring the
site by providing bottled water to potentially impacted area residents. At that time, EPA explained that it
would provide bottled water as an interim solution until an alternate water supply could be arranged by
NJDEP. Thereafter, NJDEP determined that the most appropriate method to supply potable water was to
install individual carbon units at the potentially impacted homes. NJDEP installed the carbon filter units
during the spring/summer of 1989, at which time bottled water delivery was discontinued.
The site was proposed for inclusion on the National Priorities List (NPL) in June 1988. In March 1989, the
site was formally placed on the NPL, thus making it eligible for federal funds to investigate the extent of
contamination and to clean up the site. In June 1990, EPA released a Focused Feasibility Study (FFS)
report and EPA's Proposed Plan for the construction of a water line extension to provide the potentially
affected residents with an alternate water supply. On September 24, 1990, EPA issued a first Record of
Decision (ROD) that selected an interim remedy to connect the potentially affected residents to an existing
water supply. The water supply was later installed for residents along State Highway 518.
Remedial investigation (RI) field work was conducted by EPA's contractors in two phases: from March
1990 through January 1992, and from February 1992 through March 1992. Ground-water, surface and
subsurface soil, surface-water, sediments and suspected source area (through test pit excavation) samples
were collected and analyzed during Phase I of the RI. Phase II of the RI, which included the excavation of
additional test pits and sampling, was conducted to investigate other potential sources of contamination.
The RI effort also included a characterization of the site hydrogeology including water level measurements
and slug, packer, and pump testing to determine hydraulic parameters. A risk assessment conducted in
conjunction with the RI report indicated the site posed unacceptable risk to human health. A feasibility
study of remedial alternatives was also prepared in 1992. Based on this RI and FS, a second ROD to
address ground water contamination at the site was prepared in September 1992. The selected alternative
included the installation of extraction wells and construction of a treatment plant that would include metals
treatment by precipitation and ion exchange and removal of volatile organic compounds by air stripping. A
remedial design was prepared in 1994 and construction activities began in 1995. A third removal action
occurred in 1996 which included removal of drums, containers, and contaminated soil encountered during
on-site road construction to the treatment plant. The plant began operating in May 1998.
1.5.3 SITE CHARACTERISTICS
1.5.3.1 HYDROGEOLOGY
The Higgins Farm Superfund Site is located in the Piedmont physiographic province and overlies rocks of
the Newark Basin. The local bedrock is the Rocky Hill Diabase, a medium to dark gray rock composed of
feldspar and pyroxenes, with lesser amounts of olivine, biotite, garnet, quartz, and hornblende and traces of
-------
other minerals. The Rocky Hill Diabase is part of a sill (sub-horizontal tabular intrusion) that underlies
much of this area of New Jersey and is exposed along the Hudson River as the Palisades. The diabase is
estimated to be over 1000 feet thick in southern Franklin Township. The upper portion of the diabase is
weathered and fractured and is overlain by residual soils (primarily silty), fill, and traces of glacial deposits.
Alluvial deposits are found near nearby streams. Overburden is generally less than 15 feet thick and in
some places on site less than five feet thick.
The degree of fracturing in the diabase diminishes with depth. The unweathered rock is quite competent
and has infrequent fracturing. Rock quality designation (RQD) values, which describe the degree of
fracturing (higher values indicate larger fracture spacing), are very high. Continuous cores were obtained
during drilling of the Westbay system monitoring wells. Very few sub-horizontal fractures were
encountered, though some appeared to be open and often iron or manganese oxide coated. Others were
filled with chlorite, pyrolusite, epidote, and less common minerals. Steeply dipping fractures that were
encountered were largely completely healed by calcite, chlorite or other minerals. During coring, water
loss was often negligible, though some zones were associated with significant water loss, indicating
significant fracture permeability.
Though of high quality, the diabase is, however, offset by a significant fault zone in the large trap rock
quarry just southwest of the site. The fault zone strikes N 15° E and dips approximately 50 degrees to the
southeast. Offset may be as large as 1700 feet (CH2M-Hill RI, 1992). Rock near this fault zone is more
fractured and weathered. At least one possible fault zone was encountered while drilling at the site (at
approximately 145-foot depth in MW-11) and slickensides were observed on fractures.
Ground water movement is believed to be restricted to the overburden and fractures within unweathered
bedrock. Most flow is probably concentrated in a limited number of connecting open fractures. Not
surprisingly, well yields at the site are low, and seepage into the nearby quarry was relatively small but
noticeable during a site visit on August 16, 2003 by the EPA RPM and James May of the USAGE Engineer
Research and Development Laboratory located in Vicksburg, MS. Movement is controlled by the preferred
orientation of the fractures.
1.5.3.2 SITE GROUND WATER CONDITIONS
Ground water conditions have been characterized at two general depths - shallow (less than 20 feet depth)
and deep (20- 200 feet depth). Shallow monitoring wells are constructed with casing and five to ten feet of
well screen. Deep monitoring wells installed during the RI have a surface casing into competent bedrock
and are completed as open holes to total depth. As such, the deep monitoring wells represent conditions
over very long vertical intervals. In 2000, additional monitoring points were installed around the perimeter
of the site. These wells were completed as multi-port Westbay wells to depths of approximately 200 feet.
Water levels are typically not taken from these ports, though they are sampled for chemical analyses.
Static water levels in monitoring points are typically less than 2 feet to over 25 feet below grade at the site.
Piezometric levels are typically 275-235 feet above sea level. Ground water flow directions based on
piezometric levels measured in the RI monitoring wells are comparable for the two depths. Ground water
flow is primarily from the north-central part of the site to the south and southeast. Hydraulic gradients are
typically around 0.015 ft/ft with steeper gradients near the extraction system and on the western part of the
site. At times, the gradients in the central part of the site drop to as low as 0.005. Hydraulic conductivities
are highly variable, particularly in the fractured rock. Slug testing on shallow wells conducted in the RI
indicated hydraulic conductivities of 2E-3 to 3E-5 cm/sec. Packer testing on intervals in the deep open
monitoring wells indicated hydraulic conductivities as high as 4E-3 cm/sec to many intervals (presumably
essentially unfractured) displaying conductivities of less than 1E-7 cm/sec. Higher conductivities were
typically identified in the upper 70 feet of the deep wells. Pump testing conducted at the site yielded an
estimated transmissivity value of 112 ft2/day. This value was used in the original basis for the extraction
well layout. The Design Basis report recognized that additional wells may be needed to account for the
uncertainty in the estimated transmissivity.
-------
1.5.3.3 SITE CONTAMINATION
Contaminants at the site include, as would be expected given the nature of the release, a variety of organic
compounds and some metals. The primary organic compounds found in ground water are tetrachloroethene
(PCE), trichloroethene (TCE), with lesser amounts of 1,2-dichloroethane (1,2-DCA), cis-1,2-
dichloroethene (1,2-DCE), chlorobenzene (CB), benzene, 1,1,2-trichloroethane (1,1,2-TCA), 1,1,2,2-
tetrachloroethane (1,1,2,2-PCA), and trace amounts of vinyl chloride (VC). Other contaminants had been
identified in soils at the site, including pesticides. Past removal actions by EPA and others have addressed
the major source areas through the removal of drums and significant soil contamination.
Although characterization done during the RI identified widespread contamination at the site, ground water
extraction has reduced the on-site extent of contamination. Whereas, the on-site plume was not initially
well defined, now several individual plumes appear to emanate from specific past disposal sites. The
largest mixed organic compound plume extends southeastward from the Excavation Area apparently
beyond the Higgins property boundary and has a width of 750-800 feet. Another mixed organic compound
plume extends southward from the NJDEP fenced area. This plume widens southward to over 1000 feet
due to the source location on the slight ground water divide extending through the site. Finally, another
plume, consisting primarily of 1,2-DCA with some 1,1,2-TCA extends westward from an indistinct source,
perhaps near the Higgins residence. A geophysics survey in 1996 did not reveal drums in the area.
Given the presence of cis 1,2-DCE and trace amounts of VC, there is possibly some reductive
dechlorination occurring at the site.
Ground water at the site is largely aerobic and dissolved oxygen levels are typically above 2 ppm, though
there are some wells with DO concentrations as low as 1-2 ppm. Iron levels are variable, but elevated in
some samples. Eh values are almost all positive relative to the reference electrode. Elevated turbidity is
noted from some wells, including some evidence for suspended solids from certain extraction wells.
Monitoring wells with elevated turbidity over 20 NTU (as of March 2003) include MW-2S, -2D, -3D, -6S,
-6D, -8S, and -10S.
1.5.3.4 SITE AND NEARBY LAND USE
The region around the Higgins Farm Site is used for agricultural, rural residential, mining, and recreational
purposes. There has been an increase in residential development in the area. The Higgins site is currently
used as pasture for livestock. Portions of the Higgins property have been identified as wetlands. Wooded,
undeveloped land lies to the south and southeast of the site almost to State Highway 27, though there is a
law enforcement pistol range in the area. The Trap Rock Industries quarry uses land to the southwest.
Areas to the north and east are used for a mix of agricultural and residential purposes. Most homes in the
immediate vicinity of the site, with the exception of residences along State Highway 518, have domestic
water wells. EPA provided a number of residences along highway 518 with municipal water in a 1990
interim remedial action.
-------
2.0 SYSTEM DESCRIPTION
2.1 SYSTEM OVERVIEW
The groundwater remedy for the Higgins Farm Superfund Site includes the following items:
• groundwater extraction from 20 wells located on site;
• above ground treatment (equalization, coagulation and precipitation, filtration, ion exchange, air
stripping, pH and alkalinity adjustment, solids handling, and discharge to a tributary to Carters Brook).
The groundwater treatment system operations began in May 1998. During the first year system
commissioning phase, the system experienced hydraulic and process related problems. The secondary ion
exchange system was not operating properly and taken off line due to short runs and inefficient operation.
Metals of interest (iron, manganese, barium, zinc, barium, nickel, magnesium and copper) are generally
below discharge levels following the weak acid ion exchange resin contactors. The plant was designed to
have a maximum capacity of 100 gpm, although the surface water discharge standards in the permit
equivalent are based on an average effluent flow rate of 75 gpm. Start up and shake down testing identified
a maximum capacity of approximately 60 gpm. Additional problems associated with scaling down stream
of the reaction tank and clarifier further exacerbated the problem. The capacity of the well field was
slightly below the 60 gpm limitation of the plant, and the plant has been able to treat all the water extracted.
Plant operations transitioned to the current contractor Sevenson Environmental Services (SES) in 1999.
The new operations contractor proceeded to diagnose hydraulic and operational system issues and
reinforced the previous contractors finding that the chelating ion exchange system was not needed to meet
metals discharge standards. SES revised the operation of the plant to include more chemical addition to the
reaction tank to ensure the softening step which was used as a pretreatment step for the ion exchange units,
maximize iron and calcium removal upstream of the ion exchange beds, resulting in longer filter runs and
media life. Pump control wiring and iron fouling in the extraction pumps continue to consume a good deal
of the operator's time each day.
2.2 EXTRACTION AND INJECTION SYSTEM
The ground water extraction system includes 20 extraction wells, seventeen of which have been installed in
a "U"-shaped configuration around the western, southern, and southeastern edge of the site. These wells
are typically spaced approximately 150 to 400 feet apart. Three extraction wells have been installed just
south and west of the excavation pit area. The extraction wells are all six inches in diameter and
approximately 150 feet deep with a screen and filter pack extending from essentially the bottom of the
borehole to within 5 feet of the surface. The 20 feet of screen in the interval of unconsolidated materials
and weathered bedrock has a slot-size of 0.010-inch and is surrounded by filter sand (Morie OON sand).
The remainder of the screen set into competent rock had a slot size of 0.100-inch and was surrounded by
pea gravel. Following drilling and prior to well construction, 16 of the 20 wells were hydraulically
fractured to increase flow rates from the wells. Based on the report prepared by CH2M-Hill on the
installation of the wells (CH2M-Hill, 1996), the hydraulic fracturing was only partially successful at
increasing production from the extraction wells and resulted in an average increase of less than 1 gallon per
minute.
