W912DR18R0055_-_Attachment_E_-_OU-6_Optimization_Remedy_-_7-20-2018.pdf

arcadis.com P:\Tech Memos\OU6 GW Optimization\Final\OU6 Optimization Remedy - 7-20-2018.docx

Arcadis U.S., Inc.

2101 L Street NW

Suite 200

Washington

DC 20037

Tel 202 912 8100

Fax 202 887 5168

Page:

1/14

MEMO

To:

Steve Wood, USACE

Brian Shedd, USACE

Scott Forbes, USACE

Copies:

Courtney Ingersoll, Arcadis

Kathleen Hamrick, Arcadis

Scott Potter, Arcadis

Mathew Schnobrich, Arcadis

From:

Christopher Spooner, Arcadis

Date Arcadis Project No.:

July 20, 2018 02118201.0001.01900

Subject:

DSCR OU 6 Remedy Optimization

INTRODUCTION AND PURPOSE

This memorandum presents a recommended remedial approach for the Operable Unit (OU) 6 groundwater impacts at Defense Supply Center Richmond (DSCR). The recommended remedial approach presented is based on a review of current site conditions, historical site information, previously implemented remedial actions, and performance monitoring data assessment.

OU 6 consists of impacted groundwater located beneath and downgradient of the Open Storage Area

[OSA 9 (OU 1)], Area 50, a former landfill (OU 2), and the National Guard Area [NGA (OU 3)] located in the central portion of DSCR. Two areas (OU 1 and OU 2) are suspected to be the major sources of groundwater contamination at OU 6. Historical activities in the third area, OU 3, likely contributed to groundwater impacts historically, but based on recent groundwater data does not appear to be a major source of impacts to groundwater. The impacted groundwater consists of chlorinated volatile organic compounds (CVOCs) found in the Upper Water Bearing Unit (UWBU) and the Lower Water Bearing Unit

(LWBU). The plumes of these constituents extend off the installation boundary toward No Name Creek to the east. Multiple remedial actions have been implemented to manage and remediate dissolved-phase impacts at OU 6, including: groundwater extraction and treatment, dual phase extraction (DPE) pilot testing, in-situ bioremediation (ISB), and monitored natural attenuation (MNA). The current MNA program includes a combination of long-term monitoring (LTM), institutional controls (ICs), and air sampling.

Historical remedial actions have focused primarily on the dominant source area transport pathways and diffuse groundwater impacts opposed to direct source treatment. As a result, progress demonstrates that

P:\Tech Memos\OU6 GW Optimization\Final\OU6 Optimization Remedy - 7-20-2018.docx Page:

2/14

DSCR OU 6 Remedy Optimization source mass discharge is ongoing and contributes to plume migration. While source area remediation was attempted, pre-remedial investigation to understand the vertical or horizontal distribution of impacts appears to have been insufficient. The resulting limited understanding of the distribution of source area impacts likely constrained the effectiveness of previous source treatment attempts. Based on a review of previous OU6 remedial actions as well as the primary treatment objectives described in this technical memorandum, the potential exists for expedited source area and dissolved-phase treatment via remedy optimization.

BACKGROUND

Geology

Geologic cross-sections were developed by AECOM (2013) at two separate areas that included the primary source areas of OU 6. These areas include the northwestern portion of OU 6, which includes a large northern part of OU 1 and a smaller northwestern part of OU 2 (identified from here on as OU 1

Area) and the southeastern portion of OU 6, which includes OU 3 and a smaller southeastern part of OU 2

(identified from here on as OU 2 Area). The stratigraphic interpretation of OU 1 and OU 2 Areas are as follows:

OU 1 Area (northwest plume area)

The UWBU consists of several units of unconsolidated sediment. The top unit below the ground surface generally consists of clay (upper clay). Within this area, this clay layer ranges in thickness from approximately 11 to 20 feet. The saturated thickness of the clay ranges from approximately

2 to 7 feet. Across a portion of the cross-section, the clay is overlain by a thin unit of gravel and asphalt. The upper clay is underlain by an approximately 4 to 7 feet thick unit of sand that is underlain by a coarser layer of gravelly sand approximately 1 to 6 feet thick. Collectively these units comprise the UWBU.

The UWBU is underlain by a 3 to 8 feet thick lower clay unit that generally acts as an aquitard between the UWBU and LWBU.

Below the lower clay unit, gravelly sand and well sorted sands comprise the LWBU. A layer of gravelly sand 8 to 10 feet thick is underlain by at least 10 feet of well sorted sand.

OU 2 Area (south-east plume area)

Like the OU 1 Area the UWBU consists of several units of unconsolidated sediment. The highest unit in the sequence (i.e. directly below the ground surface) generally consists of clay. Within OU-

2, the thickness of this upper clay ranges from 2 to 22 feet. The saturated thickness of the clay ranges from 0 feet near No Name Creek (off installation) to approximately 16 feet on the western end of the plume source. The upper clay on the western part of the plume source is overlain by 1 to 8 feet thick organic material and fill. The upper clay is underlain by a 2 to 5 feet thick sand unit.

The sand unit of the UWBU is underlain by the lower clay layer with a thickness of 4 to 16 feet.

This lower clay unit is believed to be contiguous with the lower clay unit described above (OU-1 area).

The lower clay is followed by 8 to 16 feet of silty sands and a basal unit of glauconitic sands with gravel and cobbles. Collectively these units comprise the LWBU.

