supcrfund record centes r site: '••i-^.^okc 2.?. ollifck · the instrumen t mus bte able...

Beach, Richard

Supcrfund Records Center SITE: '••i-^.^OKC hk l -AK: 2.?. Ol l i f ck :

From: Sent: To: Cc:

Subject:

Beach, Richard Wednesday, October 03, 2001 5:30 PM Jim Rhea (E-mail); Andy Silfer (E-mail) Susan C. Svirsky (E-mail); Moore Darrell (E-mail) (E-mail); Bob Wagner (E-mail); Farrell Timothy; dePersia, Lee R. FW: PRIVILEGED: Oct 3 2001 - Supplemental SW Study SOP for GE/QEA/NEA

Jim, Andy,

Please find attached the Supplemental SW Study SOP that incorporates all of the current changes.

I am also copying Bob Wagner, NEA, for lab coordination purposes..

Depending on the weather, Rich Totino (Weston task manager) anticipates that we will be able to run the trial event on Monday (possibly Tues) next week and then run the baseline event on Wed-Thurs if the trial goes well (and we have baseline conditions).

Previously, we discussed having a QEA person help on the processing. At this late date, we are not dependent on that support. If you do want to observe the processing, please call me as I currently plan to be there Monday pm and thru the majority of the week.

Rick Beach Roy F. Weston, Inc. 610-701-3473 610-701-3125 (fax)

The information contained in this e-mail message is intended only for the personal and confidential use of the recipient(s) named above. This message may be an attorney-client communication and as such is privileged and confidential. If the reader of this message is not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error and that any review, dissemination, distribution, or copying of this message is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail, and delete the original message.

STANDARD OPERATING PROCEDURE for SUPPLEMENTAL SURFACE WATER STUDY

October 3, 2001

1.0 Objective

The supplemental surface water study will measure the total, particulate, and dissolved phases of PCBs and organic carbon (OC) in surface water in the Housatonic River Primary Study Area (PSA). The PSA begins at the confluence of the East and West Branches of the Housatonic River, and ends at the Woods Pond dam. These data will be used to determine the distribution of PCBs and OC between the two phases of these parameters, and the results will be utilized as part of the modeling study of the Housatonic River (River). Additional parameters such as total suspended solids (TSS) and chlorophyll-a will also be measured and correlated with the PCB and OC data in support of this objective.

2.0 Technical Approach

A baseline study of the surface water during non-storm water conditions was conducted monthly at seventeen (17) stations along the River in the PSA from August 1998 to September 1999, during a period when no removal actions were occurring. PCBs and OC and were measured as part of the study protocols. However, the organic carbon was measured analytically as the total and dissolved components, and the particulate phase calculated by the difference in the results. This approach resulted in increased uncertainty in the measurement of the particulate organic carbon ("POC"), as the POC phase was small relative to the magnitude of the dissolved phase. Additionally, these analyses were performed on two separate samples of water collected side by side. In many of the samples, negative values for POC were calculated from the subtraction of dissolved OC (DOC) from the total OC (TOC). The current study will analyze the TOC, POC, and DOC phases directly.

The baseline study provided data on particulate and dissolved phases of PCBs as Aroclors and congeners. During the development of the Conceptual Model for the modeling study (in collaboration with General Electric's consultant, Quantitative Environmental Analysis, LLC. (QEA), a need for additional data for these parameters was identified in concert with a need to improve the assessment of the OC phases in the surface water. The supplemental study will include the analyses of these phases directly, and will utilize personnel and resources from both Weston and QEA. The costs associated with the PCB analyses will be paid directly by General Electric (GE) to Northeast Analytical Laboratories, Schenectady, NY (NEA). Roy F. Weston will contract other analyses.

2.1 Analytes and Data Quality Indicators

1. PCBs - bulk (particulate and dissolved) water sample, (determined on an aliquot of the sample prior to filtration). Extraction Method NE-118_03.SOP Revision 03. Analytical Method NE-013J34 Revision 04. (Congener technique by GC-ECD based on "Green Bay

Supplemental Surface Water Study - October 3,2001

Method" for the identification of 112 peaks. The laboratory will report the homolog and Aroclor distributions).