The wells are equipped with Grundfos submersible pumps, with % horsepower motors, that are nominally
rated for flow at 5 gpm. The well flows from the western and southern portions of the site are conducted to
-------
the treatment plant through a combined 3-inch-diameter fiberglass-reinforced plastic (FRP) header pipe.
Wells along the southeast edge of the site and the three wells near the excavation pit are connected to the
treatment plant by separate 2-inch-diameter FRP pipe. Each well has a flow meter, sample port, solenoid-
controlled valve, and pressure gauge. The wells are completed with a concrete vault with locking metal
doors. These vaults do not have a drain or access ladder. A few vaults have sump pumps. Piping within
vaults are heat-traced and insulated. A separate vault contains the electrical panel for the well or wells
nearby. Electrical supply and control wiring are buried. In 2003, control wiring was replaced at several
wells with radio communication equipment. Wiring for the remaining wells was completed in January
2004. The antennas need to be protected from damage caused by cattle present at the site.
2.3 TREATMENT SYSTEMS
The Higgins Farm groundwater treatment system was originally designed to operate at 100 gallons per
minute (gpm). The actual flow rate varies between 35 and 60 gpm depending on water levels, number of
pumps operating as well as other factors. The maximum design influent and discharge concentrations at
the plant are shown in Table 3.
The treatment system consists of the following elements:
• duplex equalization tanks
• aerated chemical reaction tank/flocculation tank/clarifier
• duplex sand media pressure filters
• weak acid ion exchange system
• chelating ion exchange system
• duplex packed tower air strippers
• VOC off gas venting system to the atmosphere
• backwash holding tank
• chemical storage and feed
• effluent holding tank
• discharge to Carters Brook
• filter press system for solids handling
All treatment components with the exception of the bulk chemical (caustic, sulfuric acid) storage tanks are
housed in the 100' wide x 120' long pre-engineered metal building.
2.4 MONITORING SYSTEM
The monitoring network for the extraction system consists of 21 wells; 7 shallow monitoring wells, 8 deep
monitoring wells, and six Westbay multiport wells. There are also a number of piezometers installed near
the wetland areas at the site to determine impact of the ground water extraction on the wetlands. Standard
monitoring wells were installed during the RI and are confined to the Higgins property. The Westbay wells
were installed in 2000 and all are downgradient of the extraction system with exceptions of WB-11. All
Westbay wells are located off the Higgins property with the exception of WB-11 and WB-14.
The shallow monitoring wells are typically screened at depths less than 20 feet. Deep monitoring wells are
completed as open holes to depths of approximately 200 feet in competent bedrock. These deep wells
isolate the unconsolidated materials and weathered bedrock with surface casing. The Westbay wells each
have multiple sampling ports at various depths, as described in the Table 1.
-------
Table 1: Monitoring Zones in Site Westbay Wells
Zone
4
3
2
1
Westbay Well Monitoring Zone Depths (total well depth in parentheses, units are in feet)
WB-11
(194')
N/A
47'
94'
179'
WB-12
(160')
N/A
N/A
45'
145'
WB-13
(200')
N/A
56'
145'
190'
WB-14
(200')
47.5'
112.5'
154.5'
189.5'
WB-15
(170')
N/A
30'
115'
155'
WB-16
(172')
N/A
33'
115'
157'
The conventional monitoring wells are sampled semi-annually using low-flow techniques and non-
dedicated pumps. The Westbay wells are also sampled semi-annually, but samples are obtained using a
MOSDAX sampler that is lowered inside the Westbay casing and attaches to a sampling port. The port is
opened and a stainless steel sample chamber is filled. No purging is done. Samples are also collected
semi-annually from the sample taps on the individual extraction wells. Samples are submitted for fixed lab
analyses for volatile organic compounds by EPA method 8260B and metals by EPA methods 6010, 7041,
7421, and 7470. Analyses have also been conducted for dissolved gases methane, ethene, and ethane as
well as chloride, nitrite, nitrate, and sulfate. Field analyses are conducted for dissolved sulfide, total iron,
dissolved (ferrous) iron, pH, dissolved oxygen, turbidity, and alkalinity.
-------
Table 2
Site Cleanup and Treatment Plant Discharge Criteria1
Parameter *
Benzene
Carbon Tetrachloride
Chlorobenzene
Chloroform
1, 2 DCA
1, 1 DCE
cis-1, 2 DCE
1, 1, 2, 2 PCA
PCE
1, 1, 2 TCA
TCE
VC
bis(2-chloroethyl) ether
bis(2-ethylhexyl) phthalate
Hexachlorobutadiene
Dieldrin
Aluminum
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Vanadium
Zinc
Cyanide
*BOD5
*Dissolved Oxygen
pH (units)
*PetroHydrocarbons
*TOC
*TDS
*TSS
Cleanup
Levels
OU-2 ROD
(Hg/L) 2
1
2
4
100
2
2
10
1
1
3
1
2
-
6
-
-
_
6
-
2000
4
5
100
1300
-
-
15
.
-
100
-
-
-
.
-
-
-
-
-
-
New Jersey
Surface Water
Quality Criteria
(Hg/L) '
1
I6
1
1
I6
I6
1
I6
I6
I6
1
1
5
2.5 [625] 6
1
0.002 [608] 6
100 6
5
0.5 [200.9] 6
28
I6
I6
5 [200.7] 6
10 6
5 [200.7] 6
300
0.3 6
50
0.2 [245.1] 6
13
14
47
5.2
25
5
6.5-8.5
15
50
500
40
New Jersey
Maximum
Contaminant Limit
(Hg/L) '
1
2
4
.
2
2
10
-
1
-
1
2
-
-
-
-
_
-
-
-
.
-
.
-
-
-
.
-
-
-
2
Federal
Maximum
Contaminant Limit
(Hg/L) '
5
5
100
100
5
7
70
-
5
5
5
2
-
6
-
-
_
6
2000
4
5
100
.
1300
-
15
100
-
-
all concentrations in ug/L except those identified by * are in mg/L
1 NJDEP Permit Equivalent Table Outfall 001, Higgins Farm SF Site, August 4, 1994
2 Table 16, Higgins Farm SF Site OU-2 ROD dated 30 September 1992
6 Concentrations are "Discharge Reporting Levels" (DRL's) which corresponded to the detection limits using the published
analytical method.
-------
3.0 SYSTEM OBJECTIVES, PERFORMANCE AND CLOSURE
CRITERIA
3.1 CURRENT SYSTEM OBJECTIVES AND CLOSURE CRITERIA
The groundwater extraction and treatment systems were installed as prescribed in the ROD for the second
operable unit (OU) for the Higgins Farm Superfund Site. The Regional Administrator for the U.S.
Environmental Protection Agency (EPA) Region II signed the Record of Decision for the Higgins Farm
Superfund Site on September 30, 1992. The first operable unit provided for installation of a water line to
supply area residents with an alternate water supply. The selected remedy for the second OU, as
recommended in the FS, included the installation of ground water extraction wells, installation of an on-site
groundwater treatment facility, and discharge of the water to the on-site pond that overflows into a branch
of Carters Brook. The plant effluent is required to meet the stringent surface water quality criteria
indicated in Table 3.
The following are the remedial action objectives set forth in the OU-2 ROD:
• capture and treat the contaminated ground water in an attempt to restore the aquifer to Federal and
State drinking water standards
• control or limit the future off-site migration of the contaminated ground water
• minimize the potential for direct exposure of the populace to the contaminated ground water.
The duration of the final remedy as identified in the OU-2 ROD was estimated to be between 5 and 30
years based on design investigations.
3.2 TREATMENT PLANT OPERATION GOALS
The treatment plant goals are to maximize contaminant removal, minimize plant down time, and be
consistent with the final discharge criteria specified in the ROD as identified in Table 2 and Table 3. The
plant discharge is routed through a small wetlands area and discharged to a small pond that overflows into a
branch of Carters Brook. The discharge criteria are extremely conservative, and in many cases exceed the
aquifer clean up levels specified in the ROD. Table 3 identifies the design/actual/proposed influent
concentrations, and Table 2 the effluent discharge criteria.
3.3 ACTION LEVELS
The action levels for the system are identified in Table 3 as specified in the ROD. Referring to Table 3 it is
interesting to note these criteria are actually higher than the discharge standards identified for the above
ground treatment facility discharge to Carters Brook.
10
-------
Table 3: Site and Plant Discharge Criteria
Parameter
Benzene
Chlorobenzene
1, 2 DCA
1, 2 DCE
1, 1,2,2 PCA
PCE
1, 1, 2 TCA
TCE
Vinyl Chloride
Aluminum
Arsenic
Barium
Beryllium
Cadmium
Chromium
Copper
Iron
Lead
Manganese
Mercury
Nickel
Zinc
Discharge
Criteria
Hg/L8
1
1
1
1
1
1
1
1
1
100
8
28
1
1
5
10
300
0.3
50
0.2
13
47
Influent Concentration
GWTP
/g/L
Design Actual
1200 <1
1100 1.1(1)
320 <1
13 3.6(1)
7.5 2.3(1)
740 40(1)
1100 2(1)
220 15(1)
86 <1
304,000 <100
<5 <8
1890 30(21)
25.7 <1
4.1 <1
403 <10
8750 <10
433,000 860(300)
81.4 <3
24,800 210(80)
<0.2
224 <20
811 <20
Revised
Concentration
To GWTP
Proposed4 ug/L
<1
2
2
4
6
67
5
34
0.2
<100
<5
<21
<1
<1
<1
<1
1500
<3
270
<0.2
<20
<20
Ecotoxiciry Chronic
Screening Levels 5
EPA Reg 4 (ug/L)
53
195
2000
1350
240
84
940
NA
NA
87
190
NA
0.53
0.66
117
6.54
NA
1.32
NA
0.012
87.71
58.91
RBC's/PRG Levels6
(ug/L)
EPA Reg 3 EPA Reg 9
(tap water) (tap water)
0.34c 0.34c
HOnc HOnc
0.12c 0.12c
61nc cis 61nc
120nc trans 120nc
0.053c 0.055c
O.lOc 0.66c
0.19c 0.20c
0.026c 0.028c
O.OISc 0.020c
37,000nc 36,000nc
0.045c 0.045c
2,600nc 2,600nc
73nc 73nc
18nc 18nc
55,000nc 55,000nc
ISOOnc ISOOnc
HOOOnc HOOOnc
NA nc 0.0036nc
730 nc 880nc
NA nc line
73 One 73 One
HOOOnc HOOOnc
MCL7
New Jersey
(ug/L)
1
50
2
70 cis
100 trans
1
1
3
1
2
200 (s)
10
2000
4
5
100
1300(AL)
300(s)
15 (AL)
50(s)
2
NA
5000(s)
11
-------
Notes: All values are micrograms per liter (ug/L).
1 The Risk-Based cleanup concentrations are identified in Table 2
2 Table 5-1 Section 5, Design Basis Report, CH2MHill, December, 1993.
3 Calculated based on the average concentration of the last 4 sampling results for all wells
4 Calculated based on the average concentration of the last 4 sampling results for wells to remain in service,
1, 2, 3A, 4, 5, 6, 8A, 9A, 13 and new well 11 A.
5
Region 4 Screening Values for select metals based on Hardness = 50 mg/L.
