Considerations for Applying
the Triad Approach
Hartford Area Hydrocarbon Plume Site
Hartford, Illinois
-------
EPA 542-R-06-008
United States Office of Solid Waste and August 2007
Environmental Protection Emergency Response www.epa.gov
Agency (5102G) clu-m.org
Considerations for Applying the Triad Approach
Hartford Area Hydrocarbon Plume Site
Hartford, Illinois
Prepared by
U.S. Environmental Protection Agency
Office of Superfund Remediation and Technology Innovation
Superfund Triad Support Team
In cooperation with:
U.S. Environmental Protection Agency Region V
-------
Notice
This report was funded U.S. Environmental Protection Agency (EPA) Office of Superfund Remediation
and Technology Innovation (OSRTI) under Contract Number 68-W-02-034. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
A limited number of copies of this report are available free of charge from the National Service Center for
Environmental Publications (NSCEP). Refer to document number EPA 542-R-06-008 when ordering.
Order via the Web site, by mail, or by facsimile from:
EPA/ National Service Center for Environmental Publications
P.O. Box 42419
Cincinnati, OH 45242-2419;
Telephone (800) 490-9198
Fax: (301)604-3408
This document can also be obtained electronically through EPA's Clean Up Information (CLU-IN)
System on the World Wide Web at http://cln.in. org.
Comments or questions about this report may be directed to Stephen Dyment, EPA, OSRTI (5203P),
1200 Pennsylvania Avenue NW, Washington, D.C. 20460; telephone: (703) 603-9903; e-mail:
dvment. stephen@epa. gov.
-------
Foreword
This document is one in a series designed to provide information about innovative technologies and
approaches that support less costly and more representative site characterization. These documents
include reports about new technologies as well as novel applications of familiar tools or processes. They
are prepared to offer operational experience and to communicate information about ways to improve the
efficiency of data collection at hazardous waste sites.
Acknowledgments
Special acknowledgement is given to EPA Region 5, other federal and state staff and remediation
professionals, and to the staff of Tetra Tech EM Inc. for their support in preparing this document.
-------
CONTENTS
Section Page
ACRONYMS AND ABBREVIATIONS vi
1.0 INTRODUCTION 1
1.1 THE TRIAD APPROACH 1
1.2 BACKGROUND 2
1.3 PROJECT OBJECTIVES 3
1.4 PRINCIPAL STUDY QUESTIONS 3
2.0 PRELIMINARY CONCEPTUAL SITE MODEL 5
2.1 CURRENTLY IDENTIFIED REGIONAL DATA GAPS 6
2.2 GENERAL SUGGESTIONS FOR FILLING REGIONAL DATA GAPS 6
2.3 GEOLOGY, HYDROGEOLOGY, AND CONTAMINANT DISTRIBUTIONS 7
2.3.1 ROSTResults, Contaminant Transport, and Source Areas 8
2.3.1.1 ROST Results and Product Recovery Challenges 8
2.3.2 Sand and Clay Isopach, Formation Top, and Free Product Thickness Maps 9
2.3.2.1 North Olive Sand Maps 10
2.3.2.2 Rand Stratum Maps 11
2.3.2.3 EPA Stratum Maps 11
2.3.2.4 Main Sand Maps 12
2.3.2.5 Light Range Petroleum Hydrocarbon Maps 12
2.3.2.6 Mid-Range Petroleum Hydrocarbon Maps 13
2.3.2.7 Heavy Range Petroleum Hydrocarbon Maps 13
2.3.2.8 Total Petroleum Hydrocarbon Map 13
2.3.3 Interaction of Shallow Stratums with Sewers and Utilities 14
2.4 HYDROCARBON CHEMISTRY AND GEOCHEMISTRY 14
3.0 SUGGESTIONS FOR OPTIMIZING CLEANUP SYSTEM DESIGN AND
IMPLEMENTATION 17
3.1 SOIL VAPOR EXTRACTION SYSTEM DESIGN AND OPTIMIZATION 18
3.1.1 Traditional Vadose Zone Profiling and Monitoring Techniques 18
3.1.2 Soil Sampling 19
3.1.3 Soil Permeability Testing 20
3.1.4 Vapor Monitoring Points 21
3.1.5 SVE Pilot Testing 21
3.2 PNEUMATIC WELL LOGGING TECHNOLOGY 22
3.2.1 Equipment 22
3.2.2 Permeability Profiles 23
3.2.3 Concentration Profiles 25
3.3 OPTIMIZING AN SVE SYSTEM AND IDENTIFYING SVE ASA VIABLE
REMEDIAL ALTERNATIVE 26
3.3.1 Technology Assessment 27
3.3.2 Monitoring the Effectiveness of a Cleanup Strategy 28
3.4 GEOVIS VIDEO MICROPSCOPE ESTIMATES OF IN SITU NAPL SATURATIONS
USING CPT TECHNOLOGY 28
3.4.1 Equipment 29
3.4.2 Effective Porosity 30
3.4.3 NAPL Saturation 31
3.5 ADDITIONAL SITE CHARACTERIZATION ACTIVITIES 32
4.0 INCORPORATING TRIAD-DRIVEN DYNAMIC WORK STRATEGIES 34
4.1 SUMMARY OF SUGGESTIONS 35
4.2 SUMMARY 37
5.0 REFERENCES 38
in
-------
LIST OF FIGURES
Figure
1 LOCATION MAP
2 REGIONAL GEOLOGIC CROSS SECTION
3 GEOLOGIC CROSS SECTION SHOWING FORMATION DISTRIBUTION IN AND
SURROUNDING THE HARTFORD AREA HYDROCARBON PLUME SITE
4 PRELIMINARY GEOLOGICAL CONCEPTUAL SITE MODEL
5 ROST APPROXIMATE EXTENT OF HYDROCARBON MAP
6 AVULSION OF A SEDIMENTARY SEQUENCE
7 CROSS SECTION OF A MIXED LOAD RIVER
8 ISOPACH OF A-CLAY ABOVE STRATUMS INDICATED
9 NORTH OLIVE STRATUM ISOPACH
10 TOP OF RAND AND MAIN STRATUM
11 THICKNESS OF RAND STRATUM IN FEET
12 ISOPACH OF SILTY CLAY BETWEEN RAND STRATUM OR MAIN SAND AND THE
NORTH OLIVE STRATUM OR GROUND SURFACE
13 TOP OF EPA STRATUM
14 EPA STRATUM ISOPACH
15 ISOPACH OF SILTY CLAY BETWEEN EPA STRATUM OR MAIN SAND AND THE
RAND STRATUM OR NORTH OLIVE STRATUM
16 TOP OF MAIN SAND
17 TOP OF TOTAL PETROLEUM HYDROCARBON ROST RESPONSE
18 TOTAL PETROLEUM HYDROCARBON ROST RESPONSE THICKNESS
19 LOCATIONS WHERE POROUS STRATUM (SAND UNITS) ARE POTENTIALLY IN
CONTACT WITH PRODUCT PIPELINES AND MUNICIPAL SEWERS
IV
-------
LIST OF EXHIBITS
Exhibit
1 SCHEMATIC OF PNEUMATIC WELL LOGGING EQUIPMENT
2 EXAMPLE PNEUMATIC WELL LOGGING RESULTS FOR SOIL PERMEABILITY TO AIR
3 SAMPLE OF PNEUMATIC WELL LOGGING RESULTS FOR CONTAMINANT PRODUCT
4 SCHEMATIC OF GeoVIS DIRECT-PUSH PROBE
5 ESTIMATED SOIL POROSITY (VADOSE ZONE) FROM GeoVIS IMAGES
6 VERTICAL PROFILE OF SOIL PHOTOMICROGRAPHS WITH DNAPL DROPLETS
-------
AOC
bgs
CCD
Clayton
CPT
CSM
DNAPL
EPA
Hartford site
HSA
HS
HWG
IEPA
ITRC
LED
LIF
mg/m3
MIP
MS
NAPL
NS
osc
OSRTI
PAH
PID
Praxis
RCRA
ROI
ROST
scfm
SS
START
SVE
TCE
Tetra Tech
ACRONYMS AND ABBREVIATIONS
Administrative Order on Consent
Below ground surface
Charged-coupled device
Clayton Group Services
Cone penetrometer test
Conceptual site model
Dense nonaqueous phase liquid
U.S. Environmental Protection Agency
Hartford Area Hydrocarbon Plume
Hollow stem auger (HSA)
Heavy sheen
Hartford Working Group
Illinois Environmental Protection Agency
Interstate Technology Regulatory Council
Light-emitting diode
Laser induced fluorescence
Milligrams per cubic meter
Membrane interface probe
Moderate sheen
Nonaqueous phase liquid
No sheen
On-scene coordinator
Office of Superfund Remediation and Technology Innovation
Polynuclear aromatic hydrocarbons
Photoionization detector
Praxis Environmental Technologies, Inc.
Resource Conservation and Recovery Act
Radius of influence
Rapid Optical Screening Tool
Standard cubic feet per minute
Slight sheen
Superfund Technical Assessment and Response Team
Soil vapor extraction
Trichloroethylene
Tetra Tech EM Inc.
VI
-------
UVF Ultraviolet fluorescence
VMP Vapor monitoring points
VOC Volatile organic compound
VII
-------
1.0 INTRODUCTION
Tetra Tech EM Inc. (Tetra Tech) prepared the following work products and related suggestions on
integration of the principles of the Triad approach at the Hartford Area Hydrocarbon Plume (the Hartford
site) in Hartford, Illinois. Tetra Tech prepared this document through its support to the U.S.
Environmental Protection Agency (EPA) Office of Superfund Remediation and Technology Innovation
(OSRTI), and in cooperation with the Superfund Technical Assessment and Response Team (START)
and EPA Region 5. Intermittent fires related to vapor intrusion and odor complaints at the Hartford site
have affected residences throughout the Village of Hartford. Subsequent investigations by a group of
potentially responsible parties, known as the Hartford Working Group (HWG), have detected extensive
hydrocarbon contamination beneath the site. The suggestions provided in this report are intended to
provide input to the HWG so characterization and remedial design can be optimized.
1.1 THE TRIAD APPROACH
The START project team submitted a request for OSRTI to evaluate the planned approach for field
activities to be conducted at the Hartford site, and in particular, to review the results obtained using a cone
penetrometer test (CPT) equipped with the Rapid Optical Screening Tool (ROST). OSRTI authorized
Tetra Tech to provide a set of comprehensive suggestions about the project as a whole, keeping in mind
the most urgent needs at the Hartford site and the principles of the Triad approach. This assignment was
based on review of on-going project documents and subsequent discussions with the Region 5 on-scene
coordinators (OSC) and State of Illinois Environmental Protection Agency (IEPA) representatives. The
suggestions provided are intended as a starting point for refining the existing conceptual site model
(CSM) for the Hartford site so that an effective remedy can be designed and implemented as quickly as
possible.
The Triad approach emphasizes the need for an aggressive, up-front systematic planning process to
integrate dynamic work strategies and real-time measurements during site characterization and remedial
design to streamline the cleanup process. The Triad approach also stresses a continuously refined,
interactive process that relies on innovative technologies and strategies to increase the weight of evidence
generated to support decision-making at environmental cleanup sites.
OSRTI is promoting the Triad approach as a means for streamlining site characterizations and
remediation to improve cleanup decisions at Superfund, Resource Conservation and Recovery Act
-------
(RCRA), Brownfields, and other revitalization sites. The Triad approach is becoming more widely
accepted and used by many EPA regions, states, and local governments. The principles and tools used
with the Triad approach have been demonstrated to reduce schedules and budgets required to reach
project milestones at many sites across the country. OSTRI has forged partnerships with the U.S.
Department of Energy, the U.S. Army Corps of Engineers, and the Interstate Technology Regulatory
Council (ITRC). These partnerships have been forged to document use of the Triad approach at small and
large sites to expedite reaching project milestones more quickly and economically while increasing the
level of confidence in project decisions.