3/14

DSCR OU 6 Remedy Optimization

Previous Remedial Approaches, Pilot Tests and Treatability Studies

The remedial approaches, pilot tests and treatability studies previously conducted at OU 6 are summarized below and shown on Figure 1.

i) Groundwater Extraction and Treatment (1996 – 2006) – an interim remedial action, designated as OU 9 on the northeastern part of OU 6, extracted and treated UWBU and

LWBU groundwater for approximately nine years beginning in 1996. This involved a network of 22 extraction wells along the installation eastern boundary, above-grade CVOC treatment, and a discharge system. The treatment system underwent an effectiveness evaluation and the system was placed in a standby status in January 2006. The effectiveness evaluation led to the decommissioning of the OU 9 extraction and treatment system in 2008.

ii) DPE Pilot Test (1999) – A DPE pilot test was conducted at the southeast plume of OU 6 (near

OU 3) in the UWBU and the LWBU. Pilot test results were unfavorable and thus DPE was not chosen as a remedial approach at OU 6.

iii) Enhanced ISB Treatability Tests (2007) – Enhanced ISB treatability tests were implemented at OU 2 and OU 3 portions of the OU 6 plumes. These consisted of injection of edible oil emulsion in the UWBU and the LWBU to enhance natural biodegradation of the chlorinated solvent impacts. Based on the favorable test results, ISB with emulsified vegetable oil (EVO) injection was adapted as part of the final remedy at OU 6.

iv) MNA and LTM including ICs and Air Sampling (2013 – present) – A record of decision identifying the final remedy for OU 6 was signed in September 2013. The final remedy for OU

6 consisted of ISB, MNA and LTM, including ICs and air sampling.

v) ISB (2015 - 2016) – The ISB component of the final remedy, consisting of injecting EVO through permanent injection well transects in the source areas and downgradient areas of the

OU 6 plume (Figure 1) was implemented in July/August 2015. Additional gravity feed EVO injections were completed in December 2015 and March 2016 to address hot spots and enhance distribution in additional areas of the plume.

A review of the historic chlorinated volatile organic compound (CVOC) concentration data throughout OU

6 shows that enhanced dechlorination is generally occurring; however, the presence of residual EVO in most OU 6 source area monitoring wells has greatly limited the necessary data available for performance evaluation. A more thorough source area conceptual site modle (CSM) evaluation is needed to facilitate a more comprehensive OU 6 remedial approach (i.e. to reduce the highest concentrations in source areas, eliminate migration, reduce remedial timeframes, remediate off-installation impacts and support the remedy transition to MNA/LTM).

REMEDIAL OPTIMIZATION OBJECTIVES

OU 1 Area

The following are the objectives of the remedy optimization in the OU 1 Area:

Enhance source area remediation;

Reduce remedial timeframes; and, Facilitate MNA throughout the OU 1 Area.

4/14

DSCR OU 6 Remedy Optimization

OU 2 Area

The following are the objectives of the remedy optimization in the OU 2 Area:

Enhance source area remediation;

Focus remediation of residual off-site CVOC impacts to facilitate abandonment of all off-site infrastructure prior to expiration of the current easement;

Eliminate future off-site migration of dissolved phase CVOC impacts;

Reduce remedial timeframes; and, Facilitate MNA throughout the OU 2/OU 3 Area.

CONCEPTUAL REMEDIAL STRATEGY

The recommended conceptual optimization strategy is to use existing and additional injection wells for injecting carbon donor into the UWBU within the installation boundary and the LWBU off-site east of the

OU 3 boundary. The proposed well networks at OU 1 and OU 2 Areas are presented in Figure 2 and

Figure 3, respectively. Permanent injection wells are recommended in areas where the saturated vertical thickness of the upper clay is limited and/or where access is limited. Selection of the carbon donor should be based on geology and estimated groundwater velocity in each area. The pre-design investigation (PDI), described below, should be used to confirm the appropriate carbon donor for each area.

In the OU 2 source area, direct push technology (DPT) wells should be used as part of the UWBU injection network. DPT injection points may provide more flexibility for injecting within the upper clay by allowing the use of a shorter screen and higher injection pressure than a typical injection well. Due to the presence of the thick clay layer overlying the UWBU (approximately 22 feet in the area where the DPT transect is proposed) at the OU 2 source area, DPT injections are likely more effective at delivering carbon donor in the saturated upper clay. Another potential benefit of the DPT transects is the ability to inject into the entire vertical thickness of the UWBU though a single point. Permanent wells would require nesting to enable targeting the upper clay and the lower sand and gravel. The use of DPT transects allows for adaptability during future injections to distribute carbon substrate where needed without having to install additional well infrastructure.

In addition to the optimized injection approach at the OU 2 source area, the application of hydraulic containment in both the UWBU and LWBU at the property line is recommended. The system could include the continuous use of extraction wells with either recirculation or surface water discharge of treated groundwater. The downgradient extraction wells would provide capture of impacted groundwater and prevent further off-site migration.

If designed as a dynamic groundwater recirculation system, continuous injection and groundwater extraction should create recirculation cells within the aquifer that: 1) increase the hydraulic gradient and groundwater velocity through the treatment zone, 2) allow hydraulic control over the most impacted groundwater within the treated volume of the UWBU and LWBU, and 3) provide higher subsurface flushing rates. All of which will expedite remediation of off-site impacts. The extracted groundwater would be treated ex-situ and returned via the injection wells. The hydraulic barrier could potentially also be used to supply water for injection in both permanent injection wells and DPT transects used for ISB. Additional details on the recommended optimal remedial strategies for OU 1 and OU 2 areas are described below.

5/14

DSCR OU 6 Remedy Optimization

Pre-Design Investigation

Limited investigation data has been collected to delineate the lateral and vertical extent of the source areas in both OU 1 and OU 2. To focus remediation and maximize remedial effectiveness, a PDI will be done to support historical information and define each source area with updated groundwater data. The results of the PDI will be used to confirm the placement and number of injection wells and DPT transects.

CVOC distribution will be used in conjunction with groundwater velocity estimates to strike a balance between the size of the injection network and the estimated remedial timeframe to achieve treatment between individual transect locations.

In 2015 and 2016 49 UWBU and LWBU wells in the OU 1 and OU 2 were used to inject EVO. Due to the continued presence of EVO, verifiable analytical data can’t be collected from 27 monitoring wells included in the sampling program. To expedite rehabilitation of the monitoring well network and allow for collection of representative aquifer COVC and geochemical data, redevelopment of all 2-inch monitoring wells impacted with EVO is recommended. The EVO present in the annular space and area immediately surrounding each well pressed into injection service will have to degrade before representative data can be collected; however, use of appropriate redevelopment methods should allow for sample collection to assess remedy performance. There are numerous ¾-inch wells with low redevelopment potential that should be abandoned and replaced. Off-site monitoring wells used as injection wells also likely require abandonment and replacement such that the off-site impacts can be characterized and remediated.