6 Region 9 PRG's dated October 2002; Region 3 RBC's dated April 2003 both reflect HI = 0.1 or 10"6 Increased Cancer Risk
7 Information obtained from NJDEP web site http://www.state.ni.us/dep/watersupply/standard.htm
8 Compounds/analytes listed are those detected, a complete list is included in Table 2. See note 6, Table 2 concerning DRLs
MCL = Maximum Contaminant Level NA = Not applicable NC = Not calculated AL = Action Level
nc = non-carcinogenic risk c = carcinogenic risk
12
-------
4.0 FINDINGS AND OBSERVATIONS FROM THE RSE SITE
VISIT
4.1
4.1.1
Ground Water Extraction and Injection System
WELL CONDITION
Extraction well yields are highly variable, from a consistent 5 gpm to less than one gpm. Only extraction
well RW-13 has an average capacity over 5 gpm and runs nearly continuously. The table below describes
the recent average flow rates for the extraction wells. Most well pumps cycle on and off and are not
throttled, with one well cycling 15-30 times per day. Wells require periodic rehabilitation due to fouling by
iron bacteria. The operators perform their own rehabilitation by recirculating a mixture of 50 Ibs of
sulfamic acid and 55 gallons of hot water through the screened interval using the extraction pumps.
Rehabilitation of each well is typically performed twice a year.
Table 4: Average Flow from Extraction Wells, Spring 2003
Extraction
Well
RW-1
RW-2
RW-3
RW-3A
RW-4
RW-5
RW-6
RW-7
RW-8
RW-8A
Average Flow
Rate (gpm)
1.7
0.29
0.83
0.76
0.89
1.2
0.37
1.2
2.7
2
Extraction
Well
RW-9
RW-9A
RW-10
RW-10A
RW-11
RW-12
RW-13
RW-14
RW-1 5
RW-16
Average Flow
Rate (gpm)
1.1
0.71
0.5
3.5
3.4
4
6.5
2.1
1.4
1.8
The extraction pumps require periodic maintenance due to fouling and wear. The operators perform much
of the pump maintenance and periodically scavenge parts from other unusable pumps. Fouled pump bodies
and impellers are cleaned using a sulfamic acid bath. Motors have a limited (roughly 2 year) life due to
electrical failure, corrosion, or wear on the drive shaft. The operators have not noted corrosion of the drop
pipe. A check valve in RW-10 failed at one point, allowing the reinjection of untreated water from other
wells.
The heat tracing and insulation for piping in the well vaults have been successful at preventing freezing,
though the heat tracing failed during the winter of 2002-2003 at wells RW-10 and RW-9A and piping there
froze. Given the low flow rates, velocities in the piping from the wells to the plant are not high and there is
a potential for build up of solids in the piping. The piping has not been pressure tested since it was
installed. Even when a contractor punctured the piping while installing a fence, the piping was repaired but
not pressure tested.
13
-------
4.1.2 PLUME CAPTURE AND REMEDIATION
The objectives of the ground water extraction system at the Higgins Farm Site are to contain the plumes on-
site and reduce on-site concentrations. To evaluate the system performance in this regard, the plumes must
be adequately defined, the hydrogeology must be adequately characterized to assess probable flow paths
and assess variability in ground water flow directions, water levels near the extraction system must be
measured, and the flow rates for individual wells must be known. Flow rates are available for the
individual extraction wells. Additional data regarding the other parameters would be quite useful.
The plumes are imprecisely defined on-site, though generally adequately defined relative to the placement
of the extraction wells to assess probability of capture. Plume extent is defined largely with the extraction
well concentrations near the site boundary. Additional monitoring wells are required to define the plumes
on site. For example, the extent of elevated 1,2-DCA near extraction well RW-3 is defined only by
extraction wells, the source of the plume is not certain. The northeastern edge of the main plume along the
northern boundary of the site is only defined by recent low concentrations in RW-13 and the southwestern
edge of that plume is not defined by monitoring wells within 500 feet. The very long open intervals in deep
monitoring wells and extraction wells have made delineation of the vertical extent of the plumes
problematic.
The fractured rock aquifer represents a very complex flow system. Given the highly discrete flow paths in
fractured rock and limited number of fractures at depth at the site, there is a potential, though perhaps
small, that some contaminant flow paths from the site are not intercepted by pumping of the extraction
wells. In the area near RW-11 and RW-12, the wells are approximately 250 feet apart. These wells are in
the core of the main plume and this spacing may not be adequate to capture all flow with a high degree of
certainty. The orientation of the main plume in this area is similar to the strike of fractures observed in the
nearby quarry and there may be a significant horizontal anisotropy to the flow paths. There may be
significant permeable pathways perpendicular to the line of extraction wells near RW-11 and RW-12
inhibiting complete capture. Ground water flow directions are generally stable and predictable. The
existing wells are generally located appropriately relative to the plumes' plan-view orientation. The
extraction wells are quite deep with a long screened interval. Given the uncertainty in the vertical extent of
the plumes, the use of such long interval extraction wells is of unknown benefit.
There are few monitoring points near the extraction wells to verify the piezometric surface and capture
zone. It is not possible to draw flow lines based on the piezometric contours near the extraction well lines
(and the fractured nature of the aquifer would make that an uncertain exercise, regardless of the available
data).
Overall, a water budget approach, as documented in SAIC (2002), suggests the extraction rates are
adequate to capture site ground water. Further analysis of the extraction conditions indicates that a
significant number of extraction wells recover little contaminant mass. Six of the 20 extraction wells, RW-
7, -8, -10A, -11, -14, and -16, account for approximately 95% of the extracted contaminant mass, based on
the March 2003 sampling and flow rates. Another six wells, RW-1, -2, -3 A, -5, -6, and -13, each extract
less than 0.1% of the site mass and had essentially non-detectible levels of contaminants or levels below the
ground water quality standards in March 2003.
The presence of contaminants in the Westbay wells downgradient of the extraction wells does raise the
concern of additional contaminant extent outside the capture zone of the system. This issue is discussed
further in section 5.1.
14
-------
Table 5: Mass Removal by Extraction Well
Extr Well Flow (gpm)
RW-1
RW-2
RW-3
RW-3A
RW-4
RW-5
RW-6
RW-7
RW-8
RW-8A
RW-9
RW-9A
RW-10
RW-10A
RW-11
RW-12
RW-13
RW-14
RW-15
RW-16
Total
1.7
0.29
0.83
0.76
0.89
1.2
0.37
1.2
2.7
2
1.1
0.71
0.5
3.5
3.4
4
6.5
2.1
1.4
1.8
37
Total VOC1
0
1
22
0.4
3
0.6
2
185
42
4
33
14
17
54
270
5
0
33
10
340
Total
Mass Surrogate2
0
0.29
18.26
0.304
2.67
0.72
0.74
222
113.4
8
36.3
9.94
8.5
189
918
20
0
69.3
14
612
2243.424
0.0%
0.0%
0.8%
0.0%
0.1%
0.0%
0.0%
9.9%
5.1%
0.4%
1.6%
0.4%
0.4%
8.4%
40.9%
0.9%
0.0%
3.1%
0.6%
27.3%
1Computed by summing VOCs above NJDWS
2Computed by multiplying gpm by total VOCs
4.1.3
NATURAL ATTENUATION CONDITIONS
The degradation of many chlorinated organic compounds occurs under anaerobic reducing conditions. Site
ground water is largely aerobic and unfavorable for biodegradation, though reductive dechlorination
products of trichloroethylene and tetrachloroethylene such as 1,2-DCE and VC are found at low
concentrations. Petroleum hydrocarbons, including benzene, and other aerobically degradable organics
such as acetone have been detected in some areas, and may represent localized carbon sources (in addition
to other natural carbon in shallow soils) adequate to locally deplete electron acceptors such as oxygen,
nitrate, and sulfate and allow reductive dechlorination. A few monitoring well locations have been shown
to be periodically depleted in dissolved oxygen (see Table 6, below). In addition to degradation, dispersion
and sorption are occurring at the site as these processes are common to all sites, though to different degrees.
Flow in fractured rock would not likely be subject to significant sorption as there is little if any organic
carbon in such settings. Sorption to weathering products, such as clays lining the fractures, may be
significant. The aerobic degradation of the petroleum hydrocarbons and acetone is likely at this site and
may be the cause of large declines in the concentrations of benzene and acetone in ground water. For
example, according to the RI, benzene has been present at concentrations of over 1 mg/L in monitoring
well MW-7D, but most recently has been measured at less than 1 ug/L, a decline of three orders of
magnitude. In summary, the biodegradation of the predominant chlorinated organics (TCE, PCE) is not
likely to be contributing significantly to declines in concentrations now or in the future, but is probably
greatly remediating some of the degradable organics such as benzene. TCE and PCE are likely to migrate
off-site if not captured.
15
-------
Table 6: Selected Dissolved Oxygen (DO) Measurements, March 2003
Monitoring Well
MW-1S
MW-1D
MW-2S
MW-2D
MW-3D
MW-4D
MW-5S
MW-5D
DO Measurement (ppm)
2.6
5.3
2.1
0.48*
3.2
5.5
8.9
5.9
Monitoring Well
MW-6S
MW-6D
MW-7D
MW-8S
MW-8D
MW-9S
MW-10S
DO Measurement (ppm)*
7.8
0.6*
7.8
5.4
2.1
4.4
6.0
* Values indicative of dissolved oxygen depletion.
4.2
TREATMENT EQUIPMENT
Groundwater is extracted from 20 wells through a system of 3 headers, which discharge to the groundwater
treatment plant. Once the flow enters the plant it is metered and directed to one of the two 16,000 gallon
equalization tanks. Water is pumped from the equalization tank(s) to a small aerated reaction tank where
chemicals are added to initiate the chemical reactions necessary to remove excess hardness that would
reduce the effectiveness of the ion exchange units. Following the reaction tank water flows through a
flocculation tank, and finally a clarifier where the resultant sludge settles out and is eventually dewatered
by a plate and frame filter press system. Clarified effluent is pumped to a duplex pressure sand filter
designed to remove any solids that might be carried over the clarifier weirs before undergoing treatment in
the ion exchange units that follow the filters. The ion exchange system consists of 2 separate systems, a
weak acid exchange system followed by a chelating ion exchange resin system. Following treatment in the
ion exchange systems the water is passed through 2 packed tower air strippers operating in series. After
VOC removal the pH and alkalinity is adjusted, and directed to an effluent holding tank and discharged to
the on site pond.
The system currently operates in a batch mode at a flow rate of approximately 55 gpm. The original plant
design was based on a continuous flow rate of 100 gpm. Because of hydraulic considerations and limited
instrumentation, which makes it difficult to balance chemical feed, flows, and other critical parameters, the
decision was made to operate the plant in a batch mode. Influent flow rate to the plant varies from 35 to 60
gpm depending on seasonal aquifer variations. Generally flow rates above 55 gpm are rare and short lived,
but when they occur for extended periods, are managed by restricting flow from the well field.
4.2.1
EQUALIZATION TANKS
Flow equalization is accomplished by passing flow through one of the two 16,000 gallon units. Current
operation is to use one as a standby unit available if the level in the operational unit reaches the
interconnecting pipe between the two tanks. This may occur if the inflow to the facility exceeds 55 gpm, or
if the treatment unit is down for a short period such as 6 - 12 hours. Flow is processed at a flow controlled
target batch flow rate of 55 gallons per minute, which typically exceeds the plant influent flow rate by
approximately 15 gpm. Flow is pumped from the equalization tank until it reaches the low level cut off
elevation that shuts down the entire treatment process and restarts the plant when the equalization tank
reaches an elevation of 13 feet. After activating the low level plant shut down elevation, the well field
continues to operate, and generally takes 4-6 hours to fill the equalization tank to the 13' level, the
elevation which reenergizes the plant following a low level shut off. Pumps P-2-2-1 and P-2-2-2 transfer
flow from the equalization tanks to the softening/metals removal process. Although the 55 gpm flow rate is
below the 100 gpm design capacity of the plant, it is very close to its current hydraulic capacity of 60 gpm.