1.2 BACKGROUND
The Hartford site is located in the northern portion of the Village of Hartford, Illinois, along the historical
edges of the active Mississippi River channel (Figure 1). Activities are currently being carried out at the
Hartford site to mitigate hazards from vapor intrusion identified within the limits of the Village of
Hartford. From 1966 through 1990, intermittent house fires occurred along East Watkins Street, East
Date Street, and several other streets. More recently, homeowners have registered complaints about
petroleum hydrocarbon odors that triggered the need to temporarily relocate the occupants of several
households. Because of the concern related to petroleum hydrocarbon odor, EPA identified project
objectives that included implementing effective short- and long-term vapor mitigation measures and
delineating free phase and vapor phase hydrocarbons to support final remediation objectives.
A series of documents were reviewed in preparing this report. The documents Tetra Tech reviewed
primarily address the geology and hydrogeology of the site, characterization and delineation of
hydrocarbon impacts at the site, and mitigation of vapor intrusion. The suggestions provided in this
report were prepared based on information obtained from references listed in the bibliography provided as
part of this report. Clayton Group Services (Clayton) also provided valuable support in terms of raw data
and files required to prepare these suggestions.
Tetra Tech's OSRTI support staff became involved at the Hartford site in February 2004. The data
provided in this revised report were updated to include results available as of February 2006. Overall
conclusions presented in this report have also been updated based on more recent results and reports.
In support of the project team's objectives as stated in various work plans, Tetra Tech compiled the
attached figures to support development of a refined CSM for the Hartford site. A Triad systematic
-------
planning process relies heavily on the CSM as the primary tool to focus activities where they can provide
the greatest value to decision-making and be used to identify data gaps, which may need to be filled to
achieve project milestones. The CSM is also used to identify an optimal sequence of activities.
Many practitioners are accustomed to using specific types of CSMs, such as a geological or
hydrogeologic CSM or a pathway-receptor diagram as is often used by the risk assessment community.
Triad practitioners use these forms of a CSM, along with others. A Triad-type CSM also identifies the
decision logic: a systematic process to identify and refine project decisions. Factors such as source
geochemistry, the nature of any possible remedies, and practical considerations are weighed in
establishing the most efficient and logical sequence of activities needed to address project issues and
reach project milestones.
1.3 PROJECT OBJECTIVES
Tetra Tech has identified the following project requirements based on a review of the Administrative
Order on Consent (AOC) and on discussions with the project team:
• "Abate any imminent and substantial threat to the public health or welfare in the area. More
specifically any threat to fish, shellfish and wildlife, public and private property, habitat, and
other living and nonliving natural resources" (Article 3 of the order).
• "Specific attention is to be paid to the investigation of the source and extent of contamination,
implementation of EPA approved interim measures, and design of an active recovery system
designed to abate the on-going threat of discharge to the Mississippi River" (Article 38 of the
order).
• "Conduct a vapor extraction pilot test and provide options for improving and extending the
existing vapor control system" (Article 43 of the order).
• "Implement a sentinel well monitoring program" (Article 47 of the order).
• "Establish the extent of dissolved phase hydrocarbons" (Article 51 of the order).
• "Identify preferential pathways such as utilities and pipelines and establish the extent of
vapor phase and free phase hydrocarbons which could be impacting human health and the
environment at the Hartford site" (Article 52 of the order).
1.4 PRINCIPAL STUDY QUESTIONS
Tetra Tech developed the following principal study questions based on the stated objectives in the AOC
and on review of historical data available for the Hartford site.
-------
1. What are the key, geologic, hydrogeologic, source, and or preferred pathway related factors that
might control:
(a) The release of petroleum fuel related vapors that pose a potential threat to human health and
the environment.
(b) Migration of free product and dissolved phase contamination in and away from potential
source areas?
2. How can these factors be used collaboratively along with design optimization tools to expedite
installation of:
(a) A vapor mitigation system?
(b) A free product extraction system?
(c) A release control and monitoring system for groundwater and surface water?
The following sections of this report examine elements of the preliminary CSM for the Hartford site and
demonstrate how they relate to the principal study questions. The intent is to identify physical
characteristics of the Hartford site that can facilitate planning additional investigations. As the
preliminary CSM is refined, the scale of heterogeneity and variation in environmental conditions can be
understood in sufficient detail as to support implementation of an effective remedy. In addition, a mature
CSM will allow the project team to select appropriate sample locations and sample densities and apply
innovative strategies in the most efficient way possible given the physical constraints of the project.
-------
2.0 PRELIMINARY CONCEPTUAL SITE MODEL
Efforts to mitigate vapors and other adverse environmental conditions at the Hartford site will be guided
by the project team's understanding of several key elements of the preliminary CSM. These elements of
the preliminary CSM include, but may not be limited to, the following:
• Geology and hydrogeology beneath the Hartford site
• Thickness of free product and dissolved phase contamination and proximity to preferred
pathways
• Chemistry and geochemistry of the free product and dissolved-phase contamination
Limited data are available on the chemistry and geochemistry of the contamination beneath the Hartford
site; therefore, interpretive efforts focus on the relationships among the geology, hydrology, and
contaminant distributions across and surrounding the Hartford site. Tetra Tech attempted to link potential
preferred migration pathways with these factors to identify when and where additional investigation
might be warranted. However, details on the configurations of underground utilities or sewer lines were
not available when these suggestions were developed; therefore, this link was not fully developed.
General suggestions are provided on the type and quantity of chemical and geochemical data needed to
support implementation of an effective remedy at the site.
Tetra Tech has developed work products based on the data provided in the references associated with this
report. The work products include a generalized regional cross section (Figure 2) to show the
approximate relationship between the Cahokia Alluvium and the underlying Main Sand. The Cahokia
Alluvium contains silty or clayey sand units of limited extent, such as the North Olive, Rand, and EPA
Strata, as well as fine-grained silty clay layers. The position of the Hartford site on the cross section in
Figure 2 shows the potential for hydrocarbon contamination to affect both surface water and potential
drinking water aquifers adjacent to the site.
Figure 3 is an enlargement of the Hartford site area that shows the general relationships between specific
sand units known to be present. The estimated groundwater flow direction is shown to be toward the
Mississippi River and may vary significantly between individual sand units at the Hartford site.
Based on the limited piezometric surface data that are currently available, the direction of groundwater
flow adjacent to the Mississippi River near the Hartford site can trend from directly toward the river to
-------
directly away from the river. The direction of groundwater flow may fluctuate in response to changes in
the river's elevation and local groundwater pumping. Significant changes in direction of flow between
aquifers overtime is demonstrated by the potentiometric surface maps provided in "Work Plan -
Dissolved Phase Groundwater Investigation, The Hartford Area Hydrocarbon Plume Site" (Clayton
2004d) Figures 2-5 through 2-10.
Figure 4 is a CSM prepared by compiling data from the "FPH CPT/ROST Subsurface Investigation
Report and FPH Monitoring Well and Soil Sampling Plan for the Village of Hartford, Illinois" (Clayton
2004b) and results from the "Site Wide Free Product Investigation" (Clayton 2006b). Figure 4 shows
site-specific geological relationships and the extent of hydrocarbon contamination identified along select
cross sections indicated on the block layout shown in the upper left-hand corner. The blocks extend
beyond the boundaries of the Hartford site to show expected geologic relationships; however, data were
not available for the areas surrounding the Hartford site when this report was generated. Although CSMs
of this type are useful, they also introduce an element of spatial bias in that only select cross-sections can
be presented. This same bias is not as significant in the isopach and top of formation maps discussed later
in this report and used for understanding key geologic, hydrogeologic, and contaminant relationships.
2.1 CURRENTLY IDENTIFIED REGIONAL DATA GAPS
Based on information provided by Clayton, free product does appear to be moving off-site to the
northwest. Data available for the site have been improved over the last 18 months. Figure 5 shows the
locations where ROSTdata have been collected. The extent of the free product plume and the associated
dissolved-phase plume have been adequately delineated in terms of the nature and extent, but additional
characterization may be necessary to finalize system design and optimization. Significant data gaps
remain, particularly in the design of optimized soil vapor extraction and product removal systems.
2.2 GENERAL SUGGESTIONS FOR FILLING REGIONAL DATA GAPS
ROST data have been collected in upgradient source areas from beneath the Premcor refinery (Clayton
2006b), but similar investigations are needed for other surrounding properties to assure that any proposed
remedies are reliable. Historical information on upgradient sources should also be compiled as available.
Data for soil and groundwater in down gradient areas have been used to delineate the extent of the
dissolve phase associated with the product plume (Clayton 2006a). A higher density of data is needed
-------
around source areas where geologic conditions are favorable for vapor intrusion and product removal is
possible to improve the efficiency of the soil vapor extraction (SVE) and product removal systems. The
specific locations and types of data suggested for collection in and immediately surrounding the Hartford
site are also discussed in more detail later in this report.
2.3 GEOLOGY, HYDROGEOLOGY, AND CONTAMINANT DISTRIBUTIONS
According to "Sedimentary Environments: Processes, Fades and Stratigraphy, " (Reading 1996), the
depositional environment beneath the Hartford site can be thought of as a mixed load river avulsion zone.
The Hartford site is located in an area where the Mississippi River has shifted its position in recent
geologic time, in a process referred to as an "avulsion" of the river channel. An avulsion occurs when the
river breaches its natural levee and cuts a new channel in the floodplain. The river bed load is called a
mixed load because widely variable sediment grain size — ranging from finer-grained levee deposits to
coarse sands — can be deposited across a broad avulsion band such as is shown in Figure 6. These
fluvial processes create a highly heterogeneous sediment package.
The typical sedimentary sequence includes thick sequences of sheet-like channel sands, lenticular splay
sands, fine-grained levee, and floodplain deposits. Figure 7 depicts the variety of deposits that are
generally associated with fluvial deposits in a major river avulsion band. Figures 6 and 7 are schematic
diagrams and are not site-specific, but near-surface fine-grained sediments generally grade with depth to
massive sands units. Although the cross section shown in Figure 7 is theoretical, site-specific cross
sections provided in the "FPH CPT/ROST Subsurface Investigation Report and FPHMonitoring Well
and Soil Sampling Plan for the Village of Hartford, Illinois " (Clayton, 2004b) seem to concur with this
generalized geologic sequence. Keeping in mind the two principal study questions, this geologic setting
suggests that better delineation of fine-grained sediments will yield important information on locations
where vapors might be expected to be present at the highest concentrations. Fine-grained sediments can
be substantial barriers to vapor-phase, as well as free-phase and dissolved-phase, hydrocarbons. Finer-
grained sediments can also act as long-term source locations and pose significant challenges to source
mitigation.
Tetra Tech's experience with free-product sites suggests that addressing coarse-grained contaminated
aquifers without also addressing contamination in fine-grained sediments will reduce the effectiveness of
a remedy. For example, applying high vacuums to coarse-grained sands can remove substantial quantities
-------
of free product from the sand units. However, once the system is turned off, residual contamination
bound to fine-grained sediment units can re-contaminate the aquifer.
Based on the schematic diagrams shown in Figure 6 and 7, sands tend to thicken and merge toward the
present-day river and can be in direct hydraulic communication with the river. Therefore, dissolved phase
contamination may be discharged to the river. Thick sections of more fine-grained materials or levee
deposits are also expected around the edges of the former channel sand deposits. Currently, vertical and
lateral piezometric, geologic, and contaminant distribution data are insufficient, both inside and outside of
the Hartford site boundaries, to begin to construct a detailed regional CSM adequate to address the
requirements stipulated in the AOC for the Hartford area.
