Earth Tech, Inc., Personnel: John Bassett
Reporting Period: July 31 through August 20, 2002

As requested by the U.S. Environmental Protection Agency (EPA) under Contract No. 68-W6-0037, Work Assignment No. 941-RSBD-OS29, Tetra Tech EM Inc. (Tetra Tech) is conducting site visits and performing oversight activities at the Lemon Lane Landfill site in Bloomington, Indiana. Viacom, Inc. (Viacom), is the responsible party for the site. Viacom is conducting ongoing investigations to better understand groundwater flow patterns and polychlorinated biphenyl (PCB) transport to Illinois Central Spring (ICS) downgradient of the site. During the reporting period, Viacom performed testing on piezometer PZ-F and monitoring well NN-300A.

This report summarizes (1) activities observed by Earth Tech, Inc. (Earth Tech), Tetra Tech's subcontractor, at the Lemon Lane Landfill site on July 31, 2002; (2) verbal reports received by Earth Tech of activities conducted at the site on August 6 and 20, 2002, from Viacom; (3) issues and developments; and (4) future activities at the site. Appendix A contains the figures cited in the text, Appendix B contains a photographic log of field activities, and Appendix C contains Earth Tech's field notes.

Earth Tech's activities and observations at the Lemon Lane Landfill site from July 31 through August 20, 2002, are summarized below.

Wednesday, July 31, 2002

Earth Tech conducted oversight of a flush test conducted by Viacom on piezometer PZ-F, which is installed in the epikarst at an elevation of about 820 feet above mean sea level (amsl). The goals of the test were to (1) clean sediment out of PZ-F and (2) assess the capability of the aquifer to transmit water.

The test setup consisted of a fabricated "T"apparatus (see Photograph No. 1). Water from a fire hydrant was injected into the top of the "T." Water flow rate was measured at the fire hydrant using a flow meter. Water was injected into the open-hole piezometer by a nominal 1.5-inch-diameter drop pipe installed to within 6 inches of the piezometer bottom. Washed material was returned to the surface through the annular space between the open hole and the drop pipe. This material was routed to a truck- mounted, polyethylene tank (see Photograph No. 2).

The flush test began at 12:56 p.m. under the direction of Viacom's geologist, Mike McCann. An initial flow rate of 2 gallons per minute (gpm) was used. No overflow to the tank had been noted by 1:25 p.m. The flow rate was increased to 6 gpm. Beginning at 1 :27-p.m., attempts were made to determine the water level in the piezometer using an electric water level meter. The measurements were taken inside the 1.5-inch-diameter drop pipe and proved to be difficult because of the cascading water inflow. After no overflow had been observed by 1:34 p.m., the flow rate was increased to 10 gpm. An apparently valid water level measurement taken at 1:43 p.m. indicated a water level of 22.80 feet below the measuring point (bmp). This measurement point was 2.97 feet above the steel casing of PZ-F.

At 1:50 p.m., the flow rate was increased to 30 gpm. Earth Tech checked site monitoring wells 00-387, 00-370, NN-300, and NN-300A for any evidence of cascading water. No cascading water was noted. No useful water level measurements could be obtained in PZ-F at the 30-gpm flow rate.

At 1:56 p.m., the flow rate was increased to 50 gpm because no overflow to the tank had been noted. However, no overflow occurred even at this flow rate, the highest used during the test. The flow was stopped at 2:05 p.m. after a total of 988 gallons of water had been injected into PZ-F. Water level measurements were obtained in PZ-F after the water injection ceased. These measurements are shown in the following table.

Little or no sediment was removed from the bottom of the piezometer during the test. Measurements taken before and after the flushing confirmed a piezometer depth of 55.2 feet bmp, or about 52.2 feet below the top of the steel casing.

Tuesday, August 6, and Tuesday, August 20, 2002

Viacom reportedly performed dye trace tests at piezometer PZ-F and monitoring well NN-300A on August 6 and 20, 2002, respectively. Earth Tech was not present at the site on these dates but received verbal reports of the test results from Viacom. These results are discussed below.