As described in the geology section above, the UWBU beneath the OU 1 and OU 2 Areas is overlain by a relatively thick clay layer. Within the western parts of OU 1 and OU 2 Areas where the highest CVOC concentrations have been measured, the saturated thickness of this clay unit can range between 7 and 16 feet, respectively. This suggests the presence of source mass within the clay layer that could act as a continuous source of impact to the underlying sand and gravel layers. To complete delineation of each source area, a direct push Membrane-Interface Hydraulic Profiling Tool (MiHPT) should be used to collect

CVOC data in soil and groundwater and hydraulic characteristics from OU 1 and OU 2 Areas where the highest concentrations of CVOCs have previously been measured (near MW-132 at OU 1 and near MW-

61 and MW-69 at OU 2). Results from this investigation will be used to identify locations of most significant impacts and best direct the organization of the injection network.

The MiHPT borings should be installed from the surface to the top of the lower clay. While final quantities and locations should be defined through subsequent technical communications with the USACE and the

DLA at present 20 to 30 MiHPT borings are recommended to complete source area delineation of the OU

1 and OU 2 source areas. Following installation of the MiHPT borings, soil and groundwater samples should be collected from 5-7 borings in each OU to quantify impacts identified by the MiHPT borings and correlate data from other MiHPT locations. It is anticipated that approximately two weeks should be allowed to complete the installation and analysis of the MiHPT borings at each area requiring a total of approximately 4-weeks to complete the PDI at the OU 1 and OU 2 Areas.

Optimized Injection Approach

The conceptual optimized injection approach focuses on primary transport pathways and source treatment to reduce remedial timeframes. It has been assumed that existing well infrastructure can be re-developed prior to implementation and that the achievable injectability observed previously at existing injection wells can be restored. As needed, additional injection locations may be added to address potential gaps in the well network attributed to poorly performing wells. The conceptual approach may also be adjusted pending results of the PDI investigation and collection of an updated comprehensive set of groundwater data.

6/14

DSCR OU 6 Remedy Optimization

Injections should be completed by extracting water from injection water supply wells 2-3 days prior to the start of an injection event. The extracted water will be treated and stored for later injection usage. It is anticipated the injection rate will be higher than the extraction rate and therefore the objective is to fill the frac tanks prior to starting the injection. This water will then be re-injected into the aquifer at the proposed injection wells after being amended in-line with the appropriate volumes of carbon donor. The carbon donor will be stored on-site in 275-gallon totes. A metering pump should be used to introduce the appropriate amount of carbon. An in-line static mixer should be used to mix the injected solution.

Carbon Donor Selection and Injection Transect Spacing

Carbon donor selection and injection transect spacing was evaluated for each injection location within each injection network. Additionally, selection of potential carbon donors was based on area-specific geology, estimated groundwater velocity, and the anticipated distribution in the subsurface. Groundwater velocity was estimated using water table gradients, the assumed mobile fraction of the porosity in each treatment area, and hydraulic gradients calculated from the results of hydraulic testing conducted in OU 6.

Transect spacing was evaluated based on a balance of anticipated carbon donor injection distribution, organic carbon longevity, the behavior of mobile carbon transport within the reactive zone, the estimated groundwater velocity, and the number of pore flushes required to reach cleanup goals. Collectively, these results were used to establish the estimated remedial timeframe based on different transects spacings and organic carbon injection for operational life-cycle cost determination. Selection of specific carbon donors for each area will be completed following the PDI to provide a carbon donor tailored to the geology and geochemistry of each area.

The primary differences to consider between soluble and semi-soluble donors at OU 6 are differing subsurface distribution behaviors (during injection events or between events), the unique subsurface transport characteristics, and longevity. While semi-soluble donors (i.e., vegetable oil-based products) are able to persist for over 24 months, they are sequestered within the soil matrix within several weeks following injection and do not move significantly beyond the injected radius of influence (ROI) in unconsolidated soil matrices. Due to soil straining behavior which can reduce the radial distribution during injection, delivery of these reagents typically requires larger injection volumes to achieve the desired distribution. Soluble carbon donors are more easily distributed in the subsurface and not impacted by these straining effects, so will move downgradient from the injection location with advective groundwater flow. Where meaningful groundwater velocities exist (e.g., greater than 0.3 ft/day), this natural transport allows expansion of the reactive zone for treatment and increases overall residence time. In contrast to semi-soluble donors however, soluble carbon donors persist in the sub-surface for 3-9 months and require more frequent injections to sustain electron donor for treatment.

While the longevity of each carbon donor determines the injection frequency, the distribution of each carbon donor impacts the transect spacing and remedial timeframe. Because semi-soluble donors create a barrier that does not move with groundwater, the distance between barriers must be compared to that anticipated with soluble donors to ensure that the transport time (and operational treatment time) is consistent with remedial objectives. This concept is presented graphically below.

7/14

DSCR OU 6 Remedy Optimization

Conceptual remedial strategies associated with reactive and flushing zones in soluble carbon approaches or in faster groundwater flow conditions (A), and in insoluble substrate approaches or under slower groundwater flow conditions (B).

Table 1 shows hydrogeologic characteristics for each proposed injection area used to develop carbon donor and transect spacing recommendations.

Table 1: Hydrogeologic Properties of Proposed Injection Areas

Area Gradient (ft/ft) Hydraulic

Conductivity (ft/day)

Groundwater Velocity

(ft/day)

OU 1 Area 0.0035 8.2 0.24

OU 2 Area – UWBU 0.006 8.2 0.41

OU 2 Area - LWBU 0.0035 3.5 0.12

Based on the groundwater velocity within each area, the water bearing units are generally conducive to either soluble or semi-soluble donors in all areas – but the parameters do have a bearing on transect spacing and delivery frequency. The velocity is slow enough to allow sufficient residence time in a semi-soluble donor barrier, while fast enough to allow advective expansion of the reactive zone with soluble donors. Additionally, both types of carbon donor are appropriate for the geology in the targeted injection interval in all areas, with the exception of the upper clay in the OU 2 UWBU area. Based on these considerations, the proposed optimization well networks were evaluated to support the recommended carbon donor selection and transect spacing. These recommendations are provided below for each area.