2.3.1 ROST Results, Contaminant Transport, and Source Areas
The response of the ROST to petroleum hydrocarbon contamination can be roughly correlated with the
presence or absence of product (Tetra Tech 2004). With this in mind, the ROST responses (as shown as
red, yellow, green and blue color bands depending on the range of hydrocarbons) in Figure 4 can be
examined to distinguish primary sources from areas where contaminant migration may have resulted from
transport of free-phase hydrocarbon along the top of the water table. Product source areas are generally
indicated by ROST responses at depths at the surface to 10 feet below ground surface (bgs), depending on
whether the release is suspected to have occurred at the surface or below a buried pipeline. From a
review of Figure 4, it is apparent that source material (above 20 feet bgs) is present primarily along the
eastern, western, and northern edges of the Hartford site. One exception is the area beneath the river
pipeline that runs along Elm Street. New ROST data in this area also indicate the presence office
product at depths starting at approximately 8 feet. Most of the other product contamination indicated in
the ROST responses is present near the water table or the smear zone, which is defined as the region
where the upward and downward fluctuations in the groundwater table spread hydrocarbon contamination
across a greater vertical interval of the soil. It is anticipated that additional surface source areas will be
identified as the density of data for the site increases.
2.3.1.1 ROST Results and Product Recovery Challenges
As discussed in many of the reports reviewed in preparing these suggestions, most of the recorded
incidents of fire and odors occur during high stands in groundwater. A review of Tables 2-1 through 2-3
of the work plan (Clayton 2004a) suggests that free product thickness can increase dramatically as water
-------
levels rise. This relationship is particularly evident at well HMW-22, suggesting that product recovery
may need to focus on wells screened across intervals that correspond to high stands in water levels.
Water is sometimes used to enhance the secondary recovery of petroleum in an oil field, although,
fluctuations in water levels may also act as a hydraulic pump to enhance product recovery.
The relationship of apparent hydrocarbon sources to minimum and maximum groundwater levels can be
used to focus areas where different types of cleanup might be most effective. For example, vapor
extraction technologies could be used with only a minimal need for direct free product recovery in areas
where little or no source material is located at or below a low stand in groundwater if the product present
is in the gasoline range. Conversely, the focus of cleanup efforts might be on collecting free product
during high stands in the water table where source material is present in the smear zone or below the
water table. However, it is important that current efforts focus on monitoring both changes in water
levels and in observed free product thickness on a finer scale than in the past.
At present, the project team has installed nested piezometers or extraction wells in each of the primary
sand units and screened them across the upper portion or across the entire sand unit where the thickness
permits (Clayton 2004b). This strategy may be inefficient, however, based on the observation that much
of the free product underlying the Hartford site is likely present in the smear zone below the upper sand
units. Free product recovery should be directed at those areas where thick columns of product are
observed and should focus on design of a recovery system that target zones for removal based on geologic
conditions and the proximity of product and the water table.
It is apparent from the ROST response observed in potential near-surface source areas, such as are
indicated near ROST locations HROST-6 and HROST-10, that near-surface source areas are limited in
extent. However, the heterogeneity of these areas indicates variation on a finer scale than can be
understood based on existing ROST results. Therefore, additional characterization is needed before near-
surface source areas can be addressed adequately. Outside of near-surface source areas, it may be
possible to define regional trends in geology, hydrology, and product thicknesses and then to design free
product removal and vapor intrusion mitigation systems on a more regional scale.
2.3.2 Sand and Clay Isopach, Formation Top, and Free Product Thickness Maps
Tetra Tech has developed top of stratum and hydrocarbon product thickness maps and top of formation
and product ROST response maps (Figures 8 through 16) for each of the four major strata (North Olive,
-------
Rand, EPA, and Main sands) and the silty clay units that separate them. In addition, Tetra Tech prepared
maps that show the top of free product (heavy, mid-range, and light range) and total product thickness
(Figure 17 and 18). These maps were developed to identify (1) areas where vapor intrusion issues might
be greatest, (2) areas where design of a product extraction system may be warranted, and (3) areas where
monitoring the dissolved-phase plume or where additional characterization is needed. An isopach map of
the total silty clay (Figure 12) has been developed, along with a map to indicate where the sand units may
intersect known potential preferred pathways such as sewer lines and other buried utilities (Figure 19), as
a first attempt at a more sophisticated level of interpretation.
Figure 19 is an example of the type of work product that could be important, as the CSM is refined. The
geologic, hydrogeologic, and contaminant characteristics provided in Figures 8 through 18 can be
combined on composite maps to drive a dynamic work strategy and guide future investigations.
However, any additional integration of the materials presented or discussed in this report is beyond the
scope of the support available through OSRTI to the Hartford Working Group and EPA Region 5.
The potential for vapor intrusion is likely highest where the uppermost extents of permeable sand units
are closest to the surface, the total thickness of fine-grained alluvial deposits is lowest, and the total
thickness of sand and product is the greatest. The maps discussed in the following sections attempt to
identify specific geologic, hydrogeologic, and free product relationships that could directly influence the
fate and transport of free product, distributions of vapor-phase contaminants, and distributions of
dissolved-phase hydrocarbons.
2.3.2.1 North Olive Sand Maps
It appears that the North Olive Stratum thins across the central portion of the Hartford site and terminates
here (i.e pinches out). Figure 8 shows the isopach thickness of the North Olive Stratum. This pinching
out suggests that the North Olive Stratum is not in direct hydraulic communication with either the Rand
or Main Sands except in the southeastern portion of the site, where the North Olive merges with the Main
Sand. Free product or vapors within the North Olive unit cease beyond a point, as suggested by the
general relationship between reported fires and the extent of the North Olive stratum.
Product was also detected in ROST locations HROST 51 and HROST 52 (Figure 8), where the North
Olive merges with the Main Sand. Therefore, the potential for direct communication of contaminants
10
-------
from the North Olive Stratum into the Main Sand, or vice versa, exists in this area. Multiple fire events
have occurred in this area.
As mentioned previously and as shown in the isopach of the North Olive Stratum in Figure 9, areas where
fires have been historically reported across the Hartford site seem to correspond with areas where a
measurable thickness of the North Olive Stratum has been mapped. It has also been observed that the
silty clay layers thin out above the Main and Rand strata in this area.
2.3.2.2 Rand Stratum Maps
The Rand Stratum merges into the Main Sand adjacent to the eastern boundary of the Village of Hartford,
as shown in Figure 10. Figure 11 presents the thickness of the Rand Stratum in feet. Contamination
within the Rand Stratum in the southeastern portion of the site could therefore easily migrate in the vapor
or free phase from the Rand Stratum into the area enclosed by the 12-foot bgs contour in the Main Sand.
A potential also exists for contamination within the Rand Stratum near ROST locations HROST 26 and
33 (Figure 5) to migrate up into the structural high in the Main Sand shown in the north-central portion of
the village.
Evaluation of the isopach of the silty clay above the Rand and Main sands further supports why fires have
not been recorded in and around ROST locations HROST 23 and 24 (Figure 4). The silty clay in this area
thickens to nearly 20 feet (Figure 12). The thickness of the silty clay above the Rand or Main Sand is
generally less than 12 feet and the North Olive Stratum is also present throughout the area where fires
have been recorded.
The thickening of the clay in this area may indicate that the need to mitigate vapors may be less urgent.
However, the hydrocarbons appear thickest in this area. Product recovery in this location may be
warranted because of the potential for product to move from this area toward the northwest, where the
silty clay unit thins dramatically and fires have been reported.
2.3.2.3 EPA Stratum Maps
The EPA Stratum is limited in extent, as shown in Figures 13 and 14. However, contamination in the
EPA Stratum would migrate directly up into the Main Sand in the northwest portion of the Hartford site.
11
-------
2.3.2.4 Main Sand Maps
An isopach of the Main Sand could not be created because it represents the basal glacial outwash sand
unit, which extends down to the limestone bedrock in the area. Stratigraphic information on the bottom
of the unit is not available. An isopach of silty clay between the EPA stratum of Main Sand and the Rand
stratum or the North Olive stratum is presented in Figure 15. The contour map of the top of the Main
Sand (Figure 16) shows a northwest-trending structural high along the approximate axis of fires reported
in the southern portion of the affected region beneath the village. This structural high is crudely aligned
along the expected flow direction in the Main Sand, as depicted in the work plan (Clayton 2004d). This
northwest-trending feature in the Main Sand suggests that a principal area of concern for contaminant
migration away from the site could exist northwest of the current site boundaries. Migration of dissolved-
and free-phase constituents might be expected downgradient of this structural high along the regional
northwest direction of flow within the Main Sand. As will be discussed later in this section, product
appears to extend off site and downgradient along this northwesterly trend (Clayton 2006a). The impacts
from the presence of product west and north of the village are not thought to immediately affect the
current remedial design efforts and are therefore not discussed further in this report.
2.3.2.5 Light Range Petroleum Hydrocarbon Maps
A large high in the top of the ROST response for lighter hydrocarbons is present along Elm Street
(pipelines run from the refinery to the river along Elm Street). Figure 17 shows the approximate extent of
lighter range, lower boiling point fuels in combination with the top of the ROST response. The depth of
the response appears to coincide with the approximate depth of the pipelines in this area. Numerous spills
have been recorded in this area and historical records for the pipelines indicate that these lines could have
leaked throughout the history if their use.
Heavier hydrocarbons have a greater peak height at higher wavelengths of absorbance, as indicated by
greater peak heights on the right-hand side of the ROST output file (the dwell profile). When peak
heights are greater on the left-hand side of these dwell profiles, the fuel at this spot in the contaminant
plume is likely made up of hydrocarbons such as gasoline, which have lower boiling points, usually
considered light range hydrocarbons. Diesel fuels might be considered a mid-range hydrocarbon product
with greater peak heights in the center of the profile. Motor oil or weathered product, which has been in
the ground for extended periods, is usually considered a heavy range hydrocarbon product. Based on the
12
-------
general results shown in Figure 17, it appears that the preponderance of hydrocarbons present across the
site are in the gasoline range.
2.3.2.6 Mid-Range Petroleum Hydrocarbon Maps
The distribution of mid range hydrocarbons shown in Figure 17 seems to indicate the potential for the
presence of two source areas for this type of petroleum product. One is located along the northeastern
boundary of the site near ROST location HROST 6 and 10 and another is located near the northwestern
boundary of the site near HROST 2. The nature and extent of mid-range hydrocarbons may influence
their treatability and their tendency to cause vapor intrusion and will therefore need to be examined more
closely. The specific chemistry and constituent makeup of each of the source types identified should be
examined to determine: (1) site-specific action levels for vapor intrusion, and (2) site-specific action
levels that can be used to assess the need for removal. These action levels will be driven by the chemistry
and type of potential associated risk or hazard identified for the area of interest within the site.
2.3.2.7 Heavy Range Petroleum Hydrocarbon Maps
Figure 17 shows the limited extent of heavier range hydrocarbons at the Hartford site. As expected, the
extent does not generally correspond to areas where fire hazards have been reported. Since heavy
hydrocarbons products tend to sorb to the soil and are generally more viscous, they have less of a
tendency to migrate away from primary source areas. Primary concerns in these areas should be focused
on limiting the potential for direct contact. Chemicals of potential concern in surface soil in this area
might include polynuclear aromatic hydrocarbons (PAH). A close inspection of the ROST profiles in this
area does indicate the presence of light hydrocarbons beneath these apparent heavier hydrocarbon source
areas.
2.3.2.8 Total Petroleum Hydrocarbon Map
Figure 18 shows the extent of the total ROST response to all three ranges of hydrocarbons. The largest
thickness in ROST response is along Elm Street. This supports the large high in the ROST response of
lighter range hydrocarbons in that same location.