16
-------
4.2.2 SOFTENING/METALS REMOVAL PACKAGE
The softening/metals removal package consists of three discreet components, the reaction tank, flocculation
tank, and the clarifier/thickener. The 1500 gallon reaction tank is aerated at a rate of 1.5 scfm via a diffuser
to convert any ferrous iron present into insoluble ferric iron. Samples taken September 29, 2003 identified
ferrous iron concentrations were less than 0.1 mg/L. Addition of air and the presence of a slight vacuum
present in all the gravity tanks upstream of the air strippers, also assist in the volatilization of the volatile
solvents present in the tanks. Heavy metals are removed by adjusting the pH with a 50% caustic (sodium
hydroxide) solution. A control loop that senses the pH in the reaction vessel adjusts the amount of caustic
fed to the vessel based on a preset pH value of 10. Soda ash (30% concentration) is added to the reaction
tank to aid in the softening process, in particular precipitation of non-carbonate hardness consisting of
calcium and magnesium sulfates or chlorides. The soda ash feed is flow paced at a rate of 100 mg/L as is 2
ppm of anionic polymer. The reaction tank is equipped with a constant speed rapid mixer and has a
detention time of 27 minutes at a flow rate of 55 gpm which far exceeds the detention time recommended
for flash mix units.
Flow from the reaction tank flows by gravity to the flocculation tank, which was also sized at!500 gallons
with a 27-minute detention time at a flow rate of 55 gpm. Detention times at or above 30 minutes are
generally recommended for flocculation tanks, however given the design of the clarifier/thickener, which is
actually combines the flocculation and clarification steps together, the total flocculation time between the
two units exceed the recommended 30 minutes minimum. The operators pointed out 3-4 feet of sludge
needs to be cleaned out of the flocculation tank every 3 months due to the extensive settling that occurs in
the tank. Polymer may be added in the piping between the reaction tank and the flocculation tank, or
directly into the flocculation tank, and is flow paced. The flocculation tank is equipped with a variable
speed mixer that is intended to promote flocculation and particle growth as the water flows into the
clarifier/thickener unit.
The clarifier/thickener unit consists of an approximately 40,000-gallon clarifier, with a 12-hour detention
time. The clarifier is also provided with a motor driven rake arm assembly, which is supported by the
walkway/truss, located on the top of the clarifier. The clarifier has a conical reaction or flocculation zone,
which is also supported by the clarifier walkway/truss system. The 4" discharge pipe between the two
tanks enters the above the clarifier rake and travels in about 3 feet, then angles upward horizontally for 5
more feet then up vertically for 5 feet allowing floe particles to enter the reaction well near the water
surface and settle in the clarifier. The sludge density and sludge blanket depth are monitored remotely. A
small volume of settled sludge is recycled to the reaction tank to act as seed sludge to enhance particle
growth in the reaction and flocculation tanks. Sludge is withdrawn and dewatered through the plate and
frame filter press, and clarified effluent is directed to the 1000-gallon filter feed tank.
4.2.3 FILTER SYSTEM
The filter system is intended to remove solids carry over from the clarifier/thickener and softening process
prior to the ion exchange beds designed to remove trace metals. The 1000-gallon filter feed tank receives
effluent from the clarifier and then directs the flow to the duplex pressure sand filter units. The filter feed
pumps (P-5-2-1, P-5-2-2) that supply the media filters were designed to produce enough head to force the
flow through the filter system and the weak acid ion exchange beds, chelating resin beds, and finally to the
intermediate pH adjustment tank. The filter feed pumps were noted as having quite small diameter pump
suction and discharge connections, both 1-inch in diameter. Fittings such as tee's, valves, and reducers in
the fiberglass reinforced plastic resin piping cause added head loss. Scaling further reduces the cross
sectional area in the pipes and fittings, particularly in the butterfly valves. The operators have identified the
problems caused by scaling and replaced several 2" diameter pipe segments with 4' piping. In an effort to
increase flow (65 gpm vs 55 gpm) from the existing pumps, the 7.5 HP motors were removed and replaced
with 10 HP motors. Level switches in the filter feed tank control the filter pumps
Both filters operate in parallel, except when one filter is being backwashed, which occurs about 3 times per
week per filter, which corresponds to about 105,000 gallons treated per filter unit backwashed. Head loss
17
-------
through the clean filter bed is 5 psi. A high level interlock between the filter feed tank and reaction tank
feed pumps turns off the reaction tank feed pumps when the water level in the filter feed tank reaches the
high-high level in the filter feed tank. This occurs during a backwash event when all flow is directed
through one filter due to head requirements that exceed the pump flow-head capacity. Effluent from the
filters is directed to the weak acid ion exchange units. Life expectancy for the sand media is only two years
due to scaling that occurs.
4.2.4 ION EXCHANGE UNITS
The system was designed to remove metals such as iron, magnesium, manganese, aluminum, copper and
nickel down to low ppb levels as listed in Table 2. The ion exchange resin system consists of two skids
consisting of duplex resin exchangers operating in series. The first duplex resin bed is a weak acid
exchange resin (WAR) system operated in series, piped such that either unit can operate in the lead
position. The second system is a chelating resin system that operates in a manner similar to the WAR units.
The chelating resin exchangers are off line and have been since 2000. The WAR units have been able to
remove the cationic contaminants down to levels that meet discharge standards without using the chelating
resin system. The head loss through the WAR beds is 23 psi, pH into the first bed is between 9 - 10,
following the first bed the pH is roughly 7.5, and the pH following the second bed is between 5.5 and 6 at a
residual pressure of around 8 psi. Effluent from the WAR beds discharge to the intermediate pH
adjustment tank. Backwash for the resin is manually initiated and occurs approximately every 200,000
gallons. The regeneration cycle for the WAR (manufactured by Bayer) consists of the following sequence;
backwash 1200 seconds, 200 seconds settle, 1200 seconds acid rinse with 5% H2 SO4, 300 seconds slow
rinse, 600 seconds fast rinse. Since the backwash pumps have a maximum available flow of 55 gpm
(which meets slow rinse criteria) the fast rinse sequence time is simply added onto the slow rinse cycle.
The fast rinse flow rate identified by the manufacturer should be a minimum of 85 gpm. Regeneration rinse
water is discharged to a regenerant waste storage tank where it is bled back into the reaction tank over
approximately an 8-hour period. The low pH WAR effluent is adjusted by a modulating valve system that
varies the chemical feed (sodium hydroxide) through a flow paced forward feeding loop based on the pH in
the tank influent. The pH of the air stripper one influent is approximately 7 SU and following the
stripping process rises to about 8.2.
4.2.5 Am STRIPPING SYSTEM
The primary VOC removal system consists of two 30" diameter Delta packed tower air strippers with 25'
of 1-1/2" in diameter Jaeger tripack media. Pressure loss is 8 - 10 inches of water column, blowers are 25
HP Chicago Fan unit with a flow rate of 3200 scfm. A total of three blowers are provided, one per tower
and a third as a standby unit. Flow is introduced into the first tower, and collected in the sump where it is
then pumped to a second identical unit. Off gas is collected by an induced draft fan and discharged to the
atmosphere through a stack approximately 90 feet above grade. After the second tower the flow is directed
to the 1000-gallon final pH adjustment tank. These transfer pumps were designed to pump at a flow rate of
85 gpm.
4.2.6 PH ADJUSTMENT AND PLANT WATER SYSTEM
Following the air strippers, flow is directed to the 1000-gallon final pH adjustment tank. The pH is
adjusted to neutral and using a pH probe and a feed forward control loop to adjust the necessary caustic or
sulfuric acid dosage as needed. Lime is added at a rate of approximately 50 mg/L to increase the hardness
as directed by the State of New Jersey to meet the whole effluent toxicity (WET) test criteria. After the pH
adjust tank flow is directed to the 5000 gallon backwash/service water tank. Water from the backwash tank
is used as service water, chemical makeup, dilution water, flushing lines, hose stations, backwash and rinse
water for the filters and ion exchange units. Excess water is discharged to a small wetlands area that flows
into a small surface pond located on the site and eventually to Carters Brook.
18
-------
4.2.7 CHEMICAL FEED
The plant uses 93 percent sulfuric acid, lime, 50 percent sodium hydroxide, anionic polymer and 30 percent
soda ash.
Sulfuric acid is used for the following:
pH adjustment for the intermediate pH adjustment tank following the WAR exchanger,
final pH adjustment following the air slippers,
regeneration of the WAR ion exchange units,
regeneration of the chelating ion exchange units (when operating).
Lime is used to add hardness to the finished water in an attempt to match the ambient hardness in the
surface water.
50 percent sodium hydroxide solution is stored in an insulated and heated tank and is used for the following
processes:
pH adjustment in the reaction tank,
pH adjustment for the intermediate pH adjustment tank following the WAR exchanger,
final pH adjustment following the air slippers,
regeneration of Ihe chelating ion exchange unils (when operating).
Polymer is used to enhance particle formation in the following unils:
Flocculation lank,
Sludge storage lank.
30 percenl soda ash is used in Ihe reaction lank for chemical softening of Ihe groundwaler.
4.2.8 SLUDGE STORAGE AND DEWATERING SYSTEM
Sludge from the clarifier is transported to Ihe sludge thickening tank where il is mixed and further
thickened. Other sources such as Ihe flocculalion lank are also transferred to Ihe sludge storage lank for
eventual dewalering in Ihe press. The softening process along wilh olher miscellaneous processes produce
enough sludge for a single sludge cycle per month, which is equivalenl to approximately 2 cubic yards of
50 percenl solids cake. The operators typically store an adequate volume of sludge in the thickener lank so
Ihey need only renl a dumpsler for a single month Iwice yearly. Polymer is typically added to Ihe thickener
to assisl in particle agglomeration resulting in a drier cake.
4.3 MONITORING SYSTEM AND PROGRAM
The subsurface monitoring program is evaluated for adequacy in space, lime, sampling and analytical
melhodology, and dala managemenl and reporting. As discussed above, Ihe monitoring system is
somewhal sparse bolh on-sile and off. This makes il difficull to define Ihe extenl of Ihe main plume on site,
capture near Ihe extinction wells, and Ihe small plume on the soulhwesl side of Ihe site. Additional
monitoring would be useful as described in section 4.1.2. The long screened intervals for Ihe deep wells do
nol allow adequate vertical characterization of Ihe dislribution of conlaminanls.
The most significanl dala gap lies off-site. Mosl of Ihe off-site Weslbay well ports are contaminated
suggesting portions of Ihe site plumes extend soulheasl and soulhwesl of Ihe site and extinction system. Il
should be noted lhal persislenl high pH (pH>10) values have been noted in samples laken from the
Weslbay wells (over Iwo years or more following conslruclion). The cause of Ihe elevated pH values is nol
clear. Groul was nol used in the conslruclion of Ihe Weslbay wells since Ihese sampling systems were sel
in Ihe open bedrock holes and used packers to isolate sampling zones. Groul was used for setting casing
through Ihe unconsolidaled materials al Ihe lop of Ihe Weslbay boreholes, Ihough the amounls were minor
and Ihere was ample flushing of Ihe holes during the drilling into Ihe bedrock. The range of pH observed in
Ihe extinction wells (6-7.5) mosl likely represenls Ihe nalural ground water pH over similar deplhs.
19
-------
Whatever the reason for the elevated pH values, the persistently high pH values in the Westbay wells
suggest some of the ports were set in zones that do not conduct significant ground water flow. The degree
to which the samples from these ports can be considered representative of true formation conditions is
unclear. Westbay wells WB-12, WB-13, WB-14, WB-15, and WB-16 have elevated pH. Ports in Westbay
well WB-11 have pH values similar to conventional wells. The elevated organic concentrations in various
ports of the Westbay wells may not represent concentrations at the respective depths, but may be influenced
by water communicated during/following drilling of the hole but before the packers were set in the well.
Regardless, it is likely that contaminants exist off site and may be migrating downgradient toward Carters
Brook and residential wells along Route 27 southeast of the site or toward the quarry.
The conventional monitoring wells are sampled semi-annually using low-flow techniques and non-
dedicated pumps. The Westbay wells have, in the past year, been sampled quarterly, but samples are
obtained using a MOSD AX sampler that is lowered inside the Westbay casing and attaches to a sampling
port. The port is opened and a stainless steel sample chamber is filled. No purging is done. Given the
historical data now available for these wells, quarterly sampling may no longer be needed from the
Westbay wells. Samples are also collected quarterly from the sample taps on the individual extraction
wells. This sampling is done using a low-flow-like sampling methods, though the reasons for this are not
clear. Samples from active production wells are typically drawn by directly filling the bottles from a tap.