8/14

DSCR OU 6 Remedy Optimization

OU 1 Area

Based on the 2014 solvent concentrations in this area, the above estimates of groundwater velocity, and an assumed VOC retardation factor of 1.5, it’s estimated approximately 11.5 pore flushes are needed to remediate the source area to achieve the maximum contaminant level (MCL) remedial goals. In this area, the distance between transects (as proposed) is approximately 60 feet. Based on the target ROI (15 feet) the transport distance between injection barriers is approximately 30 feet (upstream to downstream between the anticipated edges of radial carbon injection). If a semi-soluble carbon donor is used and is retained within the injection radius, approximately 11.5 years of flushing time would be needed to achieve the estimated number of pore volume flush. Utilization of a soluble donor with the same barriers, however, will enable advective carbon transport behavior downgradient of each injection barrier and reduces this distance to approximately 9 feet. This reduction translates to shorter overall operational time, with approximately 1.5 years necessary to achieve the estimated 11.5 pore flushes. These timeframes do not account for the time to establish appropriate geochemical conditions and achieve complete dechlorination within each barrier, so an additional six months of lag time was assumed for these estimates. Assuming six months for each carbon donor to facilitate reductive dechlorination, the estimated remedial timeframes associated with each carbon donor would result in approximately 4 injections of semi-soluble donor (one injection every 3 years) and 6 injections of soluble donor (one injection every 4 months).

Based on the existing and proposed transect spacing and the objective to reduce remedial timeframes across the site, a soluble carbon donor injected using the proposed network is recommended. While this recommendation results in more injection events, the reduced cost of available soluble carbon donors

(e.g., molasses) combined with efficiencies realized in the field (numerous field activities ongoing concurrently) should result in overall life-cycle cost reduction at OU 6.

OU 2 Area – UWBU

Timeframe estimates for OU 2 were based on 2014 solvent concentrations in this area, a 30-foot spacing between transects, and the assumptions specified above. It’s estimated approximately 11.2 pore flushes are needed to remediate the source area to MCLs. Based on the target ROI (5 feet) the distance for one pore flush is approximately 20 feet if a semi-soluble carbon donor is used. Based on the estimated groundwater velocity and a sorption retardation factor of 1.5, approximately 4.4 years would be needed to achieve 11.2 pore flushes. Using a soluble donor, given the additional organic carbon under advective groundwater flow, it is assumed that the number of transects could be reduced from four to two transects spaced 60 ft apart. In this scenario, the pore flushing distance with only two transects relies on a flushing distance of approximately 16 feet and requires approximately 1.5 years to achieve the total estimated 11.2 pore flushes. Similar to assumptions for OU 1, these timeframes do not account for the lag time required to establish geochemical conditions and facilitate complete dechlorination within each barrier, so an additional six months was added to the projected timeframe. The resulting operational demands associated with each carbon donor would result in approximately 2 injections of semi-soluble donor (one injection every 3 years) and 6 injections of soluble donor (one injection every 4 months).

The upper clay unit present in this area likely requires the use of DPT injection rather than permanent wells. DPT injections would rely on smaller ROIs and higher injection pressures in order to distribute the injection solution in the upper clay. Because DPT is recommended rather than permanent infrastructure, the use of a long-lasting carbon source is preferable in the upper clay. Based on the proposed transect spacing and the objective to reduce remedial timeframes across the site, a semi-soluble carbon donor injected using DPT is recommended. While this recommendation increases the remedial timeframe by approximately 3 years, the difference in remedial timeframe between the two options is not significant and

9/14

DSCR OU 6 Remedy Optimization the use of a semi-soluble carbon donor provides the desired treatment, a reasonable remedial timeframe, and overall cost efficiency.

Injection volumes and Frequency for OU 1 Area

Approximately 28 UWBU injection wells are included in the OU 1 Area conceptual injection plan (Figure

2). This includes 20 new proposed injection wells and 8 existing injection wells. Injection well spacing of 30 to 40 feet apart resulting in a targeted 15 to 20 feet of ROI is recommended. Final injection well spacing should be determined based on boring observations and CVOC mass distribution observed during the pre-design investigation, and estimated injection volumes will be adjusted based on the final well network spacing and lithology. Ultimate injection completion will be determined in the field based on confirmed breakthrough response at proximal dose-response monitoring wells. For the purposes of this tech memo a target ROI of 15 feet has been assumed to calculate per well and total injection volume. The target injection volume, Vinjection, necessary to achieve full reagent breakthrough at the targeted injection radius of

15 feet is estimated using the following equation, based on the volume of a cylinder:

Vinjection = z*π*r2*Nm

Where: Vinjection = Required injection volume z = Saturated injection interval accessed by injection screen (15 feet) r = Target radius of influence (15 feet)

Nm= Mobile porosity of soils (12-15 percent [%])

The 12-15% mobile porosity range for the injection is based on site soil conditions described during previous site activities and should be confirmed during the PDI. The estimated volume of mixed carbon solution is 9,500-11,900 gallons for per well. Using the assumptions stated above, an estimated total of approximately 266,000–333,200 gallons of 2% soluble carbon donor (e.g., molasses or lactate) should be injected into the OU 1 Area injection wells (proposed and existing) during each injection event. Final target injection volumes will be modified based on the condition of individual wells, soil characteristics, and final spacing and vertical interval of the injection wells.

Due to the presence of EVO in existing monitoring wells within the OU 1 Area source, current groundwater concentrations are unknown; however, remedial timeframes have been estimated based on historic concentrations. Based on historic concentrations, it is anticipated that 6 injection events conducted every 4 months would likely lead to an MNA end point equating to a timeframe of approximately 2 years to reach

MNA. Timeframe estimates can be refined following the pre-design sampling and used to identify the ideal balance between remedial time, injection infrastructure needs, injection volumes, and overall active ISB life-cycle cost.