13
-------
2.3.3 Interaction of Shallow Stratums with Sewers and Utilities
Tetra Tech prepared Figure 19 to identify areas where preferred pathways (sewers) might intersect
permeable stratum units, allowing hydrocarbons to accumulate at shallow depths. Figure 19 shows the
location of product pipelines, municipal sewer mains, and shallow stratum units (with upper extents
above 12 feet bgs). The map identifies the upper extents of shallow sand intervals (primarily the North
Olive Stratum, but also the Main Sand in the southern portion of the Hartford site). The 12-foot bgs
contour shown in this figure is significant because the depth of buried pipelines is approximately 12 feet
bgs, as noted in the "Utility and Pipeline Investigation Work Plan, the Hartford Area Hydrocarbon Plume
Site " (Clayton 2004f). This work plan did not indicate the depth of the municipal sewer mains, but it can
be assumed that they are above the 10-foot bgs contour. Notably, a sewer main crosses the 8- to 10-foot
bgs contour interval in the eastern portion of the village. Five buildings where fires have been reported
are located within 100 feet of this sewer main. This map, like those previously discussed, should be
considered when the working group prioritizes locations where sewer and utility investigations and
design-related activities are planned. The FiWG also may consider using the presence or absence of light
or mid-range free product in the shallow sand units as a means of prioritizing when and where to focus
remediation efforts for sewers and utilities.
2.4 HYDROCARBON CHEMISTRY AND GEOCHEMISTRY
The chemistry and geochemistry of hydrocarbon product, geologic formations, and groundwater beneath
the Hartford site will have a strong influence on the effectiveness of any remedy. These and other
physical factors such as moisture content, permeability, and effective porosity should be used in
conjunction with one another to support the design of any potential remedy. The HWG has not focused
on the chemistry of the product found beneath the Hartford site up to this point, as is indicated by
responses to comments provided by Clayton to EPA Region 5 dated June 21, 2004, and titled "Letter to
USEPA Region 5. Response to Comments to ROST Investigation Report and Work Plan. ") The response
to U.S. EPA comment 1 part A, second sentence states, "It is Clayton's opinion, based on experience at
other petroleum sites, that the design of the remediation system will be primarily based upon geology of
the area and the amount of product present not the type of product" (Clayton 2004e).
Ignoring product-specific chemistry during remedial design could limit the effectiveness of any cleanup
strategy. The petroleum industry has long recognized that the nature of various petroleum products can
pose different challenges to extraction of petroleum from an oil reservoir. Heavier products often require
14
-------
more aggressive techniques to extract. For example, methods such as steam-enhanced recovery have
been developed to address removal of heavier hydrocarbons where simple flooding methods have proven
ineffective.
Not only is further characterization of the nature of the product necessary; the physical properties of the
petroleum hydrocarbons need to be understood so their fate and transport can be estimated and input to a
model to support the evaluation of impacts to surface water in the area. Further characterization of the
product is also suggested to support risk estimation and development of field-based action levels related
to both vapor intrusion and dissolved-phase fate and transport issues.
For example, one of the questions at the Hartford site is the impact of removing the free product and
dissolved-phase hydrocarbons will have on vapor intrusion. This issue is chemistry related. Petroleum
hydrocarbons consist of complex mixtures of carbon, hydrogen, ammonia, sulfur compounds, and other
constituents, such as lead and oxygenates, used to improve fuel performance. Each mixture has a
susceptibility to treatment at a particular moisture level in soil that is related to its Henry's Law constant.
Therefore, the composition and physical properties of the mixture can affect the removal rate and
estimated risk. Liquid-phase removal may also be affected by the chemical and physical properties of the
free product, such as its tendency to form a physical or chemical emulsion that will be difficult to treat.
Detailed data on chemistry, geochemistry, and physical properties are needed to design a system and then
predict whether it can be successful in mitigating vapor or dissolved phase-related hazards.
The HWG should consider implementing a robust chemical, geochemical, and physical properties
characterization effort to begin to understand differences in product chemistry. The analytical suite
should include, but not necessarily be limited to, the following:
• Volatile organic analyses (using method 8260)
• Semivolatile organic analysis (using method 8270)
• PAH analyses (using modified method 8270 operated in the selective ion monitoring mode)
• Viscosity and density analyses
• Porosity, permeability, grain size, total organic carbon
• Nonaqueous phase liquid (NAPL) saturation
• Cation exchange capacity
15
-------
In addition to these chemical and physical property analyses, site-specific testing in the form of core
column tests might be considered. These tests can also be performed in situ using innovative tools such
as the Praxis Environmental Technologies, Inc. (Praxis ) PneuLog, which allows for the design of the
SVE removal system to be optimized once a system has been installed. Since there is an existing system
on site, this technology might be immediately applicable.
Core column studies may be conducted when there are significant questions on the applicability of one of
several alternatives for treatment, such as in the area near ROST location HROST 2. Pilot testing with
PneuLog could be used to optimize and expand an existing system design in areas where SVE already
appears to be the logical alternative, such as the area surrounding HROST 51. Chemical data, along with
concentrations present, should be used to estimate any risk that requires treatment. Additional
information on the use of core column and well product removal pilot testing can be made available from
Tetra Tech on request.
16
-------
3.0 SUGGESTIONS FOR OPTIMIZING CLEANUP SYSTEM
DESIGN AND IMPLEMENTATION
In the review of the primary study questions and the information presented thus far in this report, the
working group faces the following issues at the Hartford site that will eventually need to be addressed:
• Immediate physical hazards, such as fires that result from vapor intrusion
• Impacts to human health from vapors
• Impacts from contaminated soil in the vadose zone to groundwater
• Impacts to groundwater from free product in the smear and saturated zone
• Impacts from dissolved- and free-phase hydrocarbons to surface water
A robust set of suggestions for each of these issues is beyond the scope of the support that Tetra Tech can
provide under its current assignment for OSRTI. Therefore, the focus in this section is on providing
general observational data and suggestions for most of the key elements that should be evaluated. HWG
can then more fully evaluate the types of specific activities and decisions that will need to be made.
Installing and sampling vapor monitoring probes (VMP) is under way at the Hartford site to evaluate the
potential hazards and risks to human health from vapors. The maps and suggestions provided by Tetra
Tech in this document are intended to identify areas where the interaction among the sewer and utility
system, geologic features, and free product should be further evaluated through VMPs. In addition, the
maps and suggestions provided indicate where free product may be collecting in stratigraphic traps, such
as the area near ROST location HROST 51. Tetra Tech believes this area might be more amenable to
SVE than other areas where the presence of more fine-grained materials might pose a challenge to the use
of SVE. Suggestions are also provided that identify areas where free product extraction should be the
focus of the HWG efforts. Free product extraction may be warranted where free product is found at the
greatest thicknesses (Figure 18) and has the greatest potential to continue to contribute to migration of
dissolved-phase contamination away from source areas.
The current data set lacks sufficient information on hydrology and the chemistry and spatial distribution
of contaminants to support the design of a free product extraction or dissolved-phase monitoring and
treatment system. Therefore, this section discusses use of collaborative data sets and similar approaches
to optimize the design of the vapor extraction system, investigate the dissolved-phase contamination,
evaluate methods to remove free product from the smear and saturated zones, and implement an
integrated monitoring system to track the progress of the remedial action.
17
-------
3.1 SOIL VAPOR EXTRACTION SYSTEM DESIGN AND OPTIMIZATION
As mentioned previously, SVE may be an effective alternative in source areas at the Hartford site,
particularly where free product is present above the water table and geologic conditions are favorable.
PneuLog is a technology designed to reduce long-term operational costs and accelerate cleanup by
optimizing SVE systems in unsaturated zones. The PneuLog technology uses in-well instrumentation to
measure air permeability and contamination production continuously throughout an extraction well within
the screened interval during SVE. This technology is intended to improve the assessment of geologic
heterogeneity within the screened intervals of individual wells and identify mass transfer constraints in
the vadose zone. Data from several wells can be used to optimize a cleanup strategy and estimate
operation times needed to meet closure requirements.
Tetra Tech proposes using PneuLog technology to support the evaluation and optimization of any SVE
systems planned for the Hartford site. Tetra Tech suggests that a dynamic work strategy may be used as
an alternative to the traditional phased approach to limit the need for mobilizations and thus streamline
product removal and vapor mitigation. The PneuLog technology can be used not only to target zones
with the highest concentration in vapors; it can also be used to size pumps where contamination in
concentrated in fine-grained soils. In contrast, conventional SVE design and optimization procedures rely
on empirical data that do not adequately evaluate mass transfer constraints, limiting the effectiveness of
the remedy. As a result, conventional systems may be overbuilt, inefficient, and expensive to operate.
The PneuLog approach incorporates short-term SVE testing with pneumatic well logging to delineate the
horizontal and vertical extent of contaminants and quantify the permeability of soils throughout the
screened interval. The PneuLog test is repeatable, and multiple deployments can track the progress of
cleanup when combined with technologies such as passive or active soil gas surveys and vapor probe
analysis in a collaborative data set. When used in a number of wells, this approach provides a more
complete and accurate baseline evaluation for design and optimization of SVE systems. In addition, data
on soil permeability and airflow rate data provided by PneuLog can be used in models to estimate
removal action timeframes.
3.1.1 Traditional Vadose Zone Profiling and Monitoring Techniques
Traditional methods of delineating vadose zone contamination to implement SVE involve soil gas surveys
and multipoint vapor probe sampling. Traditional methods of developing vertical contaminant profiles
18
-------
involve installing discrete VMPs at multiple depths at a single location. Similarly, traditional methods of
developing lithology profiles require continuous split spoon sampling. SVE pilot testing is used to
develop soil vapor production rates, contaminant concentrations, and radius of influence (ROI)
information on a site-wide scale.
Before an SVE system can be constructed, the following types of data should be collected at specific
targeted locations within the Hartford site at a density sufficient to characterize the heterogeneity of the
hydrogeologic setting:
• Contaminant type and associated volatility
• Permeability of the soil
• Soil structure and stratification
• Soil moisture content
• Depth to groundwater and changes in product thickness and water level over time
The following sections briefly summarize methods Tetra Tech suggests for horizontal and vertical
profiling of the vadose zone.
3.1.2 Soil Sampling
Vertical profiling is an important component of site characterization and CSM development. Before
cleanup technologies for soil treatment can be evaluated, a site is usually investigated to characterize the
geology and vertical distribution of contaminants on a site-specific scale. Lithologic and geotechnical
parameters are usually collected at various depth intervals. New tools such as the membrane interface
probe (MIP) and other real-time sensors, or near-continuously reading instruments such as PneuLog, are
changing existing ideas about how contaminants are distributed in the environment. Small-scale
heterogeneity seems to be the rule, rather than the exception, and can severely impair the effectiveness of
a remedy.
Soil structure and stratification are important to the effectiveness of SVE because they affect vapor flow
within the soil matrix under SVE conditions. For instance, it is widely accepted that SVE is generally
less effective in moist, silty, or clayey soils. Structural characteristics (such as layering and fractures) can
create preferential flow pathways that can short-circuit SVE systems, resulting in extended remedial
19
-------
timeframes if the extraction points are positioned such that the induced airflow bypasses the area of
contamination.
Soil borings are typically completed using a hollow stem auger (HSA) or direct-push probe such as a
Geoprobe. These borings are drilled to collect a continuous soil sample and characterize the subsurface.
Instruments can be deployed with direct-push drilling equipment that can record nearly continuous
measurements of the soil's physical and chemical characteristics, resulting in development of a
continuous vertical profile. The MIP technology provides a continuous log of soil conductivity and the
volatile organic compound (VOC) concentration as it is driven into the soil. The Simulprober technology
is a modified split spoon sampler that can also be used with conventional direct-push drilling techniques
and is intended to collect continuous soil and soil gas samples and conduct in situ single-point slug tests.
These tools have significant limitations; an investigation must proceed with these limitations in mind and
should not be conducted without consulting with an expert who has significant experience with these
instruments.
3.1.3 Soil Permeability Testing
Permeability affects the rate of air and vapor movement through the soil: the higher the permeability of
the soil, the faster the movement and (ideally) the greater the quantity of vapors that can be extracted.
High moisture content in soils can reduce permeability and, consequently, the effectiveness of SVE by
restricting the flow of air through soil pores. Fine-grained soils produce a thicker capillary fringe than
coarse-grained soils. SVE is generally not effective in treating soils below the top of the capillary fringe.