2. PCBs - participate. Extract and analyze the suspended material on the filter. Extraction Method NE-005_04 (Analysis by Soxhlet). Analytical Method NE-013_04 Revision 04. (Note: The laboratory will perform an overnight drying of the filter at room temperature, followed by folding and inserting the filter into the soxhlet thimbles with sodium sulfate). (Same method and comments noted above for total PCB congeners).

3. PCBs - dissolved, (extract and analyze the filtrate from the sample filtration). Extraction Method NE-118JB.SOP Revision 03. Analytical Method NE-013_04 Revision 04. (Same method and comments noted above for total PCB congeners).

4. TOC - total organic carbon (determined on an aliquot of the sample prior to filtration). Analytical Method NE-128_03 Revision 03. This method uses a sample injected into a high temperature carbon analyzer with a non-dispersive infrared (NDIR) detection. The method may need to be altered if high levels of suspended material are present in the samples. The instrument must achieve a detection limit of 1.0 ppm (preferably a DL = 0.5 ppm), an analytical precision of + 25 RPD, and an accuracy of + 35% at a level of 2 ppm using a reference material or an alternate standard. The sample preparation will incorporate an acid treatment to remove inorganic carbon prior to analysis.

5. POC - particulate organic carbon (determined on three "punches" from the sample filter). Analytical Method NE-128_03 SOP Revision 03. This method quantifies the carbon mass on the filter punches. This method uses a high temperature carbon analyzer with a non-dispersive infrared (NDIR) detection following a modified "Lloyd Kahn" protocol. The method must achieve a sensitivity of 10 ug of OC per filter punch, an analytical precision of + 25 RPD at a level between 25 to 75 ug OC per sample, and an accuracy of + 35% to a reference material or an alternate standard. This sensitivity is comparable to a sample detection limit of 0.04 ppm POC for a sample size of 16 liters and the processing protocol defined for this study. The instrument must be able to accept a sample consisting of a 13 mm diameter "punch" from a glass fiber filter. Analysis will consist of three separate analytical measurements, one on each of the three punches. The POC will then be calculated by scaling up the average carbon mass per unit area of the punch to the total filter area and then dividing by the total volume filtered, to yield the POC in mg/L. The sample preparation will incorporate a phosphoric acid treatment to remove inorganic carbon and should not employ drying temperatures greater than 40° C.

6. DOC - dissolved organic carbon (analyze the filtrate from the sample filtration). Analytical Method NE-128_03 Revision 03. This method uses a sample injected into a high temperature carbon analyzer with a non-dispersive infrared (NDIR) detection. The instrument must achieve the same detection limit, precision, and accuracy criteria specified above for the TOC analysis. The sample preparation will incorporate an acid treatment to remove inorganic carbon prior to analysis.

7. TSS - Total Suspended Solids (determined on an aliquot of the sample prior to filtration). EPA Method 160.2. As referenced in: Final Quality Assurance Project Plan (QAPP) November 2000, and the related Volumes II and IIA for the Laboratory Standard Operating Procedures.

8. Analysis for chlorophyll-a (determined on an aliquot of the sample prior to filtration). EPA Method 10200. As referenced in Final Quality Assurance Project Plan (QAPP)

Page 2 of 11


November 2000, and the related Volumes II and IIA for the Laboratory Standard Operating Procedures.

2.2 Sampling Locations

Samples will be collected at four stations along the River.

• Pomeroy Avenue Bridge. This station on the East Branch is located immediately upstream of the PSA. This station was used as a sampling location in the monthly surface water baseline study and the 1999 storm water sampling program.

• West Branch (of the Housatonic River). This station is approximately 200 feet upstream from the confluence with the East Branch of the Housatonic River and was used as a sampling location in the monthly surface water baseline study and the 1999 storm water program.

• New Lenox Road Bridge. This station is approximately in the middle of the PSA and was used as a sampling location in the monthly surface water baseline study and the 1999 storm water program.

• Woods Ponds Footbridge. This station is the next to the last downstream sampling location within the PSA and was used as a sampling location during the 1999 storm water program. It is approximately 300 meters upstream of the Woods Pond Dam, which is the lower boundary of the PSA. The Woods Pond Dam station was used as a sampling location in the monthly surface water baseline study.

