December 6, 2002

Dear Mr. Alcamo:

As you requested, Tetra Tech EM Inc. (Tetra Tech) has prepared the enclosed current status report for the above-referenced site in Monroe County, Indiana. The report incorporates your November 14, 2002, comments on Tetra Tech's draft current status report, which was submitted to you on October 21, 2002. The report summarizes available hydrogeologic and chemical data for the site. Tetra Tech reviewed U.S. Environmental Protection Agency (EPA) Region 5 files and Viacom, Inc., files to obtain this data. Viacom, Inc., is the responsible party for the site.

If you have any questions about the enclosed report, please call me at (312) 946-6491.

Sincerely,

Jeffrey J. Lifka

This status report has been prepared for the U.S. Environmental Protection Agency (EPA) in partial fulfillment of the statement of work for Response Action Contract No. 68-W6-0037, Work Assignment No. 945-TATA-05ZZ. Tetra Tech EM Inc. (Tetra Tech) prepared this report to summarize the current status of hydrogeologic and chemical data for the Bennett's Dump site in Bloomington, Monroe County, Indiana.

Bennett's Dump was placed on the National Priorities List in September 1984 because of polychlorinated biphenyl (PCB) contamination at the site. In August 1985, EPA, the State of Indiana, Monroe County, the City of Bloomington, and Westinghouse Electric Corporation (Westinghouse) (later known as CBS Corporation [CBS] and now known as Viacom, Inc. [Viacom]) entered into a consent decree (CD). Under the CD, Westinghouse was to remediate six PCB-contaminated sites in the Bloomington area, including Bennett's Dump. Initially, Westinghouse agreed to remove and incinerate Bennett's Dump waste and monitor groundwater quality around the site. Remedial actions (RA) other than incineration had to be considered, however, when the Indiana State Legislature passed several laws intended to delay and block the implementation of the incineration remedy.

Tetra Tech reviewed available site data obtained from EPA Region 5 as well as data that became available to Tetra Tech from other parties to the CD for the site. The earliest data is from 1983 and data through June 2002 is included in this report. Geologic, hydrogeologic, and analytical data is being collected at the site on an ongoing basis and will be included in future status reports prepared for the site. This status report describes the current understanding of groundwater flow, PCB occurrence, and PCB migration based on site investigations conducted to date.

This status report consists of six sections, including this introduction. Section 2.0 contains site background information, Section 3.0 summarizes the site hydrogeologic data, Section 4.0 summarizes the site chemical data, Section 5.0 summarizes Tetra Tech's overall findings, and Section 6.0 summarizes future site activities. References used in preparing this report are listed at the end of the text. Figures and tables cited in the text appear in Appendix A and B respectively.

SITE HISTORY

Bennett's Dump is located about 2.5 miles northwest of Bloomington in Monroe County, Indiana (see Figure A-1). The main fill area occupies 3.5 acres, and another 0.5-acre fill area lies east of and adjacent to the main fill area. In addition, a small, satellite fill area measuring about 30 by 60 feet lies 750 feet north of the main fill area. The site, which is no longer active, lies in a former limestone quarry area called Bennett's Stone Quarry. The site is bordered by Stout's Creek and railroad tracks to the west, quarry access roads to the south and east, and a gravel road to the north. The site topography is characterized by numerous excavation pits remaining from former quarry operations. The main fill area lies on a westward-sloping hillside east of Stout's Creek (Viacom 2002a). Figure A-2 shows significant site features, including streams, springs, monitoring wells, and other sampling locations.

When the site was active in the 1960s, part of the quarry was used for landfilling industrial wastes, including electrical parts. During this period, landfilling operations were conducted without modern landfill control measures such as liners, daily cover, and leachate collection. Most of the electrical parts visible at the site had been crushed, burned, or otherwise torn open. Insulator wrapping papers, ceramic bushings, and other electrical parts littered the site. Westinghouse electrical capacitors, some of which contained PCBs, were among the electrical parts disposed of at the site.

EPA undertook an interim remedial measure (IRM) at the site in June 1983. The IRM consisted of removing 252 capacitors and 20 cubic yards of stained soils; isolating and containing the waste material with a 2-foot-thick, clay cap and topsoil cover; and securing the site (CBS 2000).