Pumps can be used to depress the water table; however, pumping to lower the water table is not feasible
because of the volume of water in the aquifers beneath the Hartford site. Site-specific data on water
levels and soil permeability will be integral in optimizing the effectiveness of SVE or when SVE is
selected as the preferred remedial alternative. Combining this information with characterization data on a
finer scale can help engineers understand the limitations of a proposed system. For example, a site where
the CSM indicates contamination in the finer-grained portions of the soil profile may not be effectively
remediated using SVE alone. This level of understanding on the potential for SVE as a remedy at specific
locations within the Hartford site (such as near HROST-2) can be developed only by collecting data on
soil permeability and contaminant concentration on a finer scale using pneumatic tests.
20
-------
3.1.4 Vapor Monitoring Points
VMPs are the traditional method for developing initial vertical contaminant profiles at SVE sites and may
also be used as a tool for optimizing existing systems. VMPs are constructed by installing several
relatively short-screened interval wells separated by bentonite seals within a single borehole. VMPs are
used to measure vertical variations of vapor-phase contaminant concentrations and pressure and vacuum
response (and, indirectly, permeability) along the soil profile when they are placed at varying depth
intervals. Soil vapor chemistry is assessed at each discrete depth interval by withdrawing and analyzing
soil vapor samples. The vacuum pressure required to extract the soil vapor from each individually
screened depth interval indirectly indicates permeability. Soil permeability dictates the amount of
airflow. Relatively high airflow indicates higher permeability, and relatively low airflow indicates low
permeability.
Continuous soil borings are typically used to install VMPs. This method can be labor intensive and does
not yield a continuous soil profile. Optimal locations for these points are best identified using high-
quality geologic, contaminant distribution, and pneumatic test data. Conducting MIP profiles or a
headspace soil gas survey before the VMPs are installed can improve efficiency when vapor probes are
designed and installed. Installing VMPs without these data can result in poor system design and
ineffectiveness of the remedy.
3.1.5 SVE Pilot Testing
A pilot test is recommended for evaluating SVE effectiveness and design parameters at any site,
particularly where SVE is expected to be only marginally to moderately effective. Data provided by pilot
testing are necessary to properly design the full-scale SVE system. Pilot tests also provide information on
the concentration of VOCs that are likely to be extracted during the early stages of operation of the SVE
system.
Various extraction rates and wellhead vacuums must be evaluated to estimate optimal operating
conditions. Pilot studies typically involve extraction of soil vapors for a short period (1 to 30 days) from
a single extraction well, which may be an existing monitoring well at the Hartford site. However, longer
pilot studies (up to 6 months) using more than one extraction well may be appropriate for larger sites.
More information on methods for system operation can be found in "Innovative Site Remediation
Technology, Design and Application, Volume 7, Vacuum Extraction and Air Sparging" (EPA 1998).
21
-------
Vapor concentrations should be measured at two or more intervals during the pilot study to estimate
initial vapor concentrations that may be expected during operation of a full-scale system. The vapor
concentration, vapor extraction rate, and vacuum data should also be used in the design process to select
extraction and treatment equipment.
Estimating the ROI of each extraction point is important for proper design of an SVE system. The ROI is
defined as the greatest distance from an extraction well where a sufficient vacuum and vapor flow can be
induced to adequately enhance volatilization and extraction of contaminants in the soil. Practitioners can
increase their confidence that the pilot test design accommodates site conditions with better
characterization methods and pneumatic logging techniques.
3.2 PNEUMATIC WELL LOGGING TECHNOLOGY
Pneumatic well logging is a technology developed by Praxis, which is designed to optimize SVE design
and operation. Pneumatic well logging is performed by simultaneously measuring cumulative airflow and
contaminant vapor concentrations vertically along the depth of an extraction well screen during active
SVE. To record these measurements, a flow sensor is moved up through the well while vapor extraction
and soil gas samples are continuously collected and analyzed. Collecting these measurements at a
representative number of wells can yield a three-dimensional picture of the extent of soil contamination at
a site as well as the distribution of soil permeability. These measurements, in conjunction with traditional
sampling methods, can yield a more thorough understanding of a site and how an SVE system can be
optimized. This more thorough understanding is possible because PneuLog technology provides
information that other technologies cannot, such as soil permeability and mass loading of the vadose
zone.
3.2.1 Equipment
The equipment used for PneuLog pneumatic well logging is illustrated in Exhibit 1. The PneuLog
instrument is attached to a cable, which passes through alignment pulleys and a vacuum-tight fitting at the
wellhead. The instrumentation is raised or lowered by a cable wound around a motorized reel. The
logging proceeds at a rate of 8 feet per minute along the screen in the SVE extraction well. Sensors in the
pulley assembly indicate the depth of the measurement. Electrical leads connect the flow sensor to a data
acquisition system located on the motorized reel. A vapor sampling tube connects the sample port on the
instrument to a vacuum pump, also located on the reel. The sampling pump draws a continuous stream of
22
-------
air through the sampling tube to the surface, where it is analyzed for VOCs and other compounds of
interest (such as oxygen and carbon dioxide). A photoionization detector (PID) is used to provide a
continuous reading of total VOC concentrations. Summa canister samples can be collected for off-site
analysis by gas chromatography and mass spectrometry to estimate compound-specific concentrations at
discrete depths and to calibrate the PID readings. Supplemental vapor samples can be collected in Tedlar
bags and analyzed on-site with a field gas chromatograph.
Exhibit 1: Schematic of Pneumatic Well Logging Equipment
Computer
RealTime
View of
Data
Cable
Alignment
Assembly
Vacuum on
Wellhead
Applied by
Blower
Vacuum Pump for r
Vapor Samples
Tube for Vapor Sample
Transfer
Air Velocity Sensor
3.2.2 Permeability Profiles
The airflow from each soil layer is related to the cumulative airflow by a simple mass balance. The
cumulative airflow measured below the soil layer is subtracted from the cumulative airflow measured
above the soil layer to calculate the airflow from a soil layer. The soil permeability of the interval is then
determined from Darcy's law. The data and the analyses appear similar to output from borehole
flowmeter testing in water wells. A typical cumulative gas flow measurement from PneuLog is provided
in Part (a) of Exhibit 2, below. In this example, the well is screened from 12 to 32 feet bgs. As shown,
23
-------
the airflow from the bottom half of the well is essentially zero. The airflow increases steadily from 0 to
28 standard cubic feet per minute (scfm) between 23 and 16.5 feet bgs as the instrument is raised through
the screen. The steady flow increase indicates this soil interval has a relatively uniform permeability to
air. Only 2.5 scfm of soil gas are added from 16.5 to 15 feet. The volumetric flow increases by 15 scfm
in the next 1.5-foot interval up to 13.5 feet. The top 1.5 feet of the screen adds only 1 scfm to the total.
Exhibit 2: Example Pneumatic Well Logging Results for Soil Permeability to Air
57
Q
30
35
0 20 40 60
Cumulative Gas Flow (scfm)
0 5 10 15
Soil Gas Production (scfm/ft)
Part (a)
Part (b)
The diagrams present an interpretation of the cumulative flow measurements as soil gas production
proceeded. An effective air permeability profile can be generated using the soil gas production profile
with multi-dimensional analytical or numerical airflow models. The permeability of an interval is
proportional to the change in flow across the interval, its thickness, its depth below the surface, and the
well vacuum according to Darcy's law. Part (b) of Exhibit 2 reveals five soil strata along the screen. The
permeability of the stratum intersected by the bottom half of the screen (yellow or light blocks) is
relatively low since no measurable soil gas was produced. The geologist characterized the soil of this
interval as silt. The air production rates for the soil intervals from 16.5 to 23 feet and 13.5 to 15 feet
indicate coarse sand. These two sand intervals are separated by a 1.5-foot-thick silt interval. The soil at
24
-------
the top of the screen was also characterized as silt. This characterization of the physical properties is
superior to a geological log and a typical air permeability test. The PneuLog results are usually consistent
with the geologic observations; however, geologic logs provide little or no indication of potential air
permeability. Without the pneumatic logging data, the permeability measured by typical testing is
averaged over the screen interval and of the subsurface flow profile. It therefore cannot be quantified and
well screens subsequently optimized.
3.2.3 Concentration Profiles
The measurement of VOC concentrations along the well screen indicates the distribution of VOCs in the
screened interval. An example concentration log, which was collected simultaneously with the airflow
log previously discussed, is presented in Part (a) of Exhibit 3, below. This concentration profile was
obtained from a continuously reading PID that was calibrated to trichloroethylene (TCE) concentrations
with on-site and off-site gas chromatographic analysis of vapor samples from discrete depths and the
wellhead. The vapor concentration measured is lowest near the bottom of the screen and increases
slightly up to a depth of about 28 feet. As the tool is raised higher in the well, the concentration increases
sharply to a maximum at 26 feet and remains relatively high to a depth of 21 feet. The concentration then
decreases steadily from 21 to 15 feet bgs. The concentration increases very slightly between 15 feet and
the top of the screen.
The increases and decreases in concentration observed can be combined with the depth-specific air
production in a mass balance to estimate depth-specific soil gas concentrations. The PneuLog device
simultaneously measures the flow rate and concentration versus depth. The change in the product of
these two variables over a specified depth interval divided by the flow change is equal to the contaminant
vapor concentration in the soils of the depth interval. Application of this relationship to the data shown in
Exhibit 3 Part (a) yields the contaminant vapor concentration profile presented in Exhibit 3, Part (b). The
highest concentration occurs in the low-permeability material that underlies the deeper sand interval.
This high concentration indicates that the low-permeability interval creates a mass transfer constraint to
SVE.
25
-------
Exhibit 3: Sample of Pneumatic Well Logging Results for Contaminant Product
5
10
15
•=3
a
o>
Q
20
25
30
35
100 150 200 0 100 200
3\
Cum. TCE Cone, (mg/m )
Part (a)
TCE Soil Gas Cone, (mg/m )
Part (b)
Note: mg/m3 = Milligrams per cubic meter
3.3 OPTIMIZING AN SVE SYSTEM AND IDENTIFYING SVE AS A VIABLE REMEDIAL
ALTERNATIVE
As illustrated by this example, pneumatic well logging provides a more thorough and appropriate site
characterization than will traditional methods alone. Repeating the process in a representative number of
wells can generate a three-dimensional description of the physical and chemical subsurface by correlating
between locations. The technique also provides higher-quality data that can be used to more effectively
design and optimize an SVE system. Soil strata near or below cleanup goals are quickly identified, and
the extraction flow rate can be reduced or terminated from these layers. The operation can then be
focused on strata where contaminants remain at concentrations above cleanup goals. This optimization
could lead to cost savings by accelerating cleanup and lowering operation and maintenance costs.
PneuLog can be used in conjunction with other new and improved methods of site characterization to
build comprehensive data sets that can be used to evaluate when and if SVE is a viable alternative and
even to decide when SVE is no longer needed. Real-time measurement technologies such as a MIP or
laser induced fluorescence (LIF) provide contaminant distribution data that are independent of the
26
-------
permeability of the contaminated stratum. PneuLog, on the other hand, is biased by the permeability of
the soil. Combining these two different types of data in a collaborative data set can provide practitioners
with a better idea of whether SVE will be effective and the design specifications that are most
appropriate.
3.3.1 Technology Assessment
Praxis developed the PneuLog technology to aid both site characterization and optimization of SVE
systems. Tetra Tech's evaluation of this technology revealed several advantages and disadvantages. The
primary advantages of this technology are as follows:
• A continuous vertical profile of contaminant concentration and soil permeability can be
quickly developed for each SVE well on site. This profile represents average values for each
major soil interval intersected by the vent well.
• The use of progressive extraction, vapor sampling, and pneumatic logging of the wells as
they are installed will provide guidance for locating additional wells to more adequately
characterize the Hartford site.
• The actual VOC concentrations in milligrams per cubic meter (mg/m3) produced at specific
intervals are measured.