2.3 Sampling Events

Three sampling events at the four stations noted above will be scheduled to evaluate the River at low, medium, and high flow conditions during non-storm event periods. Low flow conditions are defined here by River flows less than or equal to 50 cubic feet per second (cfs), medium flow as approximately 100 cfs, and high flow as greater than 200 cfs, as recorded by the Coltsville USGS Hydrograph Station. The specific conditions acceptable for the three flow regimes will be decided according to the weather conditions during fall 2001. A single sampling event with three replicate samples at the Pomeroy Avenue Bridge will be conducted during low River flow conditions as a trial of the protocol and the analytical procedures. Data derived from this initial sampling event will be used to evaluate methodology and may be utilized for data analysis.

2.4 Sample Collection Methods

At each station per sampling event:

1. The sampling events should be conducted during river conditions not under the immediate influence of storms (e.g. short-term changes in flows). The Coltsville hydrograph should be reviewed prior to all sampling to insure the river flows have been relatively constant for a period of at least 48 hours prior to the planned event. In addition, the weather forecast should be checked to judge whether storms are forecast for the sampling period.

Page 3 of 11


2. Use an ISCO Model 150 peristaltic pump, or equivalent, set at a nominal flow rate of 3.25 liters per minute (2.5 feet per second (fps), velocity for a 3/8" ID tubing) to continuously fill an 8-gallon (minimum) glass container from the water column. Record the water temperature in the river at the beginning and end of the sampling effort.

3. Use Teflon or Teflon-lined tubing, dedicated to each location, for the sample intake tubing. Each sample event/location should use the minimum length of intake tubing necessary to reach from the ISCO sampler to the sample intake point. Minimize the lengths of silicon-based tubing (dedicated to each location) used for the peristaltic pump head. No solvent will be used to initially wash or clean the tubing.

4. Fill and purge the sample tubing for a minimum of twenty (20) tube volumes prior to filling the sample container. Fill the container to the shoulder of the bottle. This should yield approximately 27 L of the sample for processing.

5. Position the intake of the collection tubing in the water column at each station at a depth equivalent to 0.6 of the total water depth (e.g. six foot sampling depth from the surface of the water when the total water depth is 10 feet). Do not allow the tubing to pick up sediment from the river bed during deployment or the actual sampling. Orient the intake opening of the collection tubing downstream.

6. Reverse the ISCO pump control to allow air to flush the water out of the intake tubing after the sample collection.

7. Wrap each sample container with a cool thermal wrap to maintain the sample temperature during sampling and transport to the field office. Record the temperature of the river at the sample location before and after collecting the sample. The equilibrium between the particulate and dissolved phases of organic carbon and PCBs, respectively, is a direct function of the temperature. River temperatures are expected to be in the range of 55 to 75° F.

8. The sampling velocity should exceed the velocity (preferably by 150%) in the River. River velocities during the 1999 storm water sampling program reached approximately 2.4 fps during some storm events at the selected station locations. The use of high sampling rates will reduce the possibility of particles settling in the sample tubing during transit from the River to the sample collection vessel.

9. Utilize PCB decontamination protocols for all equipment used in the sample collection and processing with the exception of the Teflon intake and silicone pump tubing. Normal decontamination protocols involve a sequence of alconox-water, acetone, hexane, acetone, and thorough distilled water rinsing. The final water rinsing should remove all organic solvent residuals and no organic odors should be evident. The Teflon and silicone pump tubing will only be cleaned with a thorough flushing/rinsing with distilled water in order to minimize residual organic carbon on the surfaces in contact with the samples. If non-aqueous phase liquids (NAPLs) are noted inside the tubing, the tubing must be discarded and the condition noted with the associated sample. Store all equipment cleaned at the end of each sampling event in double plastic bags. These procedures are a modification of the Field Sampling Plan, General SOP G-2 Decontamination, November 2000.

2.5 Other Data Collection

Page 4 of 11


Download the Coltsville hydrograph record from the USGS web site for flows and velocities for 1 week prior to and during the sampling event. Measure the staff gage readings (and convert to elevation) at each of the sampling stations during each sampling event. If available for the station at the time of the draft report, use the rating curves for the location to determine the flows.