Injection volumes and Frequency for OU 2 Area

The conceptual injection plan in OU 2 Area consists of injections in the UWBU and the LWBU. Injections in the UWBU will focus on source area remediation in the OU 2 source and be completed using DPT injection transects. The LWBU injection will focus on historical impacts off-site (Figure 3). LWBU monitoring wells off-site contain EVO and have not been sampled to confirm off-site remediation is warranted. Groundwater concentrations should be confirmed prior to implementing the LWBU injection program.

10/14

DSCR OU 6 Remedy Optimization

UWBU injections should be completed using DPT transects. Due to the complicated aquifer matrix, injection volumes will vary for each depth interval as the targeted injection volumes will be reduced in the upper clay relative to the underlying sand and gravel layer. The target ROI of each DPT borings is 5 feet to calculate per point and total injection volume targets. Based on historical cross sections, the depth of the upper clay layer in the OU 2 source area is approximately 8 feet thick and the sand and gravel layer is approximately 4 feet thick. The target injection volume, Vinjection, necessary to achieve full reagent breakthrough at the targeted injection radius of 5 feet is estimated in each interval using the following equation, based on the volume of a cylinder:

Vinjection = z*π*r2*Nm

Where: Vinjection = Required injection volume z = Saturated injection interval accessed by injection screen (8 feet upper clay, 4 feet sand and gravel) r = Target radius of influence (5 feet)

Nm= Mobile porosity of soils (8-10% upper clay, 12-15% sand and gravel)

The mobile porosity range is based on site soil conditions described during previous site activities and should be confirmed during the PDI. The estimated volume of solution per 4-foot section is 190 - 230 gallons in the upper clay and 305 - 350 gallons in the sand and gravel. It is estimated there are two 4-foot sections in the upper clay and one 4-foot section in the sand and gravel for each boring. Therefore, the total injection volume is 685 - 810 gallons for per DPT boring. Each transect is estimated to be 130 feet long resulting in 26 borings per transect. This results in a total of approximately 71,240 – 84,240 gallons of

2% semi-soluble carbon donor (e.g., Lactoil®, ABC) by weight injected into the OU 2 source area transects identified on Figure 3. Final injections volumes may be modified based on conditions of wells, aquifer characteristics, and final distances between injection locations.

There are 10 existing LWBU injection wells on-site near the property boundary. These existing LWBU injection wells should be considered as contingency injection wells pending the results of comprehensive

LWBU sampling once the EVO present in monitoring wells has been addressed through redevelopment

(see above). Target injection volume estimates for these wells should be calculated as described above if included in the final optimized injection plan. If groundwater data supports inclusion of these wells in the injection program, each well should be redeveloped and evaluated for injectability prior to inclusion as part of the injection program.

Due to the presence of EVO in existing monitoring wells within the UWBU and LWBU of the OU 2 Area, current groundwater concentrations are unknown and remedial timeframes have not been estimated.

Within the OU 2 UWBU area, it is anticipated that two injections are necessary to remediate the source impacts. Each injection in the UWBU would be approximately 3 years apart. Final timeframe determinations and confirmation of the proposed delivery network will be completed following the collection of PDI and updated groundwater data.

Well Construction, System Set-Up and Operation

Injection Wells

Should be constructed of 4-inch inside diameter (ID) schedule 80 polyvinyl chloride (PVC) casing and 4-inch ID stainless steel screen.

11/14

DSCR OU 6 Remedy Optimization

The screens should consist of Johnson wire wrap stainless steel 20-slot screen.

UWBU wells should be screened from the water table (between 13 and 25 feet below ground surface [bgs]) down to the lower clay layer (approximately 30 feet ft bgs). Each well should be completed with an approximate screen interval of 10 to 20 feet.

Supply Wells

Should be constructed of 4-inch ID schedule 80 PVC casing and 4-inch ID stainless steel screen.

Should have a five-foot schedule 80 PVC tail pipe (sump).

Like the injection wells, the screens should consist of Johnson wire wrap stainless steel 20-slot screen.

Will maximize well capacity via placement of the well screen within the sand and gravel layer, with an approximate screen interval of 10 to 20 feet (approximately 30 feet bgs total depth).

System Set-Up and Operation (OU 1 Area)

Submersible pumps (Grundfos Redi-Flo 5E8 or similar), capable of extracting at approximately 1 to 3 gallons per minute (gpm) should be used.

The pumps should be placed at the bottom of the screen to increase the overall available drawdown.

Submersible pumps and injection pumps will be powered by rented generators.

All equipment hoses will be run above grade.

Supply wells will only be operated immediately prior to injection events, as discussed above, and shut down once injections are complete. Any water remaining in the frac tanks at the end of an injection would be injected.

System Set-Up and Operation (OU 2 Area)

If the groundwater extraction and reinjection system is installed concurrent with injections, conveyance pipe could be installed to deliver amended water to each injection area.

If injections are completed prior to installation of the recommended groundwater extraction and reinjection system, potable water should either be delivered to the site via flatbed truck or taken from a local fire hydrant (if available) and stored in frac tanks.

Performance Monitoring

To support injection timing considerations performance monitoring of existing wells near the treatment areas at OU 1 and OU 2 will be done periodically to monitor the effectiveness of the ISB process, track changes in the geochemical parameters, and evaluate total organic carbon (TOC) longevity over time.

The sampling frequency will be quarterly for two years and semi-annual thereafter until 2 years following the final injection in each area. Note that the performance monitoring discussed here is inclusive of the property line monitoring associated with the hydraulic containment system proposed for OU 2 area discussed below. Additional diffuse plume monitoring may be added following final design of the

12/14

DSCR OU 6 Remedy Optimization recirculation system. Annual MNA/LTM is also required in accordance with the current approved sampling work plan.

Select performance monitoring wells at OU 1 and OU 2 Areas (Table 1 and Figure 4) will be sampled for the following parameters:

o Field parameters, including dissolved oxygen (DO), pH, oxidation/reduction potential

(ORP) and turbidity;

o CVOCs using United States Environmental Protection Agency (USEPA) Method 8260B;

o Dissolved gases (methane, ethene, and ethane); and o TOC o Anions (sulfate)

Numerous wells are ¾-inch wells which should be abandoned and reinstalled as part of the well rehabilitation effort completed during the PDI or prior to initiating injections such that baseline data can be collected from each location.