• The permeability, flow rate, and total VOC production for a section of screened interval can
be estimated. This information is useful in optimizing or modifying vent wells and for sizing
blowers and vapor treatment equipment for new or modified SVE systems.
• The data are presented in a manner that is easy to interpret and highlights significant
variations between intervals. When combined with other, more detailed, methods that can
measure contaminant distribution regardless of permeability, the data can be used to decide
when and if SVE will be effective or whether to modify the system.
The primary disadvantages are as follows:
• An SVE extraction well must be installed if one does not already exist. If SVE does not turn
out to be appropriate for the Hartford site, then this site characterization method may be more
expensive because installation of a well will generate soil cuttings and is labor intensive.
• Pneumatic logging provides limited data from soil intervals that are not within the screened
interval of the well. The ideal screened depths cannot be identified before the vent well is
installed. However, PneuLog testing in a single-well pilot test could be used to more
effectively locate well screens in a full-scale, multi-well SVE system.
• Contamination from an overlying low-permeability layer may be detected at dilute
concentrations in an underlying high-permeability layer. High levels of VOC contamination
may be entering the vent well from one direction and be diluted by clean soil gas from other
directions. The vent well tends to average VOC concentrations, and the PneuLog tool can
measure only the average VOC concentration inside the well and the average permeability of
the soil interval.
27
-------
Overall, a technology such as PneuLog is warranted because of the size of the Hartford site and the need
for efficiency in implementing a cleanup strategy. Additional information and design considerations
should be evaluated in conjunction with Praxis or an equivalent vendor of a similar technology.
3.3.2 Monitoring the Effectiveness of a Cleanup Strategy
A direct measurement approach should be used to monitor the impact of any cleanup strategy for the
Hartford site. Passive soil gas probes placed at regular intervals in and around treatment zones to measure
relative changes in concentration can be an economical way to accomplish this task. Alternately, VMP
samples can be collected over time using a focused analytical program. Any analytical program for
monitoring the effectiveness of the remedy needs to include not only contaminant-specific analysis, but
should also monitor for explosive levels of less toxic petroleum fuel-related constituents. Action levels in
the vapor phase will need to be agreed upon by all stakeholders before a cleanup strategy can be
implemented.
Real-time measurement tools such as a mobile gas chromatograph should be considered initially as a
method to economically increase the density of vapor sampling. More sensitive vapor probe
measurements such as Summa canisters may be required as concentrations decrease and vapors are
mitigated. The PneuLog itself can also be used to evaluate yields and the benefit of continued operation
of an SVE system.
3.4 GEOVIS VIDEO MICROPSCOPE ESTIMATES OF IN SITU NAPL SATURATIONS
USING CPT TECHNOLOGY
In addition to traditional methods for evaluating the potential for product removal at the site, Tetra Tech
suggests that HWG consider GeoVIS as a method to increase the project team's understanding of
hydrocarbon saturations. Hydrocarbon saturations can be used to estimate where additional permeability
and productivity testing using high-vacuum extraction may be warranted. Currently, the HWG proposes
to base its product removal system design on one or two key locations where core data were collected and
conditions were favorable for further testing (Clayton 2006c). Use of a limited set of results in the
manner proposed could result in an ineffective remedial design and wasted project funds.
The GeoVIS video microscope has the capability for collecting real-time in situ images of the subsurface
soil environment for use in estimating soil porosity and light NAPL saturations. The GeoVIS system uses
28
-------
a miniature digital video camera coupled with magnification and focusing lens systems integrated into a
cone penetrometer probe. The soil environment is imaged through a window in the probe and the video
signal from the camera is returned to the surface where it can be viewed in real-time on a video monitor,
recorded on a video recorder, or digitized (or any combination). When combined with lithology
information obtained from CPT probe data and soil contamination estimations from LIF data, GeoVIS
provides the small-scale tools necessary to identify thin layers of highly permeable material that provide a
potential pathway for contaminant transport and removal, which could be overlooked easily through
conventional means. It also provides a direct means for locating contamination source zones that have
been difficult to localize using traditional sampling approaches.
3.4.1 Equipment
The equipment used for GeoVIS is a direct push penetrometer mounted on a CPT platform. It is equipped
with a vertically mounted charged-coupled device (CCD), a mirror to reflect a side view of the soil into
the camera, and a sapphire viewing window (Exhibit 4). The GeoVIS uses light from four light-emitting
diode (LED) light sources (Xenon lamps) to distribute diffused light evenly across a sapphire viewing
window, resulting in even reflected light from the soil. The standard GeoVIS optics system provides a
viewing field of approximately 2 by 3 millimeters and a magnification of 100 when viewed on a standard
13-inch monitor. Approximately eight unique (non-overlapping) photomicrographs can be collected per
inch of soil or 96 unique images per foot of video push. All soil photomicrographs are collected using a
frame capture device and can be saved as bitmap files. The pores between sand and gravel grains and
contents of the pores (such as dense NAPL, or dense nonaqueous phase liquid [DNAPL]) are generally
readily observable and easily definable from these soil microphotographs.
29
-------
Exhibit 4: Schematic of GeoVIS Direct-Push Probe
Mirror
Lense/Focusing System
CCD Color Video Camera
Sapphire Window
3.4.2 Effective Porosity
Porosity between fine-grained materials is part of total porosity that cannot be observed within the
photomicrographs; therefore, total porosity cannot be rendered. However, effective porosity as it relates
to the specific yield of the soil is extracted and quantified since the large pores between granular materials
can be observed. Pixels of grain and matrix materials are converted to pure white and pixels of pore
space are converted to pure black. The number of white versus black pixels is used to estimate the
percent pore space in the photomicrograph. Area percentages calculated from two-dimensional
photomicrographs can be used to estimate porosity, saturation, and void volumes by the consecutive
volume slice method. If a sufficient number of compositional determinations of two-dimensional slices
are conducted on a three-dimensional volume, then the composition of the volume can be reliably
estimated (Exhibit 5).
30
-------
Exhibit 5: Estimated Soil Porosity (Vadose Zone) from GeoVIS Images
Select area within white box and calculate
effective porosity (percent white vs percent black
area = pores). Area to volume assuming
consecutive slices.
Navy PWC San Diego SCAPS Project
3.4.3
NAPL Saturation
Obtaining NAPL saturations is more problematic than processing GeoVIS outputs for porosity. NAPL
color varies based on the thickness of the NAPL in the pores, the NAPL type, background reflectance,
pore size, and soil type. Dark, globular DNAPLs are easily rendered, whereas lightly colored fuels and
diesel are not rendered well. Another problem that may arise is that dark mica or other dark minerals can
also be misinterpreted as free product; therefore, the percentage of dark minerals must be known before
images can be processed for free product. It is recommended that images of fluorescing NAPL induced
by LIF be collected to overcome the highly variable nature of NAPL colors and reflectance under most
field conditions. After the NAPL areas are obtained from each photomicrograph, the areas can be
converted to NAPL saturation by dividing the area by the effective porosity average for the push. Exhibit
6 shows the results of the DNAPL image processing for a soil video profile. Photomicrographs of
DNAPL droplets are shown on the left. Results of the DNAPL image processing, presented in black and
white, are shown adjacent to each photograph. Pixel counts as total image area and DNAPL saturation
results are presented on the right. Saturation results are a percent within the pore space, assuming a
consistent porosity of 43.1 percent. The photomicrographs and image processing results both show a
large drop in DNAPL droplet numbers and sized and in DNAPL saturation with depth.
31
-------
Exhibit 6: Vertical Profile of Soil Photomicrographs with DNAPL Droplets
F'holonitci ograph
8 1 ft
8 3fl
11.1 B
DNAPL Area ('»<.) S^
:~ 11"..
t. "*•••..
I I •'".,
I."'!..
The information on this technology was adapted from "Confirmation ofCPT Video Microscope Estimates
of In Situ Soil Porosity andNAPL Saturations" (Sinfield, 2004).
3.5
ADDITIONAL SITE CHARACTERIZATION ACTIVITIES
In addition to the continued refinement of a CSM to support the evaluation of vapor intrusion issues at the
Hartford site, information should also be gathered to support the evaluation of the nature and extent of
dissolved- and free-phase contamination at depth and in off-site areas. The data available from the
investigations conducted thus far have focused on the presence or absence of free product immediately
beneath the Hartford site. Future investigations should also consider the engineering and characterization
data needed to optimize the proposed remedies for the site.
32
-------
As a first step in this process, the preliminary CSM should be expanded to include potential source areas
north, east, and west of the Hartford site. Without any information on these potential source areas, it is
difficult to provide detailed suggestions on delineation of dissolved-phase hydrocarbons at this time. The
project team should consider the following activities:
• Compiling existing data from areas surrounding the Hartford site.
• Continued piezometer installation and monitoring real-time flow directions and water levels
using pressure transducers in and around source areas where product recovery and SVE will
likely be applied.
• Additional water sampling from small-gauge, multi-level wells both on and off site. Small-
gauge wells should be installed with screened intervals within, and below, the identified
boundaries of the free-phase and dissolved-phase plume to monitor product thicknesses and
water levels.
• Additional source term characterization should be conducted such that mass loading can be
estimated and the fate and transport of petroleum contaminants can be predicted.
• Fate and transport parameters such as groundwater flow velocities and directions should be
mapped and calibrated using intrinsic tracers, or existing plume extents and characteristics,
where possible.
• Natural attenuation parameters should be added to the set of monitoring parameters for wells
in the distal fringe and surrounding the contaminant plumes.
Nested wells are needed throughout these areas for product removal, and monitoring wells should also
extend to some depth below the plumes for monitoring.
The well network currently proposed should be augmented with small-gauge wells or temporary
monitoring points whose locations are optimized based on the work products provided in this report. A
dynamic approach is suggested to limit project cost and improve performance. As mentioned in the CSM
portion of this report, there are obvious pathways for contaminant transport that will need to be refined
and targeted.
33
-------
4.0 INCORPORATING TRIAD-DRIVEN DYNAMIC WORK STRATEGIES
The general approach presented by Clayton and in the "Technical Memorandum, Vapor Control System
Upgrade Design " (Clayton 2004c) and the utility and pipeline investigation is a traditional static or
phased approach. Tetra Tech suggests that the project team instead consider adopting a dynamic work
strategy to guide future activities. A dynamic work strategy explains how the decisions will be guided by
field measurement results and how spatial uncertainties will be identified and addressed as the field
activities proceed. Ideally, stakeholders build consensus on project objectives and key decisions that are
based on agreed-on action levels before the new field activities are undertaken.
For example, if a sampling location is found to contain free product, the dynamic work strategy might
discuss how its extent will be delineated and what type of data will be used to support delineation.
Another example might include evaluating how headspace soil gas surveys, VMPs, PneuLog, or GeoVIS
will be used to optimize treatment system design. Data from these types of evaluations then might be
used to select specific VMP locations and locations where additional design data should be collected.
These types of strategies should be laid out using a series of flow charts and diagrams before field crews
mobilize.
Tetra Tech understands some of the basic reasons behind the phased approach Clayton proposed.
However, Tetra Tech believes it is imperative for the project team to clearly state the specific rationale
that will be used to select when and where various activities will be considered. The rationale should be
based on a well established CSM and address every aspect of the project design, including installation of
piezometers, direct-push soil borings, VMPs, soil and ambient air gas samples, full-size wells and
vacuum monitoring probes. Establishing clear guidelines for decision-making should be the first step in
the systematic planning process.
The revised CSM should be developed and used to select the most appropriate set of innovative
technologies for use at the site. A data management and communication platform that operates in real
time should be considered. Many three-dimensional data presentation tools are available. A web portal
with a relational database should also be considered to expedite communication of results. This portal
will facilitate project decision-making on a real-time basis and swift communication of results to all
regulators and affected stakeholders.