2.6 Sample Processing Methods

1. Processing of each sample should be initiated within 4 hours of sample collection. 2. Record the temperature of the composite sample before and after processing the sample.

Record the ambient room temperature in the sample processing area. 3. While mixing the sample thoroughly with an overhead shaft mixer system, use a

peristaltic pump (with silicone tubing dedicated to each location) to fill two 1-liter amber glass bottles for TSS analysis (second bottle is backup for possible transit breakage), four 40-ml volatile organic (VOA) glass vials for TOC analysis (first two vials for analysis, second two vials for possible transit breakage), and three 1-liter amber glass bottles for total PCB analyses (the third bottle is a backup for possible transit breakage). The mixing activity and pumping rate must be adequate to suspend and sample the particulates present in the sample. In order to evaluate the processing and analytical variability in samples, the trial sampling event will involve the collection of triplicate samples for all parameters. See Table 2 for a summary of analytical samples.

Note: TSS analysis for the study will utilize a glass fiber filter with a nominal pore size of 0.7 urn. A pre-study experiment was conducted to determine variability in TSS results between filters with a 0.7 um pore size and a 1.5 um pore size (the standard filter pore size used in standard TSS analyses). The results are presented in Table 1 below and indicate no significant difference in TSS levels as measured by the two pore sizes. The pore size of 0.7 um will be utilized for this study as the smaller pore size will be closer to "dissolved" conditions.

Table 1 Summary of TSS Results with Varying Filter Pore Size

Sample ID Filter Pore Size TSS (mg/L)

H2-SW00051 -0-00021 0.7 28.2 H2-SW00051 -0-00022 0.7 28.3 H2-SW0005 1-0-00023 0.7 35.3 H2-SW00051 -0-00024 0.7 32.9 H2-SW00051 -0-00025 0.7 33.7

Average Concentration 31.7 + 3.2 (1 SD)

H2-SW00051 -0-00026 1.5 27.8 H2-SW00051 -0-00027 1.5 26 H2-SW00051 -0-00028 1.5 31.6 H2-SW00051 -0-00029 1.5 29.8 H2-SW00051 -0-00030 1.5 30.3

Average Concentration 29.1 + 2.2 (1 SD)

Page 5 of 11


4. Add approximately a 0.5-liter portion from the sample collection container into the upper pressure chamber of the filtration unit. Rinse the filtration unit by gently swirling the water, and discard the rinsate immediately before assembling the unit for sample filtration.

5. Add approximately a 1-liter aliquot from the mixed collection container (using the same peristaltic pump used for the initial sample aliquots and the rinsing) to the pressure chamber of the stainless steel (SS) filtration unit (nominal inside diameter, ID, of 142 mm). The units have an effective filtration ID of approximately 120 mm (the actual filtered area will be measured on each sample filter).

6. The filtration unit will utilize a 142-mm diameter Whatman GF/F glass fiber filter (nominal filter pore size of 0.7 microns) or equivalent that is oven "cleaned" at 450 °C (minimum, 525 °C maximum) for one hour, weighed, and is supplied in petri dishes from the analytical laboratory. The laboratory should run at least one filter blank for the particulate analyses of PCBs and OC for each sampling event. Once a specific filter manufacturer and pore size is selected, the same filter selection and preferably specific batch should be used for all analyses.

7. Using nitrogen gas, increase the pressure (as little as possible, but no higher than 15 psi, a pressure that may result in the filter "blowing out" the sides of the filtration unit) to the sample pressure chamber to filter each aliquot from the mixed sample collection container.

8. Add additional aliquots for filtering if the sample processor believes the entire aliquot can be filtered without changing the filter.

9. Collect all of the filtrate from each sample into 1-liter glass bottles. The bottles should be processed sequentially and labeled. Measure the sample volume in each bottle. The amount of filtrate can be measured by recording the change in mass between a tared bottle and a filled bottle and converting the water mass to volume based on the density of water of 1 gram per milliliter. Report the sample filtrate volume with the sampling report. Record the temperature of each liter of sample immediately after the bottle is filtered.

10. When all of the sample composite has been filtered, or when the processor believes another aliquot cannot be pressure filtered within an additional 2 hour period through the single filter, the processor should allow the nitrogen gas from the pressure housing to gently purge the excessive water through the single filter to complete the filtration step. If possible, filter up to 16 liters of composite sample through the single filter.