ISSUES AND DEVELOPMENTS

Figure A-1 is a hydrograph of piezometer PZ-F water levels after completion of the flush test. The figure shows the depths to water levels measured with 2.97 feet above the steel casing as the reference point. An increase in depth to water level corresponds to a decrease of water level in the piezometer. It is apparent that voids within the epikarst at this location are capable of accepting a considerable volume of water, as no circulation return was observed at flow rates as high as 50 gpm. Water levels measured during piezometer recovery displayed a nonuniform rate of water level decline. Because the water level initially declined slowly from a level of about 20 feet bmp, the principal voids accepting the flow were probably located above this level. A pronounced increase in the rate of water level decline occurred 9 to 17 minutes into the recovery period. This is probably due to the sudden opening of an epikarst void in contact with the well bore and the release of water into the void. As the water level appeared to resume its initial rate of decline below a level of about 45 feet bmp, the void may have opened up at about that level.

Viacom later reported to Earth Tech by telephone that the flush test conducted on July 31, 2002, resulted in a 45 part per billion (ppb) PCB spike at ICS downgradient of the site. The ICS sample was collected 4 hours after injection, so the actual peak concentration may have been higher. Based on Groundwater travel time estimates, the peak concentration was not expected to arrive at ICS until about 12 hours after injection. Subsequent ICS samples had PCB concentrations of about 11 to 12 ppb. Viacom also reported that water levels in nearby wells in the phreatic zone rose about 0.2 foot within 1 hour of injection. This occurrence is also an indication that water injected into PZ-F reaches the phreatic zone quickly.

Viacom also reported the results of the dye trace test conducted at PZ-F on August 6, 2002, to Earth Tech by telephone. During the test, 225 grams of Rhodamine WT dye was injected into PZ-F, followed by 1,000 gallons of water. The dye was detected in monitoring wells 00-370 and 00-300A as well as at ICS. Figure A-2 is a graph of the dye concentrations detected at 00-370, 00-300A, and ICS. 00-300A is open to the vadose zone (840 to 855 feet amsl), while 00-370 is in the phreatic zone (795 to 825 feet amsl). The wells are located about 70 feet apart. The detection of dye in both wells indicates that a portion of the water traveling from the area of PZ-F remains in the vadose zone until it is very near 00-370, at which time it migrates vertically into the phreatic zone.

In addition, Viacom provided Earth Tech with a verbal summary of the dye trace test conducted at monitoring well NN-300A on August 20, 2002. Beginning at 9:15 a.m., 225 grams of Rhodamine WT dye was injected into the well. The dye flush ended at 9:46 a.m. Dye was visually noted at monitoring wells 00-370 and 00-300A beginning at 9:51 a.m. No dye was visually detected at monitoring well 00-387, which is only 17 feet west of 00-370.

Finally, Viacom supplied draft, raw data regarding the dye trace test performed at monitoring wells MW4i and MW-6 on July 10, 2002. This test is described in Tetra Tech's field oversight summary dated July 22, 2002. Briefly, 225 grams of Rhodamine WT dye was injected into MW4i, and 50 grams of fluorescein dye was injected into MW-6. Viacom monitored the injection wells, nearby wells, ICS, and Quarry B spring by testing for the presence of dye. Figure A-3 is a graph of the dye concentrations detected in ICS and Quarry B spring. Two significant findings are apparent in this graph. First, no significant fluorescein peak was detected at ICS (background concentrations from previous dye tests were detected), which suggests that groundwater in the vicinity of MW-6 does not follow a path to ICS. No fluorescein has been detected at ICS since the test. Second, no significant Rhodamine WT peak was detected at Quarry B spring (background concentrations from previous dye tests were detected), which is downgradient of ICS. This finding (1) suggests that the Rhodamine WT dye that emerged from the ICS orifice was captured by the ICS treatment plant (which is located between ICS and Quarry B spring) and (2) demonstrates that no bypass occurs under low-flow conditions.