RE Response to Work Plan Comments- Diffusion Sampler Pilot Test

From: KENT Mavis D

Sent: Wed Nov 30 10:28:00 2005

To: 'Chris Kimmel'

Cc: Sarah Prowell; Schlappi, Wayne R; Clint Jacob; Terry McGourty

Subject: RE: Response to Work Plan Comments- Diffusion Sampler Pilot Test

Importance: Normal

Thanks for your response. Let me know as we get closer in time what time period you will be installing the diffusion samplers. If I can, I would like to come out and see what you are doing. I will put a copy of this email in the file.

Mavis D. Kent

Cleanup & Emergency Response Section

DEQ Northwest Region - East Side Office

1550 NW Eastman Parkway, Suite 290

Gresham, Oregon 97030

503-667-8414 X55008

503-674-5148 Fax

kent.mavis.d@deq.state.or.us

-----Original Message-----

From: Chris Kimmel [mailto:CKimmel@landauinc.com]

Sent: Wednesday, November 30, 2005 9:56 AM

To: KENT Mavis D

Cc: Sarah Prowell; Schlappi, Wayne R; Clint Jacob; Terry McGourty

Subject: FW: Response to Work Plan Comments- Diffusion Sampler Pilot Test

Hello Mavis,

Thank you for your prompt review and approval of the Diffusion Sampler Pilot Test work plan . As indicated in the work plan, we will conduct the pilot test during the January 2006 sampling event. Response to your comments is provided below. Please call or email if you have any additional comments or questions.

Thank you,

Christine Kimmel, L.G.

Landau Associates

Project Geologist

main (425) 778-0907

fax (425) 778-6409

The information contained in this Email should be considered confidential and is intended only for the person(s) to whom it is addressed. Should you receive this transmission in error, please notify Landau Associates immediately and delete the Email.

From: KENT Mavis D [mailto:KENT.Mavis@deq.state.or.us]

Sent: Tuesday, November 15, 2005 8:36 AM

To: Clint Jacob; Sarah Prowell

Subject: Diffusion Sampler Pilot Test

I have reviewed the Diffusion Sampler Pilot Test work plan dated November 8, 2005. I agree with the work plan in general and approve it, but have a few questions.

1. Diffusion Bag Placement. The work plan indicates you will place the bags above the pump intake where there is a dedicated pump, or where you normally place a pump during regular sampling, and otherwise place one bag for every other 20 feet of screen. The diffusion bags will represent the zone of the well adjacent to where they are placed. Several factors can influence what the data from a diffusion bag represents including vertical flow within a well and variability of contaminant concentration in layers within the formation along the screened interval. For all the wells, do you have a sense of whether there is vertical flow in the well and/or vertical contaminant concentrations within the formation?

Its hard to say. Those are possible reasons that the diffusion bag and standard purge samples might not agree, but we'll have to wait and see the results. However, the boring log information indicates that the TSA sandstone unit and the TSA conglomerate unit are both relatively uniform with out significant layers of fine or course material that could cause variations in transmissivity across the screen and vertical flow in the well. Also, placing one diffusion sampler per 20 ft of screen is intended to evaluate potential vertical flow in the well in the event that results for the diffusion bag at the pump depth do not agree well with the purged sample. The diffusion bag to be placed on top of the pump is most likely to have concentrations that match the purged sample.

Is your regular sampling done using a low flow method?

No, historical groundwater sampling has been conducted by Cascade and Boeing using standard purge sampling methods. We feel that switching to low-flow purge sampling for the pilot test would complicate the evaluation by adding another variable that would require consideration when comparing the pilot test results to historical data.

If not, how do you foresee comparing a high volume purge sample with the diffusion sample?

Again, hard to speculate prior to seeing the pilot test results. However, field data collected during historical groundwater sampling events indicate minimal groundwater draw down during standard purge (i.e. compliance with one requirement of the low-flow purge method).

Where you install two diffusion bags in one well, will you be obtaining a regular purge sample at the depth of each of the two diffusion bags?

No, we intend to collect a single purged sample with the pump at the normal depth. The wells planned for multiple diffusion samplers are irrigation wells with heavy pumps not easily raised and lowered. The additional diffusion bags to be placed in these wells are intended to evaluate potential vertical flow, not to match purged sample and diffusion sampler results at multiple levels.

If you use a high volume purge sample method typically, and you find a significant difference in the diffusion sampler data versus the purge sampler data, will you consider a second step of installing a higher density of samplers in a well to better understand the vertical variability to perhaps better use the diffusion sampler method to the advantage of the project?

If we observe a significant difference in analytical results between the diffusion sampler and the conventional sample, we will consider additional investigation to determine if diffusion samplers placed at a different level will better match historic purged sample results. However, our purpose for the pilot test is to show a correspondence between the two sampling methods, not to detail the vertical distribution of contaminants across the well screens.

2. Sampling materials and analysis. You will be using diffusion samplers from two different companies and samples will be analyzed by two different labs. Do you now have samples from the two groups of wells (Boeing wells and Cascade wells) analyzed at the two identified labs? Are the diffusion bags to be obtained from the two labs comparable in size and performance? Will all diffusion bag samplers in the future be analyzed by these same two labs for the same sets of wells, assuming this pilot test is successful?

Historically samples collected by Boeing have been submitted to Analytical Resources Inc. (ARI) and samples collected by Cascade have been submitted to Columbia Analytical Services (CAS). Both Cascade and Boeing plan to continue with the labs they have historically used for each remediation area. The diffusion bags provided by each laboratory will be the same length (24-inches). The CAS bags are 1.25-inches in diameter and the ARI bags are 1.75-inches in diameter. The difference in volume will be inconsequential over the 2-week equilibration period.

3. Knot. I am curious. What type of knot will you use?

Good question; we've all on occasion lost bailers due to faulty knots and certainly don't want to be fishing for the diffusion bags. We will use a Fisherman's knot to attach the monofilament to the diffusion bag and to the well seal; the Fisherman's knot is used for tying fishing line to hooks and lures. The security of all knots will be tested prior to deployment of the bags.

4. Data Evaluation. Will you be doing any kind of statistical analysis, or just a straight comparison of values obtained for each method?

We plan to use the data evaluation methods described in the ITRC Technical and Regulatory Guidance Document (ITRC 2004). We anticipate comparison of diffusion sampler results to historic data and to the split sample results (i.e. standard purge results), as follows.

1) The diffusion sampler results will be compared to the last two years of analytical data for each well to evaluate if the diffusion results fall within or above the range of historic results.

2) The split samples (diffusion sample and purged sample) will be evaluated as duplicate samples per our existing quality assurance plans. Detected concentrations from the split samples will be compared using using relative percent difference (RPD) calculations if the reported concentrations are greater than five times the reporting limit with a 20 percent acceptance criterion. If split sample results are less than five time the reporting limits, then we will evaluate difference between the concentrations as a multiple of the reporting limits. Agreement of split samples may also be evaluated graphically by plotting the split results relative to a 1:1 concentration line to observe scatter.

Those are all the questions that I have. I will put a copy of this email in the file indicating that I have approved the Diffusion Sampler Pilot Test work plan.