Following initial injection, evaluation of the performance monitoring data will be focused on confirming TOC distribution and ensuring TOC is sustained above approximately 10 milligrams per liter (mg/L) in select performance monitoring wells (Table 2).

Based on the TOC sample results, the injection frequency may be adjusted (either up or down) or the TOC loading modified to maximize electron donor longevity between injection events.

As the active bioremediation system entails enhanced biodegradation, it is anticipated that shifts in daughter products (e.g., trichloroethene to cis-1,2-dichloroethene and vinyl chloride) will be observed over the course of the program. CVOC concentrations monitored within the treatment area will be compared to background trends, and these trends will also consider shifts in daughter products (which may increase initially) as part of holistic CVOC declines throughout the plume.

Based on the results of the data, ongoing refinement of the remediation program should be conducted and may include modification of the carbon substrate, pH modification, bioaugmentation, or carbon dosing and frequency of delivery events.

Table 2. Performance Monitoring Wells

Well Classification OU 1 Area OU 2 Area UWBU

Wells

OU 2 Area LWBU

Wells

Performance

Monitoring Well ID

MW-227c1 MW-231c MW-236c

MW-132 MW-691 AEHA-31B

MW-228c1 MW-601 MW-94

MW-131 MW-611 MW-147

MW-128 MW-66 MW-148

MW-1291 MW-135 MW-146

MW-130 MW-65 --

MW-127 EBF-01S --

13/14

DSCR OU 6 Remedy Optimization

Well Classification OU 1 Area OU 2 Area UWBU

Wells

OU 2 Area LWBU

Wells

MW-126 AEHA-30A2 --

MW-1232 AEHA-31A2 --

MW-1242 -- --

MW-222c2 -- -- Notes:

All wells will be sampled for general field parameters (DO, pH, ORP and turbidity) and CVOCs

Highlighted wells are ¾” wells for abandonment and replacement 1 Evaluated for presence and utilization of TOC and used to make injection frequency determinations 2Wells that will be sampled for TOC and dissolved gasses (methane, ethene and ethane); all others will not be sampled for these constituents

Hydraulic Containment - OU 2 Area UWBU and LWBU

Concurrent with or after the optimized injection remedy approach at OU 2 Area is initiated, a hydraulic containment remedy is recommended for the installation boundary and off-site LWBU impacts. This will include 9 downgradient extraction (and injection water supply) wells that will operate continuously. The groundwater system will have the following design elements:

The extraction wells should be screened across the UWBU between approximately 15 and 35 feet bgs. Extraction well screens may be adjusted based on geologic observations (e.g., won’t be screened across the upper clay)

The extraction wells should be screened across the LWBU to approximately 55 ft bgs with 30 ft of screen. Extraction well screens may be adjusted based on geologic observations (e.g., won’t be screened across the lower clay)

Extraction wells will be constructed of at least 6-inch ID schedule 80 PVC casing and 6-inch ID stainless steel screen.

If recirculation of treated groundwater is implemented, re-injection of the groundwater will be balanced, with the total injection rate of approximately 20 gpm equal to the total extraction rate of approximately 20 gpm. If a balanced system cannot be achieved, excess treated water meeting

Virginia Pollution Discharge Elimination Standard Permit requirements will be discharged to the on-site storm water sewer.

All conveyance lines will be installed below grade to and from the treatment system

Extracted groundwater will be treated through an enclosed ex-situ treatment system using granular activated carbon and monitored weekly to perform system maintenance, assess system performance and identify opportunities to optimize system operation.

The system will be designed such that treated groundwater can be periodically amended with an organic carbon source (specific to the target injection wells) and sent directly to injection wells or the DPT injection area. When carbon injections are not ongoing, extracted water will be sent to the clean water injection wells.

Performance monitoring of existing wells within the recirculation cell will be done periodically to monitor the effectiveness of the recirculation cell and to monitor changes in geochemical and

14/14

DSCR OU 6 Remedy Optimization

CVOC concentrations over time. The wells used for this performance monitoring will be determined following final placement of the injection wells. However, this likely includes sampling of 5-8 monitoring wells for CVOCs using USEPA Method 8260B.

Service Layer Credits: Source: Esri, DigitalGlobe, GeoEye, Earthstar Geographics, CNES/Airbus DS, USDA, USGS, AeroGRID, IGN, and the GIS User Community

@A

@A

@A

@A

@A

!A

@A@A

@A@A

@A

@A@A

@A @A

@A@A

@A@A @A@A

@A @A

@A@A

@A @A

@A@A

@A

@A @A

@A@A

@A @A

@A@A@A @A

@A

@A

@A

@A

@A@A

@A@A@A@A@A@A@A

@A@A!A

@A @A

@A

@A

@A @A

@A !A

@A

@A

@A

!A

@A

@A

@A

@A

!A

@A

@A

@A@A @A @A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A @A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A@A