Site soil sampling was conducted for dioxin and furan analysis by Ecology and Environment, Inc. (E&E), in 1984. One background surface sample, two on-site surface samples, and one on-site subsurface sample were collected. Dioxins were detected in each of the four samples. Cumulative concentrations of the dioxin isomers analyzed for (tetra, penta, hexa, hepta, and octa) ranged from 0.3 parts per million (ppm) in the background sample to 3.6 ppm in the on-site subsurface sample (E&E 1986). Furans were not detected in any of the samples.

In 1985, a CD was signed by Westinghouse, the City of Bloomington, the Indiana Department of Environmental Management (IDEM), the Monroe County Health Department (MCHD), and EPA (the CD parties). The site remedy included removal of PCB-contaminated materials, long-term Groundwater monitoring, sediment removal, and sediment monitoring in Stout's Creek on the western boundary of the site. In late 1987 and 1988, Westinghouse installed an additional cap at the edge of the main fill area and removed contaminated sediment along 1,600 linear feet of Stout's Creek adjacent to the site (CBS 2000).

Soil, sediment, surface water, groundwater, and fish sampling was intermittently performed at the site from 1983 through 1998. Sample analytical results indicated that an RA was necessary. A record of decision (ROD) amendment for the Bennett's Dump site was signed on October 16, 1998. The ROD amendment allowed for a remedial alternative that did not include construction and use of a Toxic Substances Control Act (TSCA)-approved, dedicated, municipal solid waste-fired incinerator, which was called for in the CD but was subsequently blocked by the Indiana State Legislature. The objective of the site cleanup was to meet low-occupancy or industrial standards for possible site reuse. Based on groundwater monitoring data collected prior to site remediation, the expectation was that main source removal would reduce and subsequently eliminate the discharge of PCBs from the springs on the site. In addition, the continued release of PCBs into Stout's Creek since the interim sediment cleanup in 1988 indicated that additional sediment cleanup was required in Stout's Creek.

The major components of the site remedy selected in the ROD amendment include the following:

• Excavation and off-site disposal in a permitted, commercial, chemical waste/TSCA landfill of approximately 55,000 cubic yards of contaminated soils containing PCBs exceeding 25 ppm on average
• Excavation and off-site incineration in a permitted, commercial incinerator of capacitors containing PCB oil
• Long-term Groundwater monitoring and implementation of deed restrictions

From December 1998 through June 1999, CBS completed a geophysical investigation and delineation drilling to prepare an RA work plan for the site. The work plan was initially submitted to EPA and other governmental parties on June 25, 1999. The work plan was approved by EPA in August 1999 (CBS 1999), and remedial work was completed at the site by November 1999 (CBS 2000).

• H-7 Center (1,100 ppm) and H-7 Northeast (170 ppm)
• F-6 Southwest (51 ppm)
• G-8 Northeast (67 ppm)

The locations of these grids are shown in Figure A-3. The CD parties allowed Viacom to terminate excavation in each grid after removing as much material as possible and collecting a final bottom grab sample. The average PCB concentration in the eight subgrids composing grid H-7 was 122 ppm, in the five subgrids composing grid F-6 was 25.3 ppm, and in the five subgrids composing grid G-8 was 33 ppm. The satellite fill area was excavated in a similar manner to meet the maximum and average PCB concentration remediation goals. The RA is documented in a CBS report dated February 7, 2000 (CBS 2000). During the RA, Tetra Tech collected nonsplit samples at the site for dioxin and furan analysis. The dioxin and furan concentrations in these samples were within acceptable ranges (Tetra Tech 2000).

In May 2000, CBS removed soil with PCB concentrations greater than 2 ppm from the east bank of Stout's Creek. Soil was removed from three discrete, 50-foot-long segments along the bank. A total width of 6 feet was excavated along each segment, and a minimum of 1 inch of surface soil was removed. The soil was consolidated in grid G-8 and capped with 1 foot of clean, clay backfill (Viacom 2002a).

In June 2000, CBS collected soil samples from the east bank of Stout's Creek adjacent to Bennett's Dump. PCB concentrations in the soil samples ranged from 0.28 to 2.7 ppm. Soil with PCB concentrations greater than 2 ppm was excavated as part of a small removal action in September 2000 (Viacom 2002a).