Note: The use of a single filter for the entire sample will "load" as much of the sample particulates onto one filter as possible from a sample volume of 16 Liters when low suspended solids levels are present. A pre-study experiment was conducted to assess the optimal volume of sample to filter. The optimal value would be the minimum sample volume that saturated the sorptive sites on the filter and apparatus, and allowed PCBs to pass through the filter at levels representative of the dissolved concentrations in the river sample. The results of the experiment did not suggest any trend in PCB concentrations in the sequential 1-liter filtrate samples (See Figure 1). Although no differences were noted in this experiment, the study will use the last filtered sample volumes first to minimize the impacts of the potential sorptive losses on the first aliquots of sample filtered. Processing techniques used by TAMS Consultants, Inc. on surface water studies on the Hudson River required

Page 6 of 11


filtering a total of 16 L sample volumes in order to minimize the sorptive loss of PCBs onto the filter and associated processing equipment.

Note: The PCB results in Figure 1 include the sum of calculated peak groups (5 peaks per Aroclor) from Aroclors 1016, 1242, 1248, 1254, and 1260. While the results were calculated consistently for all samples the results do not represent a quantitation comparable to those from standard PCB analyses. The quantitation approach was necessary to provide a basis for the consistent interpretation of the filtrate results that were reported by the laboratory as below the practical quantitation limits (PQLs) for individual Aroclors.

Figure 1: Dissolved PCB Concentrations Through Sequentially Filtered Sample Volumes

Dissolved PCBs - Sum of Aroclors Grouped Peaks

025

015 m o a.

1 2 3 4 5 7 8 9 10 11 12 13 14 16 17 18 19 20

Sequential Sample (L)

11. Select the LAST two of the one-liter sample bottles of the filtrate for laboratory analysis of the dissolved PCBs and DOC. The laboratory should analyze the two bottles as one sample in order to lower the detection limit. If one bottle is broken in transit, The NEA laboratory will contact the Rich Totino (Weston Task Manager) and analyze the other bottle as the sample by itself.

12. Disassemble the filtration unit. The suspended solids will be contained on the filter in a circle of approximately 120 mm diameter. Measure and record the diameter (to the nearest mm) of the sample on the filter. This measurement should be constant for each filtration unit and may be the same between units depending on the manufacturing tolerances of each of the custom built units.

13. Remove the filter with clean, smooth tipped forceps and place on a clean plastic cutting board.

Page 7 of 11


14. Using a clean, metal cork borer and a plastic cutting board, gently cut or "punch" out three filter punches located around the filter to represent the filtered material. Use a # 7 cork borer, which represents a diameter of 13 mm. The borer may later be changed to achieve a different cutting diameter in order to match the detection range of the laboratory instrument for OC analysis. A specific randomized pattern of filter punches is not necessary, but the processor should avoid any areas of the filter that are void of particulates. These areas may occur if a bubble displaced particulates during the filtering process or if the filtered material was "washed off of some areas of the filter during the sample introduction to the filter housing. The filter punches will be proportioned to areas of the filter with lighter or heavier accumulation of particulate matter. The locations and characteristics of the filter punches will be noted in the report.

15. With narrow tipped forceps, gently place the filter punches in three separate pre-cleaned sample vials. Place the containers in a horizontal holder that will protect the samples and prevent them from agitation or vibration. Submit samples to the laboratory for the POC analyses.

16. Fold the remainder of the filter (after the punches were removed) with the smooth tipped forceps three times (along the radius into a size 1/8 of the original circumference) and then in half across the radius. Place the folded filter in a pre-cleaned 4-oz wide mouth jar for submittal to the laboratory for particulate PCB analysis. Place the jars with the folded filters in a holder or box within the shipping cooler and mark the cooler with a request for vertical handling.

17. Perform all handling carefully to minimize the potential loss of particulates from the filter surface. The samples will be transported by the laboratory courier instead of being shipped by FedEx to minimize sample disturbance.

18. Request the laboratory use the first aliquot of the PCB extraction solvent to rinse the jar into the extraction vessel to capture any particulates that may have fallen off the large filter in transit. The laboratory will report the PCBs levels on a per filter basis.