Samples are submitted for fixed lab analyses for volatile organic compounds by EPA method 8260B and
metals by EPA methods 6010, 7041, 7421, and 7470. Analyses have also been routinely conducted for
dissolved gases methane, ethene, and ethane as well as chloride, nitrite, nitrate, and sulfate. Dissolved
methane, ethene, and ethane have been non-detectible using method RSK-175. Field analyses are
conducted for dissolved sulfide, total iron, dissolved (ferrous) iron, and alkalinity.
Reports are generated to present results of each sampling round for whatever wells were sampled that
quarter. The reports provide useful information including site background, maps of water levels and
concentrations, tables of results for the wells with other recent sample results from the same wells and the
range in all historical values. The plume maps in the reports have rough computer generated contour maps
which honor the data, but do not consider site ground water flow. The analytical results are managed
electronically in Excel format. Data are provided from the lab in hard copy and not electronically.
4.4 OTHER OBSERVATIONS
4.4.1 WATER DISCHARGE ALTERNATIVES
As part of the RSE analysis, the potential for alternative discharge locations for the treated (or untreated)
water was considered. The potential to reinfiltrate the treated water into the subsurface was discounted
given the low-permeability of the near-surface materials, the heterogeneous but low bedrock permeability,
and the difficulties in maintaining injection wells in such conditions. The other alternative considered was
discharge to a sewer line connected to a publicly owned treatment works (POTW). Sewers do not serve the
Higgins Farm Site and the sanitary waste generated at the site must be removed from the site by truck (note
that two different trucks must serve the two site trailers since the connections are different). There is also
no sewer service along Routes 518 or 27.
Contacts were made with two municipal sewer departments that provide services within 1.5 miles of the
site (Franklin Township Sewer Department and South Brunswick Sewer Department) and two regional
treatment authorities (Stony Brook Regional Sewer Authority [SBRSA] and Middlesex County Utility
Authority [MCUA]) to see what potential there may be for discharge to a POTW. The site lies within the
service area of Franklin Township and, according to Mr. Bill Goodheat of Franklin Township, the nearest
sewer connection to the site would be located near the intersection of Bunker Hill Rd. and Route 27, a
distance of more than 1.5 miles. The sewage from Franklin Township is treated by the MCUA. In
discussing the requirements for discharge of materials from the Higgins Farm Site with a representative of
the MCUA, the authority would not accept ground water for a period of more than 5 years. Pretreatment
requirements would be determined on a case-by-case basis, but she gave the impression that some
20
-------
treatment would be required before they would accept the water. There are some flow constraints faced by
the MCUA treatment plant so they are trying to end discharge of ground water to the system. Costs for
treatment were very high- $7.88/1000 gallons which translates into $165,000 peryearfor40 gpm. Sewer
construction costs would be $500,000 to the nearest sewer connection, based on an estimate prepared with
the RACER estimating tool.
According to Mr. Bob Griggs of the South Brunswick Sewer Department, the nearest connection of
adequate size in their system is in Kendall Park more than a mile east of the Higgins Farm Site. However,
they have near-term plans to extend sewer service into the Little Rocky Hill area along Route 27. This
would be less than 0.75 miles downhill from the site. The SBRSA was contacted and they would be
interested in the water and would only charge $1.20/1000 gallons plus $1000/year. This translates to only
about $27,000 per year at 40 gpm. Sewer construction south of the site to Route 27 would cost
approximately $250,000 as per a RACER estimate. There are a substantial number of institutional issues
that would need to be addressed before such a discharge could occur. Since the Higgins Farm site is not
located within the South Brunswick sewer service area, the Franklin Township Board would have to
approve the connection, as would the member municipalities sponsoring the SBRSA. In addition, the
Waste Management Plan for the area would have to be modified and approved by NJDEP. Mr. Griggs did
not sound optimistic that all necessary agreements could be achieved. Such a modification to the
discharge of water would require substantial time and effort to coordinate with no guarantee of success.
4.4.2 SANITARY FACILITIES
Sewer lines do not serve the site and sanitary facilities are available only in the office trailers. Different
contractors service each trailer due to incompatible connections with the septic trucks. The servicing of the
septic systems is costly ($4300/year). There is a small sink in the treatment plant for hand washing,
although larger basin would be useful for parts cleaning.
4.4.3 OTHER SUGGESTIONS FROM OPERATORS
The operators noted that an automatic restart on the air compressor would improve the operability of the
system. A new PLC and computer would also reduce operation effort.
4.5 COMPONENTS OR PROCESSES THAT ACCOUNT FOR MAJORITY
OF ANNUAL COSTS
The system costs more than $900,000/year for the O & M contractor to operate and maintain. This
includes labor, materials, sampling and analysis, residual disposal, utilities, and up to $100,000 in plant
repairs and spare parts, but excludes U.S. Army Corps of Engineers and EPA oversight costs. The two
operators spend essentially 100% of their time on this system and the project manager spends about 50% of
her time at the site. Mr. Fite spends approximately 50% of his time on reporting and about 25% of his time
sampling. Almost 25% of the operators time is spent in maintenance of the extraction wells and pumps.
Mr. Russell had been spending about 50% of his time performing non-routine maintenance on the system to
get it to run properly. The rest of the operator's time is spent in routine maintenance of the wells and plant.
Operations and maintenance costs approximately $210,000 per year including labor, travel, office space
and supplies, equipment, and profit. Expendable materials including polymer, soda ash, caustic , acid, and
limestone and office supplies and tools ($7,200) for a total annual cost of approximately $21,000 per year.
Analyses of effluent, influent, and other plant samples (including toxicity testing) cost approximately
$59,000 including G&A and profit. Sampling of monitoring wells and analyses of the samples for from
monitoring wells cost approximately $161,000/year for labor (including travel and reporting), $38,000 in
equipment and materials, approximately $150,000/year in analytical services, and $27,000 in
hydrogeological review for a total of $376,000 or about 40% of the total project annual costs. Corrective
repair materials are budgeted at $100,000 per year. Significant efforts have been made by the operator in
the past year to improve the system to reduce this amount and it is likely that these costs will decline in the
future. Disposal of spent materials, including protective equipment and filter cake, cost approximately
$8,500 per year. The only utilities provided at the site are electricity, water, and phone services. Electricity
21
-------
accounts for most ($120,000) of the $130,000 annual cost. Phones for both USAGE oversight and
contractor use cost about $8,000/year. Water service costs about $1000/year. To summarize annual costs:
Operations:
Labor and Travel: $ 182,000
Equipment, Services: $28.000
Total $210,000
Consumable Materials (for operations): $14,000
Subsurface Monitoring: $225,000
Chemical Analyses: $220,000
Utilities: $129,000
Disposal of Residuals: $8,500
Plant Repairs, Spare Parts, Well Maintenance, etc.: $100,000 (Budgeted)
Management: $24.000
Total: -$930,000
The contract for operations considers the total flow through the treatment plant. The contractor is
motivated to keep the plant running and to pump as much as possible from the subsurface, but is not
motivated to optimize the subsurface performance. In the contract there is a budget for the operator to
repair and replace items and these funds could be used for modifications to optimize the system, but there is
no strong motivation to consider major changes to the plant. The operator has properly invested time and
effort to improve the existing treatment system performance.
4.6 RECURRING PROBLEMS OR ISSUES
4.6.1 LIMITED HYDRAULIC CAPACITY
The plant design documents indicate the calculated plant capacity was set at approximately 100 gallons per
minute (gpm). During the operation of the plant from 1998 to the present time, the maximum sustained
flow rate through the system has been limited to approximately 55 gpm. While discussing the operation
with the staff they identified several potential "hydraulic bottlenecks". The first area is the reaction tank,
flocculation tank, and clarifier area. In particular the 4" gravity line between the flocculation tank and the
clarifier. The elevation difference between the flocculation tank and clarifier water levels appears to be
inadequate to provide enough driving force (head) to transfer 100 gpm through the 4" line interconnecting
the two units. The problem is exacerbated by the number of direction changes that occur in the line as well
as the scale and precipitate that coats the interior of the line reducing the cross sectional area, and
eventually further reducing the flow capacity below 55 gpm. Various pumps throughout the plant appear to
be unable to transfer the maximum design flow through the process units as designed. This hydraulic
capacity deficiency results from several factors, small diameter pump discharge connections, undersized
piping, numerous high head loss pipe fittings and valves installed in the discharge lines. The operators
identified several of the pumps (7 of 15) as having inadequate capacity. Those pumps with sub standard
capacity include; the filter feed pumps (P-5-2-1/P-5-2-2), the final pH adjustment feed pumps (P-10-1-1/P-
10-1-2), and the ion exchange regenerant water pumps (P-10-6-1/P-10-6-2).
4.6.2 SLUDGE BUILD-UP IN THE FLOCCULATION UNIT
The operators have indicated the flocculation unit requires cleaning on a quarterly basis. The particulates
resulting from the precipitation and flocculation processes do not make their way to the clarifier prior to
settling out. Approximately 3-4 feet of sludge accumulates in the flocculent tank due to suspected
hydraulic problems in the piping between the flocculent tank and clarifier. Also a potential issue is the
quiescent conditions that occur when during periods when there is no flow through the plant allowing
particulates to settle in the basin. The small diameter interconnecting pipe and numerous bends in the flow
path to the clarifier may account for some of the sedimentation.
22
-------
4.6.3 POOR QUALITY BACKWASH WATER FOR ION EXCHANGE UNITS
Backwash water used to rinse the ion exchange units should be of high quality, in this case free of metals
and cations that might use up valuable exchange sites on the resin. The current configuration at the Higgins
Farm site requires the final conditioned water be used for the backwash/service water throughout the plant.
Unfortunately as part of an agreement with the state of New Jersey, lime is added to increase the hardness
of the water to match that of the receiving stream and improve the outcome of the whole effluent toxicity
(WET) test. By adding lime, Ca(OH)2 to the ion exchange rinse water, the actual process of regenerating
the bed actually uses up metal adsorption exchange sites on the resin since it will remove Ca++ ions from
the rinse solution. The easy solution would be addition of a separate tank downstream of the air strippers
yet upstream of the lime addition. Removing the lime from the resin rinse solution could extend ion
exchange bed regeneration intervals.
4.6.4 SULFURIC ACID PIPING LEAKS
The sulfuric acid piping was originally installed in the plant using 316 stainless steel threaded pipe and
fittings rather than black steel which is compatible with 93 percent sulfuric acid. Such stainless will rapidly
deteriorate if the moisture content of the acid rises such that it diluted to a concentration below 91 percent.
The operators have had to replace sections of the piping with a more resistant stainless alloy (alloy "20" a
high-nickel-content stainless steel). The operators discovered that the previous operators failed to change
the desiccant on the tank vent which may have allowed condensation of moisture in the tank that would
dilute the acid slightly. The operators now routinely check the desiccant that has a color indicator, and
change the desiccant on 6 months intervals.
4.6.5 CAUSTIC PIPE PLUGGING
Caustic is stored in a heated, insulated 9000-gallon tank. Caustic is fed to several points in the treatment
plant from a recirculation loop. Chemical metering is done at each use point with a bleed valve. Excess
caustic is sent back to the storage tank. When the plant is turned off, and the equalization tanks are in the
fill mode, caustic trapped in the recirculation line adjacent to the tank is heated causing the water in the 50
percent slurry to evaporate out leaving the crystalline salt which scales out in the piping and fittings that
eventually occludes flow through the pipe back into the caustic tank.