In April 2002, EPA approved Viacom's long-term groundwater monitoring plan for evaluation of PCB concentrations in site groundwater, surface water, and springs (Viacom 2002a). Activities conducted under this plan include the following:

• Quarterly sampling of Middle Spring and Mound Spring
• Sampling of other flowing, on-site springs as appropriate
• Quarterly sampling at upstream and downstream locations in Stout's Creek under low- flow conditions
• Annual, high-flow condition sampling at upstream and downstream locations in Stout's Creek, Middle Spring, Mound Spring, and other flowing springs during the receding limb of a major storm event
• Fish and sediment sampling in Stout's Creek in the year 2003
• Semiannual sampling of monitoring wells MW-6I and MW-6D
• Purging PCB oil from monitoring well MW-5 and making quarterly observations to determine whether PCB oil reappears in the well

EPA also approved Viacom's groundwater investigation plan (Viacom 2002b) for the ongoing evaluation of the hydrologic flow system at the site. Activities conducted under this plan include measuring

• Groundwater elevations
• Water elevations in adjacent quarries
• Spring discharge rates
• Flow rates in Stout's Creek

The long-term Groundwater monitoring plan and the groundwater investigation plan are discussed in detail in Sections 3.0 and 4.1, respectively.

In August 2002, EPA completed a 5-year review of the Bennett's Dump site (EPA 2002). The first 5-year review was not scheduled to be completed until September 2005; however, the schedule was accelerated as it became apparent that the source removal did not diminish the quantities of PCBs in water released from on-site springs that flows directly into Stout's Creek. The following key points are made in the EPA's 5-year review report:

• Site issues exist with regard to erosion of soil cover, poor site drainage, and PCBs being released from site springs into Stout's Creek.
• In response to these issues, soil cover was repaired, liners placed over areas where surface water travels, surface runoff was channeled around the site, and Viacom implemented a groundwater investigation plan to determine whether the ROD amendment will require modification.
• The continuing release of PCBs from the on-site springs into Stout's Creek requires further investigation. A groundwater investigation is underway to obtain the necessary data to evaluate the continuing release of PCBs.
• The long-term protectiveness of the RA will be verified once the groundwater investigation plan is executed and a determination is made regarding modification of the ROD amendment.
• Implementation of deed restrictions has been delayed because of the groundwater investigation underway. The deed restrictions will be reevaluated once a determination is made regarding modification of the ROD amendment.

EPA intends to issue its next 5-year review report for the site no later than June 2007.

2.2 SITE GEOLOGY AND HYDROGEOLOGY

The site and the immediate surrounding area have been extensively disturbed by anthropogenic activity, including dimension stone quarrying, backfilling of quarry pits and other low-lying areas, and grading. A layer of unconsolidated material overlies the bedrock at the site. The unconsolidated material consists of fill material and indigenous soil. The fill material ranges in thickness from 4 to l l feet and consists of black to brown clay, sand and gravel, and rubble (Blasland & Bouck Engineers, P.C. [B&B] 1987). The area in the vicinity of the site contains large, surface and subsurface limestone quarry blocks.

The surficial bedrock units at the site belong to the Sanders Group (Mississippian). These rocks are assigned to (in descending order), the Salem and Harrodsburg limestones and the Ramp Creek Formation. The Salem limestone is the surficial bedrock unit and is the formation that was quarried for building stone in the area. The principal Salem Ethology is massively bedded biosparite. The unit is extensively quarried in the Bloomington and nearby Bedford areas, and its stone is marketed nationwide as "Indiana Limestone." The massive bedding character and homogeneous texture of the unit make it ideal for dimension stone quarrying. The Harrodsburg limestone and Ramp Creek Formation of the Lower Sanders Group underlie the Salem limestone and outcrop at lower elevations near the site (B&B 1987). The Salem limestone ranges in thickness from 90 to 100 feet at the site. A thin shale bed is present at the base of the Salem limestone. Figure A-4 is a generalized geologic cross section trending west to east across the approximate center of the site.