@A

@A

@A

@A

@A @A

@A

@A

@A @A

@A

@A

@A @A

@A @A

@A

@A @A @A

@A @A

@A @A

@A @A

@A

@A @A

@A

@A

@A

@A @A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A@A@A @A@A @A

@A@A @A

@A @A @A@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A @A

@A

@A

@A @A@A

@A

@A

@A @A

@A

@A

@A @A

@A

@A

@A

@A

@A

@A @A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

@A

!A

!A

@A @A@A@A

@A @A@A @A@A

@A@A@A@A

@A@A @A@A

@A@A @A@A

@A @A@A @A

@A@A@A@A

@A@A@A@A

@A@A@A@A

@A

@A

@A @A

@A

@A

@A

@A @A

@A

@A

@A@A

@A

@A

A R d

G Rd

E R d

1st St

2nd St

Da leb roo k D r

De fen se G en era l

Alcott Rd

Libwood Ave

Key Ave

Fonda St

Se na te St

Cherry Park Pl

E R d

Key Ave

NO NAME CREEK

OU 1

OU 3

OU 2

OU 9

CSMMW9C

PZ9B

PZ8B

MW96

MW95

MW94

MW93

MW92

MW43 MW42

MW41

MW35 MW34

MW33

MW236

MW181

MW180

MW171

MW170MW168

MW167

MW166

MW165

MW160

MW159

MW157

MW154

MW153

MW151

MW148 MW147

MW146

MW145

MW142

MW141

MW138

DMW9B

DMW7B

ANGA4

ANGA3

ANGA2

ANGA1

INJ210

INJ212

INJ213

INJ215

INJ199

INJ197

INJ196

USGSP3 USGSP2

USGSO3

USGSO2

USGSN3 USGSN2

USGSM3 USGSM2

USGSM1

USGSL3 USGSL2

USGSL1

USGSK3

USGSK2

USGSI3 USGSI2

USGSE5

USGSE3

USGSE2

USGSD3

USGSD2

LAWMWQ

EBF08D

DMW34B

DMW28B

DMW16B

DMW12B

DMW11B

DMW10B

CMT127

CSMMW9B

AEHA34B

AEHA33B

AEHA32B

AEHA31B

AEHA29B

AEHA28B

AEHA26B

AEHA25B

AEHA24B

AEHA23B

AEHA22B

AEHA21B

AEHA19B

AEHA18B

PZ5

PZ4

PZ3

PZ2PZ1

PZ7A

PZ6A

MW91

MW90

MW89

MW88

MW87

MW86

MW85

MW84

MW83

MW82

MW81

MW80

MW79

MW78

MW77

MW76

MW74

MW73

MW72

MW71

MW69MW68

MW67

MW66

MW65

MW64

MW63

MW62MW59 MW58

MW57

MW56 MW55

MW54

MW53

MW51

MW50

MW49

MW48

MW46

MW45

MW234

MW235

MW225

MW231

MW229

MW230

MW227

MW228

MW223

MW222

MW182

MW179

MW178

MW169

MW158

MW156

MW155

MW150

MW149

MW144

MW140

MW139

MW135

MW134

MW133

MW132

MW131

MW130

MW129

MW128

MW127

MW126

MW125

MW124

MW123

MW121

MW120

MW119

MW118

MW117

MW116

MW115

MW114

MW113

MW112

DMW9A

DMW8A

DMW7A

AEHA5

AEHA4

AEHA3

AEHA2

AEHA1

INJ204

INJ205

INJ206INJ241

INJ237

INJ193

INJ191

INJ190

INJ189

INJ188

INJ187

INJ200 INJ183

INJ184

INJ185

INJ186

USGSO4

USGSM4

USGSL4

USGSK4

INJ173

INJ172

MWNGA5

MWNGA3

MWNGA2

DMW18A

DMW17A

DMW12A

DMW11A

CSMPZ8

CSMPZ7

CSMPZ6

CSMPZ5CSMPZ4CSMPZ3

B150MW2

B150MW1

AEHA34A AEHA33A

AEHA32A

AEHA29A

AEHA26A

AEHA25A

AEHA22A

AEHA21A

AEHA19A

AEHA18A

AEHA16A

AEHA15A

AEHA12A

MWA50-38

MWA50-36

MWA50-35

DMW8E

DMW28E

DMW16C

USGSBR2

USGSBR1

AEHA16C

MW1

PZ9B

DMW61

Bldg. 54

Warehouse 11 Warehouse 14 Warehouse 15

Warehouse 59Warehouse 60

Bldg. 40

Bldg. 151

Bldg. 150

Bldg. 153

T-124

Bldg. 154 Bldg. 155T-123

Bldg. 56

Bldg. 41

Bldg. 149

Bldg. 85

Bldg. 522

Shed

T-125

MW217

MW216

INJ44

INJ32

INJ211

INJ214

INJ198

USGSP1

USGSO1

USGSN1

USGSK1

USGSI1

USGSE1

USGSD1

MWNGA9

MWNGA8

MWNGA7

MWNGA6

EBF04D

EBF02D

CMT126CMT125CMT124CMT123CMT121

MWNGA10

AEHA30B

AEHA27B

MW75

MW70

MW61MW60

MW52

MW47

MW39MW38MW37

MW31MW30MW29

MW224

MW226

MW152

MW143

INJ40

INJ28

INJ207

INJ208

INJ209

INJ240

INJ239

INJ238

INJ192

INJ194

INJ195

INJ201

INJ202

INJ203

USGSP4

USGSN4

USGSI4

USGSE4

USGSD4

INJ177

INJ176

INJ175

INJ174

MWNGA4MWNGA1

EBF03S

EBF01S

DMW10A CMT122

CSMMW9A

AEHA31A

AEHA30A

AEHA28A

AEHA27A

AEHA24A

AEHA23A

AEHA14A

MWA50-37

PZ8B

ANGA5

MWNGA12

11787000

11787000

11788000

11788000

11789000

11789000

Legend U.S. Government Property

OU-6

Other Operable Unit (part of OU-6) OU-9/OU-6 Pump & Treat (1996-2006)

OU-9/OU-6 Pump & Treat Extraction Well (1996-2006) OU-6 DPE Pilot Test Cross-Section (1999) Approx. Bioremediation Area

Monitoring Wells @? Upper WBU Injection Well @A Upper WBU Monitoring Well @A Lower WBU Monitoring Well !A Saprolite Monitoring Well !A Bedrock Monitoring Well

Monitoring Well WBU Unknown &, Upper WBU Piezometer &, Lower WBU Piezometer &, Bedrock Piezometer

TCE Plumes (UWBU) (Spring 2017) * 5 µg/L 50 µg/L

Figure 1 OU6 Remedial Technologies and Pilot Tests

Operable Unit 6 Defense Supply Center Richmond

Richmond, VA

Data Sources: AECOM, GIS Data;

ESRI, ArcGIS Online, Aerial Imagery Coordinate System: VA State Plane, South Datum: NAD 83 Units: Feet

0 100 200

Feet

³

OU-6 Pump & Treat (1996 - 2006) Consisted of groundwater extraction using a network of 22 extraction wells, CVOC treatment and discharge system.