4.6.6 EXTRACTION WELL SYSTEM CONTROLS
The plant programmable logic controller controls the operation of the well pumps, based on transducer-
measured water levels. Though the transducers have not had many problems, the communications between
wells and plant have been highly unreliable. Significant expense in parts and labor are incurred in
replacing link couplings (connecting the transducer leads to the communication cables). Parts of the cable
system have failed entirely. Wells RW-6 and 7 were down at the time of the site visit due to
communications problems. The cable was buried in the trench with the water piping at depths of over 5
feet. To date, the contractor has installed transceivers to replace the cable connections in three vaults
controlling several wells. These have worked well and include automatic restart capabilities if the wells go
down due to power loss. The USAGE is negotiating for the replacement of the rest of the cables with
transceivers at the time of the site visit. The RSE team supports the replacement of the cable-based control
system as a means to reduce the maintenance load on the operators. Approximately 25% of the senior
operators' time and part of the other operator's time is consumed in maintenance of the wells,
communications, and pumps.
23
-------
4.6.7 WELL FOULING AND PUMP FAILURE
The extraction wells require periodic rehabilitation due to iron bacteria fouling. The operators perform
their own rehabilitation by recirculating a mixture of 50 Ibs of sulfamic acid and 55 gallons of hot water
through the screened interval using the dedicated pumps. Rehabilitation of each well is typically performed
twice a year. The rehabilitation does not involve any surging or brushing of the well screen and there have
not been any downhole camera surveys. Most wells cycle and are not throttled. The wells cycle up to 15-
30 times per day.
The extraction well pumps require periodic maintenance due to fouling and wear. The operators perform
much of the pump maintenance and periodically scavenge parts from other unusable pumps to replace
others. Fouled pump bodies and impellers are cleaned using a sulfamic acid bath. Motors have a limited
(roughly 2 year) life due to electrical failure, corrosion, or wear on the drive shaft. The operators have not
found a more durable alternative.
4.7 REGULATORY COMPLIANCE
Discharge permit levels for the water are addressed in Table 2. Historically, test results have shown the
plant is consistently reducing the volatile organic compounds and metals levels to below the discharge
criteria. An analysis of the influent metals concentrations at the Treatment Plant indicate with the
exception of iron and manganese, the levels are already below the discharge criteria. Clean up standards
were compared to ecotoxicity and screening levels in Table 3. In all cases the calculated influent
concentrations are below the ecotoxicity screening levels, except 1, 2 DCA, 1, 1, 2, 2 PCA, PCE, TCE and
VC exceed the Region III RBC's, and Region IX PRG's. 1, 1, 2, 2 PCA, PCE, and TCE also exceeded
MCL's. Many of the discharge limits were noted to be quite conservative and should perhaps be
reevaluated.
No compliance issues related to plant operations were noted except difficulty in meeting whole effluent
toxicity (WET) test parameters. Plant personnel coordinating with the state regulators have corrected the
problem.
4.8 TREATMENT PROCESS EXCURSIONS AND UPSETS,
ACCIDENTAL RELEASES
4.8.1 TREATMENT PROCESS EXCURSIONS AND UPSETS
As identified in section 1.5 Site Location, History and Characteristics, the plant was installed in 1997 and
began operations in May 1998. During the initial 12 months of operation by the construction contractor the
plant had several operational problems. Significant issues related to the operation of the softening, ion
exchange systems, and general hydraulic capacity limitations. The current operations contractor has been
systematically diagnosing and as funds allow correcting these chemical feed, and hydraulic deficiencies.
4.8.2 ACCIDENTAL CONTAMINANT RELEASES
There have been three accidental releases of extracted ground water that were not contained within the
plant. In each case, the mass of contaminant released was determined to be below the reportable quantity.
During installation of fencing at RW-14 in April 2001, a contractor punctured the piping. Though the pipe
was subsequently repaired, it was not pressure tested. In July 2002, the equalization tank overflowed and
allowed water to flow outside the building. The pump check valve on extraction well 10 stuck open on one
occasion in June 2003, allowing water from the common discharge header to flow back into the well
similar to an injection well and eventually overflowed the vault.
24
-------
4.9 SAFETY RECORD
There have been no lost-time accidents reported during the plant operations. Housekeeping practices and
general condition of the facilities, including the wells are commendable.
25
-------
5.0 EFFECTIVENESS OF THE SYSTEM TO PROTECT HUMAN
HEALTH AND THE ENVIRONMENT
5.1 GROUND WATER
The protectiveness of the ground water extraction systems depends on the containment of the plume and
the prevention of impacts to production wells or surface water. The extraction system is probably capturing
the on-site plumes, though this is subject to some uncertainty as described in section 4.1.2. There are no
known impacts to surface water or existing active water supply wells near the Higgins Farm Site, though
there has been no recent offsite monitoring of surface water or domestic wells southeast of the site or of
seeps at the nearby quarry. Though the extraction system is likely to be capturing the on-site plume, the
current offsite extent of contaminants is unknown. As such, the protectiveness of the ground water
remedies is unknown. The main portion of the plume extends in a direction toward an area with known
active domestic wells and no current alternate water supply. Downgradient domestic wells are likely to be
constructed to depths comparable to or possibly deeper than on-site extraction wells.
5.2 SURFACE WATER/WETLANDS AND SEDIMENTS
The EPA following consultation with a biologist has designated several areas of the site as wetlands.
Though at least two of the wetlands were dry at the time of the site visit, these areas are seasonally wet.
There is no indication of any site contaminant impacts to these areas, nor is there evidence of impacts due
to ground water extraction at the site. The cattle at the site have substantially trampled the wetlands within
the pastures used for grazing. The discharge of the treated ground water to a small pond appears to have
enhanced wetland and aquatic habitat at the site. The discharge has also provided a consistent flow in
Carters Brook, though this was not verified during the site visit. The plant has maintained compliance with
the discharge requirements and should not be impacting surface water quality; however, the influent
concentrations are substantially less than expected. There is a potential for contaminated ground water to
discharge to surface water southeast and southwest of the site into the Carters Brook drainage and the
nearby quarry, but there are no data to determine if these impacts are occurring.
5.3 AIR
The plant influent is treated using air strippers and no off gas treatment is provided. Influent volatile
organic compound concentrations are approximately 100 ug/L and total extraction flow rates are around 40
gpm. This translates to less than 0.1 pound per day, primarily of TCE and PCE. Given the low mass
loading on the atmosphere and the rural location, there is no significant impact on air quality.
26
-------
6.0 RECOMMENDATIONS
6.1 RECOMMENDATIONS TO ENSURE EFFECTIVENESS
6.1.1 PLUME DEFINITION & POTENTIAL IMPACTS TO GROUND WATER USERS
The off-site extent of the contaminant plumes is unknown and there is a potential for impacts on ground
water users, particularly southeast of the site. There is an urgent need for the evaluation of current ground
water concentrations off-site and in domestic wells southeast of the site. A reconnaissance survey of
domestic wells southeast of the site along Route 27, including well depth, pump rate, and water use, should
be conducted and the sampling of selected wells based on the results should be implemented soon. The off-
site extent of the plume should be further defined by a number of monitoring well clusters. In particular,
wells are required southeast of Westbay wells WB-12, -13, and -14. For estimating purposes, three well
clusters are recommended. Given the predominance of solvents in ground water, the sampling program
should include a one-time evaluation of the presence or absence of 1,4 dioxane.
6.1.2 SURFACE WATER SAMPLING
Surface water sampling for VOCs in Carters Brook and any other tributary that may capture contaminated
seepage is recommended. If possible, samples of the seepage into the northeast face of the Trap Rock
quarry should also be obtained. If contamination is detected, the potential for human exposures and the
degree of ecological impact should be determined. Assuming a two person team collects 4 samples over a
one day period, the anticipated cost is approximately $2,500. (8 hours x $60/hr = $480, VOC analysis 5 x
$250/sample = $1250, planning 10 hours x $60/hr = $600).
6.1.3 ON-SITE PLUME CAPTURE
Though it seems likely that the on-site plumes are being captured by the existing system, a new extraction
well (RW-11 A) between RW-11 and RW-12 is strongly recommended to assure the complete capture of
the main plume. This location would fill a 250-foot-wide gap at the axis of the main plume and is likely to
capture additional ground water contaminated with high concentration of VOCs at a reasonable flow rate,
given the relatively high flow rates of the nearby wells. Costs are based on a RACER estimate of well
design and construction. RACER is a parametric cost modeling system that uses a patented methodology
for estimating costs. RACER cost technologies are based on generic engineering solutions for
environmental projects, technologies, and processes. The generic engineering solutions were derived from
historical project information, industry data, government laboratories, construction management agencies,
vendors, contractors, and engineering analysis. RACER incorporates the most technologically up-to-date
engineering practices and procedures to accurately reflect today's remediation processes and pricing. The
new well would cost approximately $60,000 exclusive of oversight costs or contingencies. This assumes a
new well similar in design to existing wells.
Additional monitoring points should be installed between, upgradient of, and downgradient of the
extraction wells capturing the major site plumes to clarify the capture zone extent. At least eight additional
monitoring points installed to approximately 20 feet into bedrock would be adequate. These should be
installed near the main plume between RW-10A and RW-11, between the proposed RW-11A and RW-12,
between RW-10 and RW-10A, downgradient of RW-11, and upgradient of the proposed RW-11A. There
should also be additional monitoring points between RW-7 and RW-8, RW-8 and RW-9, and downgradient
27
-------
of RW-8. Downgradient wells should be placed approximately 75-100 feet downgradient of the extraction
well line, based on a computed stagnation point assuming typical site parameters. Note that if additional
investigations determine a significant off-site extent of the site plumes such that alternative remedial action
objectives are considered, the installation of these monitoring points should be postponed until the utility of
the current extraction system is verified. These monitoring points are estimated to cost approximately
$55,000 based on a RACER estimate and a 10% design cost. No contingencies or oversight costs are
included. These new wells would be sampled quarterly for the first year (32 samples at 2 person-hours per
well). This would increase monitoring costs approximately $20,000 for the first year. During subsequent
years, these wells would only need monitoring semi-annually.
6.2 RECOMMENDATIONS TO REDUCE COSTS
6.2.1 INVESTIGATE BYPASSING THE GROUNDWATER TREATMENT PLANT
Given the low influent concentrations of the contaminants of concern, both volatiles and metals, the
groundwater treatment facility could likely be taken out of service and extracted groundwater discharged
directly to a POTW. Analysis shows the total volatile organic compound concentration is in the 100 to 120
ppb range, consisting primarily of TCE and PCE, while the metals concentrations generally meet drinking
water standards with the exception of iron and manganese. The anticipated flow rate would be
approximately 30 gallons per minute. Flow would need to be pumped northwest of the existing plant to
Washington Street (Route 518) and then flow by gravity northeast for approximately 1 mile to the
intersection of Highway 27 where a second lift station would be required. The lift station would discharge
to a manhole approximately 1000 feet away where the line will flow by gravity for another 2000 feet to a
Middlesex County Utility Authority manhole at the intersection of Highway 27 and Bunker Hill Road.
Costs associated with this option include O&M fees for treating the water prior to discharging to the
authority of about $ 122,000 annually, and a one-time capital cost based on the RACER estimating tool of
$500,000 for 2 lift stations, 3000 feet offeree main, and 1.4 miles of 8-inch diameter gravity sewer.
Significant cost reductions for operations, maintenance, sampling and analysis, chemicals, plant heating
and power would be realized. Health and safety concerns would also be lessened due to elimination of
chemical feeding and handling and confined space entry while cleaning tanks. The estimated total
reduction in current contract operation and maintenance costs was estimated at 60 percent. Approximately
$320,000 per year could be saved bypassing the Treatment Plant and discharging directly to the Middlesex
County Utility Authority.
There are significant institutional hurdles to be overcome for discharge of site ground water to a POTW to
occur. The Middlesex County Utility Authority has flow limitations and is only granting temporary
permits for ground water discharge as proposed here. This alternative will only be appropriate for a
maximum of 5 years, unless a permanent arrangement could be negotiated with the Authority. If water
would be discharged to the South Brunswick Sewer and treated at the Stony Brook treatment plant,
significant negotiations would be required with Franklin Township, the SBRSA, and the SBRSA member
municipalities. Modifications to the waste management plan for the area would have to be developed and
approved by the state. The costs for these negotiations would be substantial but have not been estimated.