Initial groundwater monitoring efforts at the site involved use of groundwater monitoring wells. In August and November 1987, seven monitoring wells (MW-1, MW-2, MW-3, MW-4, MW-5, MW-6I, and MW-6D) were installed in the shallow bedrock by Westinghouse (see Figure A-2) to study groundwater flow and obtain groundwater samples. The construction details for the wells are presented in Table B- 1. Based on water level measurements, the groundwater flow is consistently to the west-northwest, and shallow groundwater is believed to discharge directly to Stout's Creek (B&B 1989; Viacom 2002a). Table B-2 shows historical groundwater elevation data. Figure A-5 is a typical groundwater elevation contour map demonstrating the west-northwest flow direction observed in 1988. Groundwater gradients vary from 0.05 to 0.03 foot per foot in the southern portion of the site and from about 0.03 to 0.02 foot per foot in the central and northern portions of the site. The depth to groundwater ranges from about 2.5 to 14 feet bgs.

Rock cores were obtained from the bedrock units at monitoring wells MW-1, MW-2, MW-3, MW-4, and MW-6D. Few enlarged solution voids were noted in the rock cores. Caliper and video logging of several wells by Westinghouse did not reveal the presence of major voids or conduits, but video logging did reveal small solution conduits in the upper 10 to 15 feet of the bedrock. Packer tests conducted during monitoring well installation indicated that the bedrock at the well locations had a relatively low watertransmitting capability. Estimated hydraulic conductivities ranged from 2.5 x 10-3 to 9.1 x 10~ centimeter per second (cm/s), with most estimates falling between 1 x 10- and 1 x 10-5 cm/s; the tested intervals in well MW-5 showed no response (Westinghouse Environmental Services and B&B 1989). Attachment A contains a copy of the Westinghouse report describing the well installation and associated activities. Wells MW-2 and MW-4 were abandoned during the 1999 RA.

The groundwater flow system associated with the site is composed of recharge and discharge areas. The recharge occurs in topographically higher areas east and south of the site. Recharge is probably accentuated by the numerous water-filled quarry pits; large areas of exposed bedrock; and backfilled quarry pits present in these areas. A large, water-filled pit referred to as "Icebox Quarry" is present about 100 yards east of the northern section of the site. Another water-filled pit referred to as "Wedge Quarry" lies south of Icebox Quarry (see Figure A-2). Wedge Quarry is part of a large, interconnected quarry complex extending southeast from the site. Many of the quarries in this complex have recently been backfilled as part of new highway construction at the State Route 37 and 46 interchange.

Four springs are located within the vicinity of the site. Two springs, Middle Spring and Mound Spring, emerge in the former main fill area (which has been excavated and backfilled), and flow during all but the driest times of year. The discharges from these springs flow into Stout's Creek. Two other intermittent springs, North Spring and Mid-North Spring, lie at the perimeter of the former main fill area. A 24-inch-diameter culvert is present at North Spring and drains into Stout's Creek. The locations of the springs and culvert are shown in Figure A-2. North Spring has not flowed since before the RA excavation in 1999. Mid-North Spring flows during wet periods.

Middle and Mound Springs first appeared in 2000 after the completion of the RA, but Groundwater discharge had been noted previously in the area (Weston 1983). The emergence of these springs may have been the result of re-establishment of prior hydrologic flow pathways following the l999 remediation. Observations made by an EPA on-scene coordinator on May 12, 1983, indicated the presence of a pond in the eastern part of the site (Weston 1983). The report also describes an adjoining pond that appeared to drain west to Stout's Creek and a ponded area adjacent to the railroad tracks along the western part of the site. In June 1996, contractors for the Indiana Department of Transportation noted seeps flowing across the railroad grade into Stout's Creek just south of monitoring well MW-5 (Viacom 2002b).

The hydrogeologic characteristics of the site observed in the past are being further evaluated to assess the impacts of the 1999 RA as well as the construction of the State Route 37 and 46 interchange. This ongoing evaluation is discussed in Section 3.0.

Viacom is conducting ongoing investigations to gain additional information on hydrogeology, groundwater recharge, and PCB migration at the site. These investigations have included the collection of continuous water flow and elevation monitoring data at Middle Spring, Mound Spring, Stout's Creek, Icebox Quarry, Wedge Quarry, monitoring well MW- 1, and Piezometer H.