19. Field duplicate samples and matrix spike/matrix spike duplicate (MS/MSD) samples will be collected at one of the locations during a typical sampling event. Equipment blanks (as both water and filter samples) will be collected by sampling and processing deionized water through the standard protocol following the trial sampling event.

Page 8 of 11


Table 2 provides a summary of analytical samples for the study.

Table 2: Analytical Sample Summary

Sample # Chlor Organic Carbon PCBs Type Locations -a TSS TOC POC DOC Total Partic. Diss.

Trial 1 3 3 3 3 3 3 3 3 Low Flow 4 4 4 4 4 4 4 4 4 Med Flow 4 4 4 4 4 4 4 4 4 High Flow 4 4 4 4 4 4 4 4 4 Field

Duplicates 1 1 1 I 1 1 1 1 1

MS/MSD 2 2 2 2 2 2 2 2 2 Eqpt. Blank 1 1 1 1 1 1 1 1 1

Total 19 19 19 19 19 19 19 19

2.7 Reporting

The results of each sampling event will be summarized and presented as a report memo. At a minimum, the report will include the following:

1. Documentation of conditions and activities likely impacting the amounts of PCBs, OC, and suspended material in the surface water for 7 days prior to and during each sampling event. This will include a daily summary of any remedial activities and storm events during this period. The flow and velocity information from the USGS Coltsville Station will be presented in graphical format for the same time period prior to and during the sampling event.

2. Sample collection information on the sampling date and time, river temperature, and river and sampling observational notes. Any deviations from the decontamination or collection protocols will be noted.

3. River elevations at the time of the sampling for each station will be determined from the staff gage readings. If a rating curve for a station is available at the time of the report draft, the curve will be presented with the estimation of the flow during the sampling at each station.

4. Sample processing information on the sample volume filtered through the single filter, the volume of sample in each 1-liter bottle, the temperature of the sample during mixing and each 1-liter sample after filtration, and information regarding the submission and transport of the samples to the laboratory for analysis.

5. Summary tables of the sample results for:

• Total PCBs and congeners (analyzed directly, reported as ug/L) on the bulk sample,

Page 9 of 11


• Particulate PCBs and congeners (reported as ug/filter), • Dissolved PCBs and congeners (reported as ug/L), • Total organic carbon (reported as mg/L) on the bulk sample, • Particulate organic carbon (reported as ug/filter-punch for each punch), • Dissolved organic carbon (reported as mg/L), and • TSS (reported as mg/L) on the bulk sample.

The particulate concentrations on a volume basis will be calculated and reported from the ug/filter data supplied from the laboratory divided by the liters of filtered sample water.

Page 10 of 11


3.0 References

Standard operating procedure for the determination of total organic carbon and particulate organic carbon. NE-128_03. SOP Revision 03. Northeast Analytical Laboratory.

Standard operating procedure for the extraction and cleanup of aqueous samples by EPA method 3520c (continuous liquid/liquid extraction) for PCB analysis. NE-118_03. SOP Revision 03. Northeast Analytical Laboratory.

Standard operating procedure for the extraction and cleanup of soil, sediment, and solid samples by Soxhlet for PCB analysis. NE-005_04. SOP Revision 04. Northeast Analytical Laboratory.

Standard operating procedure for congener-specific polychlorinated biphenyl (PCB) analysis method for congener specific polychlorinated biphenyl (PCB) quantification and identification by capillary column/gas chromatography with electron capture detection. NE-013_04. SOP Revision 04. Northeast Analytical Laboratory, {modified "Green Bay" method}

Final Quality Assurance Project Plan (QAPP) November 2000. Roy F. Weston, Inc. General Electric (GE) Housatonic River Project, Pittsfield, Massachusetts {and the related volumes II and IIA for the Laboratory Standard Operating Procedures}.

Field Sampling Plan - General SOP G-2- "Decontamination," November 2000. Roy F. Weston, Inc. General Electric (GE)/Housatonic River Project, Pittsfield, Massachusetts.

Page 11 of 11

supcrfund record centes r site: '••i-^.^okc 2.?. ollifck · the instrumen t mus bte able...

Documents