There is no guarantee that such negotiations would be successful. Capital cost for this option based on the
RACER estimating tool is approximately $250,000. Annual fees would be approximately $27,000.
Approximately $420,000 per year based on cost avoidance using current operations contract costs, could be
saved bypassing the Treatment Plant and discharging directly South Brunswick Sewer and Stony Brook
Treatment Plant.
6.2.2 EXTRACTION SYSTEM REVISIONS
A significant amount of time and effort is expended in maintaining the twenty on-site extraction wells.
Since many of the wells produce very low to non-detectable concentrations of contaminants at low to
moderate flow rates, a number of these wells could be eliminated from the system without sacrificing
effectiveness of capture. Based on the concentrations, the cessation of pumping wells RW-1, -2 (if
concentrations remain low), -3 A, -4, -5, -6, -8A, -9A (if concentrations remain low), and -13 (if
28
-------
concentrations remain low) is recommended. This would reduce well electrical use by approximately 30%
(about $150/month or $l,800/year), reduce well maintenance hours by 40% (reduce labor by approximately
10% of one operator's time = 0.10*2,000 hours/year * $56/hour with overhead and profit = -$11,000), and
reduce sampling costs for extraction wells (assume 8 wells * 4 times/year * $400/sample analytical =
-$13,000 year, neglecting sampling labor and equipment). The reduction of the flow into the plant would
also reduce the cost of chemicals. The flow would be reduced by about 25% (assuming a new well is
brought on line between wells RW-11 and RW-12) so materials and sludge disposal costs would be
decreased by 25% or $5,500/year (based on cost avoidance using current operations contract costs). The
total cost savings would be approximately $31,0007 year.
6.2.3 REDUCE THE SIZE OF THE Am STRIPPING SYSTEM
Caveating our recommendations as identified in 6.2.1, analysis of the projected influent indicates the
existing air stripping process consisting of 2-30" diameter packed towers each equipped with 25 HP
blowers delivering 3200 cfm per tower, far exceeds the treatment requirements for removing relatively low
concentrations of a variety of volatile organic carbon compounds. The primary VOC compounds present
are: TCE, PCE, 1, 2 DCE, 1, 1, 1 TCA, 1, 1, 2 TCA, benzene, chlorobenzene, 1, 2 DCA, vinyl chloride,
and 1, 1, 2, 2 PCA with concentrations of 34, 67, 4, 0, 5, 0.3, 2, 2, 0.2 and 6 ppb respectively. The most
difficult compound to remove is the 1, 1, 2, 2 PCA.. Calculations based on generic and proprietary
software indicates a single packed tower 24 inches in diameter containing 20 feet of 1.25" packing at a
water flow rate of 35 gpm and air flow rate of 600 cfm will remove the VOC contaminants of concern to
levels below the discharge criteria. A second option using a tray type air stripper would require a 900 cfm
7.5 HP blower and 6 trays. And finally a third option would involve taking one of the existing stripper
units off line altogether and replacing the existing 25 HP, 3200 cfm blower in the other stripper with a
smaller 1200 cfm, 10 HP blower. Approximate capital costs based on vendor quotes and typical
installation costs are as follows: a new packed tower air stripper $40,000, a new tray type unit $25,000 and
shut off one of the two strippers, and retrofit the remaining 30" diameter unit with a new blower $5000.
Annual cost savings based on power savings alone (reduction of HP from the two strippers from 50 to 10)
would be approximately $17,000.
6.2.4 EVALUATE ALTERNATIVE MANGANESE REMOVAL TECHNOLOGIES.
Notwithstanding our recommendations as identified in 6.2.1, a different iron and manganese removal
process other than the current softening, filtration, and ion exchange system would be much simpler,
compact, and economical. Data from the last year of operation identifies iron and manganese are the only
metals of concern, which exceed the effluent criteria of 0.3 and 0.05 mg/L (the secondary drinking water
standards) prior to discharge. Doing away with the current metals removal system that was intended to
remove a variety of heavy metals will eliminate several processes, namely the reaction tank, flocculation
tank, filter, and ion exchange systems, as well as all the chemical feed and hydraulic problems associated
with them. The proposed system subjects the influent to a 3 - 5 ppm dose of potassium permanganate
(KMnO4) or other oxidant such as chlorine, which converts the iron and manganese to its insoluble
oxidized forms. After iron and manganese is oxidized, the stream comes in contact with the specially
treated media or a natural manganese greensand media that adsorbs the iron and manganese, and filters the
water. The systems are designed to remove 3-5 times the iron and manganese concentrations present in
the Higgins Farm influent stream and produce an effluent with iron and manganese concentrations below
0.3 and 0.05 mg/L respectively. The system would require backwashing about twice a week. The common
water backwash should be supplemented with an air scour step that can be used in the event biosolids
develop on the media. Backwash water would be directed to the existing clarifier, solids allowed to settle
and water decanted back to the existing equalization tanks. Sludge volume generated would be small and
could likely be removed periodically using a vacuum truck. The system could be installed near the air
stripping system, bypassing the softening, filtration, and ion exchange processes. The pumps currently
used to feed the reaction vessel may be suitable for use as the greensand filter feed pumps. Effluent from
the greensand filter would discharge to the intermediate pH adjustment tank that feeds the air stripper. The
cost for installation including control and piping modifications based on vendor quotes and typical
installation costs would be approximately $50,000. Annual cost for oxidant addition would be less than
$2,000. A one-week pilot test using small-scale equipment on site would cost between $6,000 and
29
-------
$ 10,000. Annual cost savings based on cost avoidance using current operations contract costs, from
bypassing the existing softening, ion exchange, and pH adjustment units would come from reductions in
chemicals $11,200, sludge disposal $4,300, repairs $3,000 and reduction in plant labor by 40% or $66,500
respectively for a total savings of approximately $85,000 per year.
6.2.5 CHANGES IN MONITORING PROGRAM
The RSE team has evaluated the monitoring program and costs. The contract costs for the monitoring
(both labor and analytical) appear approximately 40% high, based on conservative estimates of analytical
costs and sampling time. The RSE team suggests obtaining competitive bids for acquiring monitoring well
and extraction well samples from contractors located closer to the site. The team also recommends seeking
a laboratory independent from the operations contractor to avoid a potential conflict.
Extraction Well Sampling. The quarterly sampling of the extraction wells is appropriate, provided the
operations of the wells are adjusted based on the results. The sampling method applied to the extraction
wells should be modified to just include sample collection from the tap with annual measurement of pH,
temperature, conductivity, oxidation-reduction potential, dissolved iron, and dissolved oxygen using field
kits and a flow-through cell and sensors. This will save labor, though the amount is not certain. Assuming
20 wells and a 1/3-hour labor savings per well, at $50/hour for sampling labor, this amounts to a savings of
$330/round or approximately $l,300/year.
Monitoring Wells Sampling. The Westbay wells should be sampled only semi-annually at some point in
the near future. Since drought conditions existed for some of the period following installation, this change
can be implemented once the impact of the relatively recent return to more normal precipitation on
contaminant concentrations, if any, is identified. Given the almost four-year history of sampling from these
wells, the minimal observed rate of change in concentrations over time, and assuming no dramatic change
is observed in the next six months due to a return to more normal precipitation, the ability to identify
concentration trends would not be impacted by reducing the frequency to twice yearly. This would result
in an immediate drop of 36 samples per year (20% reduction in the analytical costs for monitoring and 72
person-hours or 25% drop in labor). This may save as much as $56,000 per year, based on applying these
percentages to the current costs for analytical and sampling labor. This savings would be offset during the
first year by additional sampling of new monitoring points as described above. Based on the existing data
on natural attenuation conditions at the site, the RSE team recommends the analysis for dissolved gases
(methane, ethane, and ethene) and ions such as sulfate and chloride be terminated or reduced to biennial
sampling. The representativeness of many of the Westbay sampling ports needs to be re-evaluated. Those
wells/ports displaying persistent pH values over 7.5 should be re-purged to assure the interval actually
yields water. Any port that is found to have very low yield should be dropped from the monitoring
program. Costs or savings associated with this have not been estimated. Lastly, strings of diffusion bag
samplers should be implemented in the deep conventional monitoring wells to determine the vertical
distribution of contaminants. This would have a one-time cost of approximately $14,000 assuming 5 bags
per each of the 8 deep wells at a cost of $300 per sample and 32 person-hours at $50/hour for acquisition
and sampling. These costs are based on past experience with similar sites using comparable equipment.
The results would clarify the site conceptual model and better focus low-flow sampling depths in these
wells.
6.2.6 REDUCE SITE OVERSIGHT BY USACE
Additional cost savings are possible by reducing the level of effort of the on-site USACE representative
from 50% time on-site to one day per week (20%). As the operators have improved the performance of the
system, the level of needed oversight has diminished. There may be times when additional oversight is
needed, especially if some of the recommendations of this RSE are to be implemented, but during routine
operations, the oversight could be reduced. If the level of complexity of the treatment plant is significantly
reduced as recommended above, the oversight could be reduced further. As a related item, during the next
contract action for operations, an effort to critically evaluate the need for the large number of submittals
should be made. Some should be eliminated, others combined to reduce level of effort in managing these
submittals. Examples include: O&M Manual updates should be submitted when changes or adjustments
30
-------
occur, rather than submitting the document quarterly even if no changes are made; Quality Control
Summary Report Information could be reported at monthly progress meetings and submitted bi-annually or
as required instead of monthly; Part J Certified Payroll could be one submitted monthly with payments
lagging one week rather than submitted weekly. These combined actions could reduce oversight costs by
more than 50% and save approximately $35,000 annually (based on the fully burdened rate for a GS 12
step 5 engineer of approximately $80/hour).
6.2.7 REPLACE TRAILER WASTE HOLDING TANK PUMP OUT CONNECTION
Each of the on site trailers is equipped with a waste holding tank. Unfortunately they were installed with
different pump out connections. These incompatible connections require a separate monthly trip from
different contractors because the trucks do not come equipped with both fitting types. Changing the pump
out connection fitting would reduce the number of contractors, and require only a single trip monthly to
empty the contents of both waste-holding tanks. Estimated cost to change the connection would be
approximately $500 and save approximately $1,000 per year (elimination of one contractor visiting the site
monthly 12 x $115 less added cost growth for the second contractor to empty the tank). These costs are
based on past experience with similar sites using comparable equipment.
6.3 RECOMMENDATIONS FOR TECHNICAL IMPROVEMENT
6.3.1 SUGGESTED CHANGES IN WELL MAINTENANCE METHODS
The extraction wells are currently rehabilitated about every six months. The operators mix a hot sulfamic
acid solution and recirculate it through the well without agitation or brushing. The use of sulfamic acid is
appropriate, but a combined effort using sulfamic or hydroxacetic acid with a dispersant and well swabbing
is likely to be more successful in maintaining good flow for a longer period of time. There are
commercially available pre-mixed products with these compounds. A downhole camera survey of a fouled
well may help identify if the problem is localized within the screen or if the fouling is at or adjacent to
pump. The project team is directed to the USAGE guidance, EP 1110-1-27, on well maintenance (see
http://www.usace.army.mil/inet/usace-docs/eng-pamphlets/epl 110-1-27/toc.htm). Additional guidance on
well rehabilitation is provided in soon-to-be-issued USAGE documents.
6.3.2 PIPING TESTING
Given the long piping runs and the past penetration of a portion of the piping during fence installation, a
pressure test of segments of the piping may be appropriate to verify integrity. Pressure drops along the
piping should be measured and compared to theoretical pressure drops as an indication of encrustation,
fouling, or sedimentation. This may be appropriate given the low velocities due to low average well flow
rates. No estimate of the cost of this activity is provided.