34
-------
4.1 SUMMARY OF SUGGESTIONS
The following summarizes the proposed dynamic work strategy to support the implementation of a vapor
mitigation program and product removal design effort at the Hartford site. The activities are designed in a
logical sequence such that the quality and utility of data that are collected is maximized. Maximizing the
utility of data will also require that the CSM for the Hartford site be continually revised as new data are
received. The CSM can communicate results to stakeholders as new data are received and can guide
subsequent actions. The CSM should be used as the basis to scope additional work and establish
contingencies and options that might need to be built into the removal strategy.
Suggested activities include, but are not necessarily limited to, the following:
• Hydrocarbon concentrations in near-surface preferential pathways such as sewers and utility
corridors should be analyzed using ambient air methods. These data are particularly
important because of the difficulty in using intrusive sampling methods near utility corridors,
such as Rand Avenue. Real-time ambient air monitoring technologies for sampling from
sewer manholes are the most promising method. A portable gas chromatograph or the
equivalent with a mass selective detection system are suggested.
• The well network should be expanded north, east, and west to answer not only vapor-related
issues, but also more long-range concerns related to impacts from the Hartford site to the
surrounding environment. Downhole transducers should be placed in wells as practicable so
that rapid fluctuations and changes in direction of flow can be monitored on a real-time basis.
Upgradient sources will likely need to be addressed to assure the long-term effectiveness of
the remedy. Technologies should be used that can increase well and screened interval
density. Small gauge multiport wells, such as can be obtained from Precision Sampling Inc.,
should be used to reduce costs and improve the project team's ability to monitor vertical and
horizontal off-site migration of contaminants.
• In addition to lithologic descriptions, any contaminant-related features such as odor, staining,
or unusual solid constituents should be noted on the logs. The visual observation of
hydrocarbon contamination should be documented using the following standardized
descriptions:
No Visible Evidence - No visible evidence of oil on soil sample;
Sheen - Any visible sheen in the water on soil particles as described by the sheen testing
method presented later in this section;
Staining - Visible brown or black staining in soil. Can be visible as mottling or in bands.
Typically associated with fine-grained soils;
Coating - Visible brown or black oil coating soil particles. Typically associated with
coarse-grained soils such as coarse sand, gravels, and cobbles;
35
-------
Oil Wetted - Visible brown or black oil wetting the soil sample. Oil appears as a liquid
and is not held by soil grains. Soils oozing petroleum typically contain 2 to 3 percent
petroleum.
• These descriptions are general and may need to be modified to more accurately reflect actual
site conditions and product characteristics.
• In addition to PID headspace and visual observations, the presence office product in soil
cores should also be evaluated periodically through direct application of a technology similar
to ROST on the open core at ground surface. This evaluation is particularly important across
the smear zone within the top of the Main Sand, where free product is most likely to be
present. As the visual evidence office product decreases, a water sheen test or SiteLAB total
petroleum hydrocarbon ultraviolet fluorescence (UVF) field test kits or the equivalent might
be considered to further understand the relationship between the measured concentration in
soil, the presence office product, and the ROST response.
• The water sheen test can be performed by placing soil in a small plastic bag or glass jar,
adding distilled water, shaking the bag or jar, and observing the water's surface for signs of
sheen. Sheen should be classified as follows:
No Sheen (NS) - No visible sheen on water surface;
Slight Sheen (SS) - Light colorless film; spotty to globular; spread is irregular, not rapid;
areas of no sheen on water surface remain; film dissipates rapidly;
Moderate Sheen (MS) - Light to heavy film; may have some color or iridescence;
globular to stringy; spread is irregular to flowing; few remaining areas of no sheen on
water surface;
Heavy Sheen (HS) - Heavy colorful film with iridescence; stringy in appearance; spread
is rapid; sheen flows of the sample; most of water surface may be covered with sheen.
Additional multiport VMP and well designs should be optimized using PneuLog.
Petroleum vapors invading sewer lines should be collected before the final SVE system is
designed and tested. Specific areas for testing should be identified through the use of the
CSM as it is revised based on the products developed during this effort and subsequent
data collection efforts.
• A near-surface monitoring network that might rely on VMP data and or headspace soil gas
should be used to evaluate the impact of free product removal and vadose zone SVE or other
methods for reducing vapors during remedy testing and before full-scale implementation.
The performance of the system should be checked against well-defined decision criteria
before full-scale implementation is considered.
• GeoVIS should be used along with other types of physical and empirical testing to evaluate
and map zones beneath the village where product removal should be considered. This type of
evaluation will improve the potential for the effective removal of product beneath the village.
36
-------
4.2 SUMMARY
The work products generated in support of the refinement of a CSM for the Hartford site are encouraging.
Tetra Tech hopes that the additional work products provided in this report will continue to expand the
project team's understanding of the Hartford site. Tetra Tech also hopes that the project team will
continue to refine and clearly state the decision logic that guides the activities in future work plans. The
project schedule should be sequenced to assure that field activities and remedial efforts are optimized.
Once optimized, locations where actions will be considered should be further tested using empirical
methods. As systems come on line, the HWG should continue to refine operating conditions and
parameters.
The Hartford site is complex, and implementation of an efficient remedy can be supported by all elements
of the Triad approach. Real-time measurement techniques can be used to make maximum use of data as
it is collected. The aggressive use of a systematic plan designed around the refinement of the CSM and
efficient communication of results is needed. Well-documented dynamic work strategies, which clearly
define how data will be used to support decision-making, will limit project delays. The collaborative use
of differing sources of information is needed to improve project efficiency.
Current data are not sufficient to address many of the principal study questions for the project. As the
project progresses, Tetra Tech suggests that the project team begin to consider ways to address as many of
the objectives as efficiently as possible through a dynamic approach designed around innovative
technologies and strategies.
Tetra Tech would be pleased to provide related available data sources on the products presented in this
report to the Hartford Working Group. If you have any questions or comments on this document, please
contact Mr. Robert A. Howe at (303) 441-7911 or via e-mail at Robert.Howe@ttemi.com.
37
-------
5.0 REFERENCES
Clayton Group Services, Inc. 2004a. "Investigation Plan to Define the Extent of Free Phase and
Dissolved Phase Hydrocarbons in the Village of Hartford, Illinois." Prepared for The Hartford
Working Group. January 7.
Clayton Group Services, Inc. 2004b. "FPH CPT/ROST Subsurface Investigation Report and FPH
Monitoring Well and Soil Sampling Plan for the Village of Hartford, Illinois." April 8.
Clayton Group Services, Inc. 2004c. "Technical Memorandum Vapor Control System Upgrade Design."
May 6.
Clayton Group Services, Inc. 2004d. "Work Plan-Dissolved Phase Groundwater Investigation, The
Hartford Area Hydrocarbon Plume Site." June.
Clayton Group Services, Inc. 2004e. "Letter to USEPA Region 5. Response to Comments to ROST
Investigation Report and Work Plan." June 21.
Clayton Group Services, Inc. 2004f "Utility and Pipeline Investigation Work Plan, The Hartford Area
Hydrocarbon Plume Site." June 29.
Clayton Group Services, Inc. 2006a. "Dissolved Phase Groundwater Investigation Report." Prepared for
The Hartford Working Group. January 4.
Clayton Group Services, Inc. 2006b. "Site Wide Free Product Investigation." Prepared for The Premcor
Refining Group, Inc. January 23.
Clayton Group Services, Inc. 2006c. "Proposal for an Active LNAPL Recovery System." Prepared for
the Hartford Working Group. February 2.
EPA. 1998. "Innovative Site Remediation Technology, Design and Application." Volume 7. Vacuum
Extraction and Air Sparging. 542-B-97-010. May.
Reading, H.G. (editor). 1996. "Sedimentary Environments: Processes, Facies, and Stratigraphy." Third
edition.
Sinfield, L.S., T. Latas, and W.E. Collins. 2004 "Confirmation of CPT Video Microscope Estimates of
In Situ Soil Porosity and NAPL Saturations."
Tetra Tech EM Inc. (Tetra Tech). 2004. Robert Howe's personal communication with Bill Davis
regarding available technology. Tricorder Inc. July.
38
-------
FIGURES
-------
Hartford
Hydrocarbon
Plume Site
— '"/blLL^t
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 1
LOCATION MAP
-------
R:\Clients\Prem Cor\Clavton\ Fiq2_PREMCOR-GEO_X-SECTION.dv»q 07/23/2007 deborgh.ford DN
A'
500'
400'-
300'-
LEGEND
L
| CAHOKIA ALLUVIUM
MAIN SAND
RENAULT LIMESTONE - MISSISSIPPIAN BEDROCK
WATER
WATER TABLE SURFACE
GROUNDWATER FLOW DIRECTION
r100'
-50'
-0
VERTICAL SCALE:
1" = 100'
APPROX.
SCALE: 1" = 2 MILES
HORIZONTAL
HARTFORD AREA
HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 2
REGIONAL GEOLOGIC CROSS SECTION
TETRATECHEM INC
-------
R:\Clients\Prem Cor\Clavton\ Fig 3_FGeoloqic-cross-section.dwq 07/23/2007 deborah.ford DN
B
SITE BOUNDARY
B'
PIPELINES
PREMCOR PONDS
MISSISSIPPI RIVER
VILLAGE OF HARTFORD
n n n n n ni
I
REFINERIES
?ZL- CAHOKIA ALLUVIUM (25-45' THICK)
LEGEND
WATER
WATER TABLE SURFACE
APPROXIMATE DIRECTION OF
GROUNDWATER FLOW
SILTY CLAY T
NORTH OLIVE SAND *i
^^•;V".-':'Vv'V;':';;V;:j:^.;:---vV-
i ' i ' ii 'K-v •'-;''•"•;•; •••;,:";.N.;-;.-:;-.
MEMBER . ', :;.
HENRY FORMATION; '. r.
(60'-150' THICK) ••••. ••'.'•••
MODERN RIVER ALLUVIUM
RAND SAND <
EPA SAND
MAIN SAND
(MAKINAW MEMBER)
RENAULT LIMESTONE
(MISSISSIPPIAN BEDROCK)
CONSISTING OF FLOOD PLAIN AND
CHANNEL DEPOSITS.
UNCONSOLIDATED, POORLY
f* SORTED SAND, SILT, AND CLAY
HENRY FORMATION - SAND AND
GRAVELS
MISSISSIPIAN AGE BEDROCK
SANDY TO CLEAN LIMESTONE
-20'
GEOLOGY REFERENCED FROM CLAYTON GROUP SERVICES
APRIL 8, 2004 "FPH CPT/ROST SUBSURFACE
INVESTIGATION REPORT AND FPH MONITORING WELL AND
SOIL SAMPLING PLAN FOR THE VILLAGE OF HARTFORD,
ILLINOIS"
0.5
0.5
1
SCALE: 1" = 1 MILE VERTICAL SCALE:
HORIZONTAL 1" = 40'
APPROX.
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD. ILLINOIS
FIGURE 3
GEOLOGIC CROSS SECTION SHOWING
FORMATION DISTRIBUTION IN AND
SURROUNDING THE HARTFORD AREA
HYDROCARBON PLUME SITE
TETRATECHEM INC
-------
SOURCE: USGS 7.5 MINUTE SERIES TOPOGRAPHIC MAP (WOOD RIVER. ILL - rev. 1994)
BLOCK LAYOUT
SCH£i r • «tf (HORZ.)
SCALE: !' - 2tf (VERT. - WROX).
POND/RIVER
CONOCO PHILUPS PROPERTY
BLOCK BOUNDARY
GROUNDWATER TABLE
RAILROAD TRACKS
REFINERY/PRIVATE ROADS
UNKNOWN BOUNDARY
VEGETATED AREA
Hartford Area
Hydrocarbon Plume Site
Hartford, Illinois
PERMEABLE LENS
SILTY CLAY
NORTH OLIVE SAND
RAND SAND
EPA SAND
MAIN SAND
SHELL PROPERTY
PREMCOR PROPERTY
FILL
FIGURE 4
PRELIMINARY GEOLOGICAL
CONCEPTUAL SITE MODEL
HEAVY RANGE PETROLEUM HYDROCARBON
MID-HEAVY RANGE PETROLEUM HYDROCARBON
MID RANGE PETROLEUM HYDROCARBON
LIGHT RANGE PETROLEUM HYDROCARBON
HARTFORD PUBLIC WATER WELLS
ABOVEGROUND STORAGE TANK
-------
HROST-118
HROST-61 V
HROST-85
HROST-1
•tff.