Beginning in 2002, the investigations have been conducted under the final groundwater investigation plan for the Bennett's Dump site (Viacom 2002b). Attachment B contains a copy of this plan. The specific objective of the investigations is to define the relationship among groundwater elevations, water elevations in the quarries, and spring flow. This information will then be related to PCB concentration information that is being collected under the long-term groundwater monitoring plan. Attachment C contains a copy of this plan.

This section summarizes the monitoring activities being conducted under the final groundwater investigation plan, and discusses continuous monitoring and groundwater investigation results.

3.1 FINAL GROUNDWATER INVESTIGATION PLAN

3.1.1 Monitoring Groundwater Elevations

To supplement the groundwater monitoring wells installed in 1987, Viacom installed a Piezometer in the H-7 remediation grid area in 2001. The purpose of Piezometer H is to measure the Groundwater elevation and PCB concentration in an area where deep contamination was found during the 1999 RA. The Groundwater elevations in monitoring wells MW-1, MW-3, MW-5, MW-6I, and MW-6D and in Piezometer H will be monitored under normal conditions and during a storm event. The Groundwater elevations will be compared to spring flow, stream flow in Stout's Creek, and quarry pit water levels.

Well MW- 1 and Piezometer H have been equipped with continuous water level monitoring equipment. Well MW-1 was instrumented in April 2001, and monitoring at Piezometer H was initiated in January 2002 soon after it was installed.

3.1.2 Monitoring Water Levels in Adjacent Quarries

3.1.3 Monitoring Spring Discharge Rates

3.1.4 Monitoring Flow Rates in Stout's Creek

Stout's Creek and the West Fork of Stout's Creek flow to the north in the vicinity of the site. A gaging station was installed in Stout's Creek at the downstream (northern) end of the site immediately upstream of the confluence with the West Fork of Stout's Creek. The gaging station consists of a staff gage and continuous water level measurement equipment. A transducer installed at the gaging station allows continuous logging of creek flow readings. The gaging station was placed in operation on April 20, 2001. A rating curve developed for the station allows water level readings to be converted into water flow rates in the creek. The rating curve, which was developed by Viacom, is shown in Figure A-6 (Viacom 2001). Stout's Creek flow is calculated as follows:

Flow in gallons per minute (gpm) = 745.53(water depth)^4.3711

PCB concentrations in the creek will be measured monthly for l year and during one storm event annually (Viacom 2001). PCB data obtained to date for Stout's Creek is discussed in Section 4.0.

3.2 CONTINUOUS MONITORING AND GROUNDWATER INVESTIGATION RESULTS

Data gathered by Viacom during the Bennett's Dump groundwater investigations is submitted to EPA on a quarterly basis. IDEM has also made periodic hydrologic measurements at the site. IDEM's activities have included water level measurements in quarries, measurements of the groundwater elevation at Piezometer H. and measurements of flow through the spring weirs. The hydrogeologic measurement data collected by IDEM is included in Attachment D. Several trends have been noted during preliminary analysis of the data. These trends are discussed below.

The IDEM data indicates that the water levels in Icebox Quarry and Wedge Quarry are higher than the groundwater elevation at Piezometer H (see Figure A-7). Both quarry areas could thus act as groundwater recharge areas for the spring discharges on site. Water level fluctuations in Wedge Quarry are greater than those in Icebox Quarry, and this may indicate a greater degree of surface water inflow to the quarry complex. Icebox Quarry shows a relatively small degree of water level fluctuation. The groundwater elevation in Piezometer H shows greater fluctuation than the Icebox Quarry water levels, and the fluctuations closely mirror the water level in Wedge Quarry.

The IDEM data also indicates that Mound Spring has a higher discharge rate than Middle Spring. Flow rates at Middle Spring are generally less than 10 gpm. Flow rates at Mound Spring are more variable and can be as high as 25 to 30 gpm (see Figure A-8). The flow rate at Mound Spring appears to be closely related to the water level in Wedge Quarry (see Figure A-9). Flow at Mound Spring stops or is greatly reduced when the water level in Wedge Quarry falls below about 740 feet above mean sea level (amsl).