6.3.3 CONTROL SYSTEM MODIFICATIONS
The plant programmable logic controller controls the operation of the well pumps, based on transducer-
measured water levels. Though the transducers have not had many problems, the communications between
wells and plant have been highly unreliable. Significant expense in parts and labor are incurred in
replacing link couplings (connecting the transducer leads to the communication cables). Parts of the cable
system have failed entirely. Wells RW-6 and 7 were down at the time of the site visit due to
communications problems. The cable was buried in the trench with the water piping at depths of over 5
feet. To date, the contractor has installed transceivers to replace the cable connections in three vaults
controlling several wells. These have worked remarkably well and include automatic restart capabilities if
the wells go down due to power loss. The RSE team supports the replacement of the cable-based control
system as a means to reduce the maintenance load on the operators. Control system modifications should
be coordinated with the findings in paragraph 6.2.2.
31
-------
6.3.4 OTHER ISSUES
The operators recommended providing auto-restart capability for the plant air compressor and the
installation of a larger sink in the plant to facilitate washing of parts, etc. The RSE team supports these
recommendations as a means to improve operationally efficiency.
6.4 MODIFICATIONS INTENDED TO GAIN SITE CLOSEOUT
6.4.1 EVALUATION OF THE ADEQUACY OF PAST SOURCE REMOVAL
A key to achieving ground water quality standards on the Higgins Farm Site is the removal of contaminant
sources. The regulatory agencies and the site owner have undertaken several removal operations. The
degree to which these removal actions were able to remove all long-term sources, including any product
that leaked from the drums and containers, was not clear in the information made available to the RSE
team. The RSE team recommends a more thorough evaluation of the past actions to assess the
effectiveness as a true source removal. If it appears there may still be significant contamination below the
water table, consideration of additional efforts to reduce mass flux from these areas should be considered.
It is encouraging that concentrations of some of the contaminants have declined substantially, especially
benzene. If a substantial mass remained in the source areas, one may not expect such declines. In any
event, it is likely on-site ground water concentrations will remain above the ground water quality standards
for a significant time to come (perhaps several decades).
6.4.2 SITE EXIT STRATEGY DEVELOPMENT
The 1992 ROD identifies remedial goals, as described in section 3.1, however the site does not have a
detailed incremental exit strategy. There is not a process to compare site conditions to specific interim
goals to quantitatively evaluate progress toward site closure goals, nor is there a process to evaluate the
concentrations in water produced by specific extraction wells to determine the need for continued
extraction from those wells. A site exit strategy should be identified that sets goals for reduction in
concentrations at specific locations based on the goals for human health and environmental protection.
These goals could be to achieve specific concentration reductions at specific locations. It may be that on-
site concentrations may eventually fall due to current actions, additional actions to address sources (if
appropriate) and natural processes to low levels such that natural processes alone could prevent
downgradient exposure and no active remediation may be needed. Interim decision points could be
developed that indicate when active extraction is no longer needed at specific locations within the
extraction system, similar to the recommendations in section 6.2.3. Such reductions in flow may reduce
future O&M costs. The interim decision points could also recommend more aggressive removal (e.g.,
additional wells, source removal) be attempted if concentration goals are not achieved or implement
contingency actions if concentrations exceed a certain threshold at monitoring points upgradient of water
supply wells. The exit strategy should include evaluation of the monitoring program to assure it is focused
on gathering the data needed to make the decisions. Periodic independent reviews (and, of course, five-
year reviews) are also part of the exit strategy. The exit strategy could be developed following the
evaluation of off-gradient contaminant distribution.
Some examples of exit strategy points relevant to the Higgins Farm Site are discussed below. It is
suggested that extraction wells be turned off when the extracted concentrations are below MCLs or when
the extracted mass represents less than 1% of the extracted mass of the system. If the iron concentrations
were to fall below some minimum value in the combined influent to the plant, the process for pre-treatment
for iron should be discontinued in the plant. If influent concentrations decline to less than the discharge
limits, the treatment plant should be bypassed entirely. The ground water extraction system should reduce
32
-------
concentrations in the plumes at some reasonable rate. Perhaps some exponential decline can be identified
based on historical data and projected into the future for specific wells. If the future concentrations for
wells such as MW-7D, MW-6S, and Westbay well 11 asymptotically approach a number more than 20%,
as an example, above the projected trend, then some additional treatment of the higher concentrations
remaining near the past disposal locations would be considered to promote further decline in concentrations
in the plume. The addition of extraction wells may be considered if the concentrations in reliable
downgradient Westbay monitoring points do not drop by (an assumed) 50% in the next two years since the
lack of a significant decline suggests there is leakage past the existing wells. Finally, if sampling of off-site
and downgradient supply wells indicates concentrations of site contaminants above 50% of the MCLs or
other risk based numbers, some additional monitoring wells should be installed upgradient of the supply
wells as sentinel wells and contingencies for additional extraction and/or treatment (e.g., well head
treatment) considered. These are only suggestions offered as a starting point for the project team. The
actual exit strategy must be determined through consensus building and may require modeling or other
studies to actually develop trigger or target concentrations for the strategy.
6.5 SUGGESTED APPROACH TO IMPLEMENTATION OF
RECOMMENDATIONS
The characterization of the off-site contamination needs to be pursued regardless of decisions about other
changes. The decision for adding a new extraction well, RW-11 A, and the installation of additional
monitoring points near the existing extraction wells should not be made until the downgradient extent is
clarified. The replacement of the existing metals treatment processes should be evaluated as other changes
are considered. This recommendation should receive the most careful consideration due to the highly
significant decrease in O&M cost even while maintaining effectiveness. Most of the other
recommendations are independent of each other and can be evaluated as individual actions. These include
development of an exit strategy, reduction in the number of active extraction wells, modifications to the
monitoring program, reduction in the air stripping process, addition of a sink in the treatment plant, change
in the well maintenance methods, change in the waste tank pump-out connections. Note that the reduction
in the number of active extraction wells and influent flow rate may perpetuate the need for batch operation
of the existing treatment processes. The upgrade of the communication and controls to the extraction wells
should not be finalized until a decision is made as to the reduction in the number of extraction wells.
6.6 RECOMMENDATIONS FOR SITE MONITORING AND
MONITORING REPORTS
The monitoring reports currently provided to USAGE for the ground water extraction and treatment system
at the site provide much useful information. The RSE team does not have many specific recommendations
for the changes in the reports. Some evaluation of the well specific capacity as an indication of fouling
problems should be provided on a quarterly or semi-annual basis. This would include indication of well
average flow rates and extraction well dynamic water levels.
6.6.1 MONTHLY REPORTS
Currently, monthly reports are submitted to the USAGE, though the reports could possibly submitted
quarterly with a significant cost savings. Assuming 10% of Mr. File's time and 5% of Ms. Cermac's time
are devoted to monthly reports, this may be about $7,000 per year (based on interviews and cost avoidance
using current contract costs. At a minimum, the operational reports for the treatment plant should present
(items apparently not currently reported are indicated by an asterisk*):
• Current operating status (operating at full capacity, limited capacity, or off)
• Operational time over the quarter (as percentage), descriptions of problems encountered
(primarily those that resulted in shut down or exceedances in the effluent), and corrective
actions taken.
33
-------
• Brief description of major routine maintenance conducted during the month (e.g., carbon
changes, filter backwashes, etc)
• Concentrations in effluent and influent, including results of sampling at intermediate points
• Average flow rates from each well over the time period along with total water extracted*
• Current draw (amperage) for each ground water pump as an indication of pump health*
• Water level measurements for each production well*
• Specific capacity estimates for each well*
• Water levels measured in prescribed monitoring points. * This data should be presented in a
table and posted on a map of the well locations. No contours need be provided.
• If sampling of monitoring wells is conducted during the month, the results should be tabulated
and key constituents posted on a map of the monitoring wells. This would include low-flow
sampling parameters such as turbidity and DO. No contours need be provided.
• Descriptions of quantities consumable chemicals used
• Major maintenance items planned for the next quarter
• Appendices of lab reports, chain of custody forms, and field sampling records
• Recommendations for equipment replacement / repair, well maintenance or rehabilitation,
process changes or optimization should be identified (approximate costs provided for major
items)
6.6.2 MONITORING REPORTS
The monitoring report for the systems should be provided semi-annually and should include the above
information along with a brief description of the site (history, hydrogeology, land use including any
changes, short description of the extraction and treatment system) and a site conceptual model. Much of
the monthly data water level information can be summarized, though if data is being reported for a month
not covered by previous reports, the information should not be summarized. The results of the most recent
monitoring well sampling should be compared to historical observations to determine trends or future
problems. This is currently and nicely done in the tables, but could be provided for key wells and
contaminants in graphical form. Any figures from monthly reports should be provided along with
interpretive text (and contaminant and ground water contours). Graphs of flow rate and concentration, and
mass removal rate, over time by extraction well are appropriate and should include historical data to show
long-term trends. Similar graphs should be prepared for influent, effluent, and total system mass removal.
A specific discussion comparing system performance to requirements (e.g., for effluent, plume capture, site
closure) must be included.
34
-------
7.0 SUMMARY
The observations and recommendations given below are not intended to imply a deficiency in the work of
either the designers or operators, but are offered as constructive suggestions in the best interest of EPA and
the public. These recommendations obviously have the benefit of the operational data unavailable to the
original designers.
The RSE process is designed to help site operators and managers improve effectiveness, reduce operation
costs, improve technical operation, and gain site closeout. In this report, several recommendations are
made with respect to system effectiveness, cost reduction, and technical improvement. The report
addresses potential ways to enhance remediation, improve reporting and data management.
The ground water extraction system is generally operating in a way that achieves containment of the
contaminant plume although additional wells are recommended to verify plume capture. The groundwater
extraction system has reduced the groundwater contaminant concentrations over the period of operation.
Several wells have concentrations at or below cleanup levels and the RSE team recommends they be turned
off, however one new extraction well is recommended along the property boundary between existing wells
RW-11 and RW-12. The treatment plant is achieving treatment goals, but has experienced operational
difficulties due to limited hydraulic capacity related to undersized piping and pumping systems. The plant
was originally designed to treat a variety of metals, which have diminished in concentration below cleanup
and discharge levels except for iron and manganese. The existing metals treatment system is much too
complex than currently needed and should be replaced. Treated ground water currently discharges to
Carters Brook but other options, namely discharge to a POTW should be investigated. Significant savings
in monitoring frequency and analytical costs were also identified.
Table 7 summarizes recommendations for improvements in effectiveness, and cost reduction, which will
also assist the customer gain site closeout.
-------
Table 7. Cost Summary Table for Individual Recommendations
Recommendation
6.1.1 Plume Definition
6.1.2 Surface Water
Sampling
6.1.3 On-Site Plume
Capture
6.2.1 Bypass Treatment
Plant to POTW
6.2.2. Extraction
System Revisions
6.2.3 Downsize Air
Stripper Blower
6.2.4 Alternate
Manganese Removal
Technologies
6.2.5 Changes in
Monitoring Program
6.2.6. Review Level of
USACE Oversight
6.6.1 Reduce Monthly
Reporting Frequency to
Quarterly
Reason
Effectiveness
Effectiveness
Effectiveness
Cost
Reduction
Cost
Reduction
Cost
Reduction
Cost
Reduction
Cost
Reduction
Cost
Reduction
Cost
Reduction
Estimated Change in
Capital
Costs
$95,000
$2,500
$115,000
$500,000 to
MCUA, or
$250,000 to
SBRSA
$0
$5,000
$50,000
plus $7500
pilot test
$14,000 for
diffusion
bag
sampling
one time
$0
$0
Annual
Costs
$0
$0
$20,000
(first year)
to $10,000
Savings
($320,000)
(MCUA)
($420,000)
(SBRSA)
Savings
($31,000)
Savings
($17,000)
Savings
($85,000)
Savings
($56,000)
(accounting
for new
mon. wells)
($35,000)
($7,000)
Life-cycle
Costs*
$95K
$2,500
$325K
($5.9M)
($8.1M)
($620K)
($100K)
($1.64M)
($1.12M)
($700K)
($140K)
Costs in parentheses imply cost reductions, refer to the paragraph cited for cost development information.
* assumes 20 years of operation at a discount rate of 0% (i.e., no discount).
36
------- |