HROST-126
HROST-22
I
HROST-23M
I HROST-25
HROST-97
HROST-19 -\
HROST-30
HROST-31a
•^ m
-1 HROST-116 x HROST-114
HROST-36
HROST-40
HROST-39
HROST-37 HROST-38a
—J HROST-43
%^r
HROST-49a
® HROST-45
; / HROST-49
SKt _
ffHROST-60
M —I ±
Legend
® ROST Sample Location
™ ™ ROST Response Boundary
Pond
Railroad
Parcel
Extent of Free Product Plume
Building
Building with Reported Fire
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 5
ROST APPROXIMATE
EXTENT OF HYDROCARBON MAP
-------
0 1000ft
0 300
Solid
highe
of levee
;X^Y \ {\ ^-Splay fill of abandoned channel
(From Reading, 1998, "Sedimentary Environments)
- V
Coalescing crevasse splays
(natural levee)
HARTFORD AREA
HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 6
AVULSION OF A SEDIMENTARY SEQUENCE
TETRA TECH EM iNC
-------
300
275 A
Levee
Cut bank
ft
Mud
250-
225-
Silt and mud
Vertical exaggeration=2Qx
Mud sheets Sand, wood and
mud fragments
(From Reading, 1998, "Sedimentary Environments)
HARTFORD AREA
HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE?
CROSS SECTION OF A MIXED LOAD RIVER
TETRA TECH EM INC.
-------
\l ^
HROST-89 (t)
F
HROST-92
HROST-93
HROST-90
® HROST-2
1
HROST-6
HROST-130
•\ \
HROST-10
\ .\\
HROST-125
HROST-23
LJ I. !
HROST-17.ERRVST
HROST-126
HROST-22
HROST-97 €
I
HROST-99
HROST-98-I—&
HROST-65
I HROST-78
HROST-77
HROST-251 •*• i V i
1—'ii.\JHROST-26
HROST-27
HROST;101
HROST-28
HROST-124
I I.-J |
HROST-117
HROST-13
\ \
HROST-38a
HROST-123
HROST-34
I
HROST-103
HROST-40
\ \
HROST-39
HROST-74
HROST-72
m
HROST-49
HROST-49a
HMW-26
HRpSJ-48
HROST-'Y06
T I J
HROST-60
HROST-112
-V \
HROST-56
1. Geology referenced from Clayton Group Services April 8, 2004 "FPH
CPT/ROST Subsurface Investigation Report and FPH Monitoring
Well and Soil Sampling Plan for the Village of Hartford, Illinois".
2. Main contour is depicted for area where Main Sand is the uppermost
sand unit. Contour also serves as isopach of silty clay above Main Sand.
I
Legend
ROST Sample Location
HMWWell
Contour of Thickness of A-Clay above North Olive (feet - depth from surface)
Contour of Thickness of A-Clay above Main (feet - depth from surface)
Contour of Thickness of A-Clay above Rand (feet - depth from surface)
Extent of Rand Sand — - Pinched-out Boundary for North Olive
Pond i— i _ .. .. 0 400
Railroad
Parcel
Building
Building with Reported Fire
Feet
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 8
ISOPACH OF A-CLAY ABOVE
STRATUMS INDICATED
TETRATECH EM INC.
-------
HROST-22
\V \
HROST-126
.,
\\
HROST-25
i\JHROST-26
HROST-40
V\
rr^- HROST-39
HROST-37 HROST-383
HMWr26HROST-49a
HROST-66
=D IPf
HROST-60
II DDE
HROST-107
OBP
HROST-111
HROST-112
\
HROST-56
NOTE: Geology referenced from Clayton Group Services April 8, 2004 "FPH
CPT/ROST Subsurface Investigation Report and FPH Monitoring
Well and Soil Sampling Plan for the Village of Hartford, Illinois".
/
Legend
ROST Sample Location
HMWWell
North Olive Thickness (feet)
Pond
Railroad
Parcel
Building
Building with Reported Fire
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 9
NORTH OLIVE STRATUM ISOPACH
TETRATECH EM INC.
-------
HROST-92
HROST-93
HROST-94
HROST-
I
HROST-128®
HROST-129
HROST-10
Y\
HROST-76
X
HROST-125V,
\
HROST-15
HROST-133
HROST-96(t)
mi
HROST-20
HROST-97
-------
W RAMP AVF
HROST-90S) HROST-3
HROST-126
•
HROST-23
HROST-20
HROST-98
HROST-65
HROST-77
HROST-27
HROST-101
HROST-31a\V-7^
HROST-44 HROST-7CK / /
' -cr rHRosW E
UHMW-26-1HI
H1 HROST-48
HROST-106
PPP
HROST-50
innn
HROST-109
HROST-110
inHi
Geology referenced from Clayton Group Services April 8, 2004 "FPH
CPT/ROST Subsurface Investigation Report and FPH Monitoring
Well and Soil Sampling Plan for the Village of Hartford, Illinois".
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
ROST Sample Location
HMWWell
Thickness of Rand in feet
Pond
Railroad
Parcel
Building
Building with Reported Fire
FIGURE 11
THICKNESS OF RAND
STRATUM IN FEET
-------
HROST-3
I
Kl HROST-5
-------
HROST-88'
HROST-2^HROST-3
HROST-92
HROST-8
HROST-9lT:
HROST-6 \ \
XHROST-12
HROST-11 f\\
HROST-14
1TW1
*-HROST-17
B28
HROST-43
/T
NOTE: Geology referenced from Clayton Group Services April 8, 2004 "FPH
CPT/ROST Subsurface Investigation Report and FPH Monitoring
Well and Soil Sampling Plan for the Village of Hartford, Illinois".
/ ; W 1ST STRICT I 4- I
HROST-57
Legend
IT
=5
ROST Sample Location
HMWWell
EPA Contour (feet - depth from surface)
Main Contour (feet - depth from surface)
Pond
Railroad
Parcel
Building
Building with Reported Fire
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 13
TOP OF EPA STRATUM
TETRATECH EM INC.
-------
\( >> HROSTr90 =
HROST-89
HROST-92
HROST-93
HROST-7
HROST-94
HROST-10
© HROST-80
f \—1—X—HROST-76
HROST-122JPI
HROST-65
HROST-77
I
HROST-27
HROST-101
HROST-124
HROST-30
I \\
HROST-31a
HROST-115
HROST-112
HROST-56
NOTE: Geology referenced from Clayton Group Services April 8, 2004 "FPH
CPT/ROST Subsurface Investigation Report and FPH Monitoring
Well and Soil Sampling Plan for the Village of Hartford, Illinois".
Legend
.
§
Q
ROST Sample Location
EPA Merges with Main Boundary
EPA Thickness (Feet)
Pond
Railroad
Parcel
Building
Building with Reported Fire
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 14
EPA STRATUM ISOPACH
TETRATECH EM INC.
-------
HROST-91
® HROST-79
\ _
f. ~T ^£ BIRCH ST \
lT^HROST-125
HROST-12
HROST-76
\\\
HROST-15
\\\
HROST-22
SfHROST-24
* i
HROST-29
QbHROST-25
HROST-30
HROST-33
\
HROST-13
HROST-39
HROST-40
I
HROST-70
± HROSTr49f
HROST-106
XJh
HROST-60 i
-------
UJ
F
HROST-114
HROST-33
HROST-13
\\
HROST-40
HROST-39
_L
NOTE: Geology referenced from Clayton Group Services April 8, 2004 "FPH
CPT/ROST Subsurface Investigation Report and FPH Monitoring
Well and Soil Sampling Plan for the Village of Hartford, Illinois".
/ ; W 1ST STRICT I 4- I
HROST-57
Legend
§
o
ROST Sample Location
HMWWell
Main Contour in feet
Pond
Railroad
Parcel
Building
Building with Reported Fire
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 16
TOP OF MAIN SAND
TETRATECH EM INC.
-------
HROST-61
HROST-118' A
HROST-12
HROST-11
HROST-19 HROST-2Q
c
HROST-97
I
HROST-65
HROST-9S
HROST-29
HROST-28
HROST-77
HROST-27
HROST-101
HROST-30V2.0
HROST-31a
Q
t HROST-78
HROST-123
WWELM&T j
HR_OST-35
HROST-34
' ' I^J-f
HROST-103/-77 4 |r
/ / HROST-74
HROST-41/y/ HROST-72'
HROST-40
HROST-39
HROST-4S
HROST-49
HROST-67
u
HROST-55
HROST-109
W. MAPLE
HROST-112
\ 1 H.
HROST-56
HROST-57
m
NOTE: Hydrocarbon thickness and extents referenced from Clayton Group Services
April 8, 2004 "FPH CPT/ROST Subsurface Investigation Report and FPH
Monitoring Well and Soil Sampling Plan for the Village of Hartford, Illinois".
/ /
Feet
Legend
• ROST Sample Location - no apparent product
9 ROST Sample Location - Light Range Product
9 ROST Sample Location - Mid Range Product
O ROST Sample Location - Heavy Range Product
— — Product Boundary
Product Contour in feet, depth below surface
Building
| Building with Reported Fire
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
Light Range Product Boundary
Mid Range Product Boundary
Heavy Range Product Boundary
Pond
Railroad
Parcel
FIGURE 17
TOP OF TOTAL PETROLEUM
HYDROCARBON ROST RESPONSE
TETRATECH EM INC.
-------
f
0 200 400
NOTE: Geology referenced from Clayton Group Services April 8, 2004 "FPH
CPT/ROST Subsurface Investigation Report and FPH Monitoring
Well and Soil Sampling Plan for the Village of Hartford, Illinois".
Legend
1
CL
CO
5 to < 10 Feet of total ROST Response
10 to < 20 Feet of total ROST Response
20 to < 30 Feet of total ROST Response
30 to < 40 Feet of total ROST Response
Building
Building with Reported Fire
Pond
Railroad
Parcel
Note: < Less Than
Sources:
Clayton Group Services, Inc. 2006a.,
Clayton Group Services, Inc. 2004b.,
Clayton Group Services, Inc. 2006c.
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 18
TOTAL PETROLEUM HYDROCARBON
ROST RESPONSE THICKNESS
TETRATECH EM INC.
-------
, -» —._. V
x x ---x
X *X >-. •'
\ N N.. ' --,.. /
Legend
Top of Sand
(feet - depth from surface)
Product Pipeline
Municipal Sewer
Industrial Sewer
Forced Main
Pond
Railroad
Parcel
Less than 8 Feet Below Ground Surface
8-10 Feet Below Ground Surface
10-12 Feet Below Ground Surface
Building
Building with Reported Fire 0
400
HARTFORD AREA HYDROCARBON PLUME SITE
HARTFORD, ILLINOIS
FIGURE 19
LOCATIONS WHERE POROUS STRATUM
(SAND UNITS) ARE POTENTIALLY IN
CONTACT WITH PRODUCT PIPELINES
AND MUNICIPAL SEWERS
TETRATECH EM INC.
-------
CO
o
o
CO
o
Oi o
CN CN g
lO "(O CD
Q.
LU
O)
,0
CD
CO
-
O CD
CD E
O CD
t E
O LJJ
CD
E
Q.
O
Q. co
9- o
<^ E
o O -O
•£. co >^
CO 2 ^
5 Q- co
.O Q. CD
CD ;« "O
"co H !§
O CD CO
O £ I
o
CD
O)
O
o
CO
m
W
o
CN
O
I o
: ^-x O)
dcN:i
•^ o co
? T- CO
CD
CO
=>
% §
IE
=1 o
m ^
"CO ±i
il
O Q.
------- |