Data gathered by Viacom during a storm event beginning on June 4, 2001, illustrates several features of the hydrologic system at Bennett's Dump. Figure A-10 shows time-series plots for several parameters monitored during the storm event, including water levels in abandoned quarries, and the Groundwater elevation in monitoring well MW-1, the stage at the Stout's Creek gaging station, and flow rates and PCB concentrations in Middle Spring. The sequence of hydrologic events was as follows:

• At the beginning of the storm, a sharp rise in discharge at Middle Spring occurred. Water levels at MW-1, Icebox Quarry, and Wedge Quarry all rose, but the largest rise occurred at Wedge Quarry, probably because of the amount of surface runoff that entered the Wedge Quarry complex.
• The stage in Stout's Creek increased and peaked about 4 hours after the beginning of the storm as water from upstream portions of the watershed reached the site. The recession to pre-storm flows occurred over a period of a few days.
• Flow at Middle Spring increased slowly during the peak and recession of Stout's Creek flow. The flow at Middle Spring did not peak until about 4 days after the storm. As a result of the storm, Middle Springs flow increased from about 2 to 8 gpm.
• The PCB concentrations of the Middle Spring discharge ranged from 1.5 to 10 ppb, but there was no consistent correlation between the PCB concentration and the discharge rate. PCBs were discharged at all flow rates. Some of the variability in PCB concentrations may have been related to the length of time that water samples remained in the autosampling equipment after sample collection.

No flow data was collected at Mound Spring during the storm event. PCB concentrations in Mound Spring were lower than in Middle Spring during the storm event and ranged from 0.46 to 2.7 ppb (see Figure A-10).

Viacom estimated flow rates for the springs and Stout's Creek while conducting sampling during the first quarter of 2002 (Viacom 2002c). The range of flow rates estimated during three first-quarter monitoring events was as follows:

• Middle Spring: 1 to 3.5 gpm
• Mound Spring: 1.5 to 8 gpm
• Mid-North Spring: 0 to 2 gpm
• Stout's Creek-upstream 200 to 700 gpm
• West Fork of Stout's Creek: 250 to 800 gpm
• Stout's Creek-downstream: 500 to 2,000 gpm

Viacom has reported the results of the first quarter of monitoring under the final groundwater investigation plan (Viacom 2002e). Monitoring under this plan was initiated in April 2002, which corresponsds to the second quarter of 2002. Figure A-11 is a groundwater elevation contour map based on data collected on April 29, 2002; the flow directions and gradients shown are consistent with those measured during sampling activities conducted in 1988 (see figure A-5). Discrete groundwater elevations and quarry and spring water levels measured under the groundwater investigation plan are summarized in Table B-3. Base on the continuouis monioring data supplied in Viacom's report (Viacom 2002e), the following observations cam be made:

• Water levels in Icebox Quarry ranged from about 738.5 to 739.5 feet amsl.
• Water levels in Wedge Quarry ranged from 1 to 3 feet.
• The stage of Stout's Creek ranged from 1 to 3 feet
• Flow rates in Mound Spring ranged from 8 to 35 gpm
• Flow rates in Middle Spring ranged from 4 to 13 gpm

Since 1983, various investigations have been conducted to characterize the extent of PCB contamination at the Bennett's Dump site. These investigations resulted in as IRM that took place in 1983 as described in Section 2.1. The IRM included removal of all visible capacitors, removal of some stained soil, and installation of a clay cap. The results of investigations conducted prior to the 1983 IRM may not be directly comparable to those of later investigations because PCB-related material was removed and clay cap installation helped prevent rainwater infiltration into the main fill area.

This section summarizes the current status of PCB data obtained for Groundwater, springs and surface water, soil and sediment, and fish tissue (Sections 4.1 through 4.4, respectively). Sample analytical results are presented for investigations when available; much of the data acquired at Bennett's Dump prior to the 1990s is not covered extensively.

The statement of work for the 1999 RA required development of an interim Groundwater monitoring plan, followed by a 5-year long-term Groundwater monitoring plan. Before excavation activities began and every 60 days during remediation, the following locations were sampled as part of the interim monitoring plan:

• MW-6I
• MW-6D
• MW-3
• Stout's Creek upstream and downstream of the site
• The West Fork of Stout's Creek upstream of its confluence with Stout's Creek