OU-6 Dual Phase Extraction (DPE) Pilot Test (1999) Result did not show a favorable outcome and was not chosen as a remedial approach OU-6 Enhanced In-Situ Biodegradation Treatability Tests (2007) Consisted of injection of edible oil emulsion in the Upper and Lower WBU to enhance natural degradation (attenuation) of the chlorinated solvent contamination. Based on the favorable results of these tests, in-situ bioremediation with EVO injection was adapted as part of the final remedy at OU6.

OU-6 MNA and LTM, ICs and Air Sampling (2013 - present) A ROD for OU6 was signed in September 2013. The final remedy of OU6 consisted of in-situ bioremediation, MNA and Long Term Monitoring, Institutional Controls and air sampling.

OU-6 In-Situ Bioremediation (2015 - present) The in-situ bioremediation component of the final remedy included in the OU-6 ROD was implemented in July/August of 2015. The in-situ bioremediation consisted of EVO injections in source areas and downgradient portions of the plume where COC concentrations were highest.

* Dashed portions of plumes are based on 2015 results and will be evaluated pending post-remedial monitoring results

Bedrock Well ID Saprolite Well ID Lower WBU Well/PZ ID Upper WBU Well/PZ ID Unknown WBU Well/PZ ID

CSMMW-9C

DMW-8E

MW171

DMW-61

MW123

Service Layer Credits: Source: Esri, DigitalGlobe, GeoEye, Earthstar Geographics, CNES/Airbus DS, USDA, USGS, AeroGRID, IGN, and the GIS User Community

Data Sources: AECOM, GIS Data;

ESRI, ArcGIS Online, Aerial Imagery Coordinate System: VA State Plane, South Datum: NAD 83 Units: Feet

0 100 200

Feet

Bldg. 85

Bldg. 54

Warehouse 59

Warehouse 60

2nd St

De fen se G en era l

E R d

11787000

11787000

11787200

11787200

11787400

11787400

11787600

11787600

11787800

11787800

11788000

11788000

11788200

11788200

11788400

11788400

Legend U.S. Government Property

OU-6

Approx. Bioremediation Area

Proposed Remedy Optimization Wells !? Upper WBU Injection Well !? Upper WBU Water Supply Well

Other Wells @? Upper WBU Injection Well

Estimated Extent of Upper WBU Plume (Spring 2017) cis-1,2-DCE - 70 µg/L PCE - 5 µg/L TCE - 5 µg/L Vinyl Chloride - 2 µg/L

Figure 2 OU6 Proposed Remedy

Optimization - AOC 1

Operable Unit 6 Defense Supply Center Richmond

Richmond, VA

³

Bldg. 150

Bldg. 153

T-125

Bldg. 154 Bldg. 155T-123

Bldg.

Bldg. 151

Bldg. 54

NO NAME CREEK

Alcott Rd Fonda St

E R d

G Rd

11788200

11788200

11788400

11788400

11788600

11788600

11788800

11788800

11789000

11789000

11789200

11789200

11789400

11789400

Legend U.S. Government Property

OU 6

Approx. Bioremediation Area Proposed Optimization DPT Transect

Proposed Remedy Optimization Wells !? Lower WBU Extraction Well !? Upper WBU Extraction Well

Contingent Wells @? Existing Lower WBU Injection Well

Estimated Extent of Upper WBU Plume (Spring 2017) cis-1,2-DCE - 70 µg/L PCE - 5 µg/L TCE - 5 µg/L Vinyl Chloride - 2 µg/L

Estimated Extent of Lower WBU Plume (Spring 2017) cis-1,2-DCE - 70 µg/L PCE - 5 µg/L TCE - 5 µg/L VC - 2 µg/L

Figure 3 OU 6 Proposed Remedy

Optimization - AOC 2

Operable Unit 6 Defense Supply Center Richmond

Richmond, VA

Data Sources: AECOM, GIS Data;

ESRI, ArcGIS Online, Aerial Imagery Coordinate System: VA State Plane, South Datum: NAD 83 Units: Feet

0 100 200

Feet

³

Service Layer Credits: Source: Esri, DigitalGlobe, GeoEye, Earthstar Geographics, CNES/Airbus DS, USDA, USGS, AeroGRID, IGN, and the GIS User Community

Data Sources: AECOM, GIS Data;

ESRI, ArcGIS Online, Aerial Imagery Coordinate System: VA State Plane, South Datum: NAD 83 Units: Feet

0 100 200

Feet

@A

@A@A

@A

@A

@A

@A@A

@A

@A

@A

@A @A

@A

@A

@A @A

@A@A

@A

@A

@A

@A

@A

@A @A

@A @A

MW94

AEHA31B

MW146

MW147

MW148

MW236

AEHA31A

MW127

MW129

MW132

MW61

MW66

MW228

AEHA30A

EBF01S

MW123

MW124

MW126

MW128

MW130

MW131

MW135

MW60

MW65

MW69

MW222

MW227

MW231

DMW28E

Bldg. 56

Bldg. 150

Bldg.

T-125 Bldg. 154Bldg. 155T-123

T-124

Bldg. 85

Bldg.

Bldg. 151

Bldg. 41

Bldg.

Warehouse 15Warehouse 11 Warehouse 14

Bldg. 54

Warehouse 59Warehouse 60

NO NAME CREEK

Alcott Rd

Se na te St

De fen se G en era l

1st St

G RdE R d

Fonda St

2nd St

11787200

11787200

11787400

11787400

11787600

11787600

11787800

11787800

11788000

11788000

11788200

11788200

11788400

11788400

11788600

11788600

11788800

11788800

11789000

11789000

11789200

11789200

Legend U.S. Government Property

OU 6

Approx. Bioremediation Area Proposed Optimization DPT Transect

Proposed Performance Monitoring Wells @A Lower WBU Monitoring Well @A Upper WBU Monitoring Well @? Saprolite Monitoring Well

Proposed Remedy Optimization Wells !? Lower WBU Extraction Well !? Upper WBU Injection Well !? Upper WBU Extraction Well

Contingent Wells @? Existing Lower WBU Injection Well @? Existing Upper WBU Injection Well

Figure 4 Proposed Performance

Monitoring Network

Operable Unit 6 Defense Supply Center Richmond

Richmond, VA

³

Saprolite Well ID Lower WBU Well ID Upper WBU Well ID

DMW-28E

MW148

MW135