GEC Comments on the Arco Sampling Plan

 June 2, 2004

Shea Jones, Remedial Project Manager
EPA Superfund Branch
61 Forsyth Street, SW, 11^th Floor
Atlanta, Georgia  30303-3104

Re: Comments on Arco neighborhood sampling, Draft Plan for Off-site Soil Sampling, Chemicals Superfund Site, Brunswick, Glynn County, Georgia; prepared by GeoSyntec Consultants, and dated May 7, 2004, for potential contamination from the LCP Chemicals Superfund Site, Brunswick, Glynn County, Georgia.

Dear Ms. Jones, 

            The Glynn Environmental Coalition (GEC) submits the following comments concerning the Draft Plan for Off-site Soil Sampling, Chemicals Superfund Site, Brunswick, Glynn County, Georgia; prepared by GeoSyntec Consultants, and dated May 7, 2004.  The Glynn Environmental Coalition is in agreement with the comments of Dr. R. Kevin Pegg, Technical Advisor for the LCP chemicals Superfund Site, that have been submitted in: 1.) LCP Chemicals Superfund Site, Technical Assistance Report.  LCP Chemicals ARCO Neighborhood Testing Plan, Volume 9, Number 1, May 2004; and, 2.) LCP Chemicals Superfund Site, Technical Assistance Report Summary.  LCP Chemicals ARCO Neighborhood Testing Plan, Volume 9, Number 1, May 2004, published in The Brunswick News June 2, 2004.  The comments by Dr. R. Kevin Pegg are included as part of the GEC’s comments by reference.

 General Comments

 1.         The GEC received the Draft Plan for Off-site Soil Sampling in the Arco neighborhood, on Saturday, May 16, 2004, from the U. S. Environmental Protection Agency (EPA) with a request to submit comments by June 1, 2004.  The following activities were initiated to facilitate community comment on the sampling plan by the EPA requested comment submittal date. 

A. Review of the Sampling Plan was started by Dr. R. Kevin Pegg, our technical advisor under the U. S. Environmental Protection Agency’s Technical Assistance Grant program.  A technical assistance report was prepared for the Arco community meeting.

B.  An Arco Community meeting was planned for Saturday, May 22, 2004, and the following were invited to present information about the Sampling Plan at the meeting.
 

- Dr. R. Kevin Pegg, Technical Advisor

- David Mellard, Toxicologist, Agency for Toxic Substance and Disease Registry

- Gary Hummel, Glynn County Health Department

- Rob Asbell, Aid to Congressman Kingston

- Shea Jones, LCP Chemicals Superfund Site Remedial Project Manager, U. S. Environmental Protection Agency

- Angela Miller, Community Relations Coordinator, U. S. Environmental Protection Agency 

C. Meeting notices were distributed door to door, mailed on cards, posted on the GEC web site, and sent to the print and broadcast media.

D. Enlargement of the Sampling Plan figures was produced to facilitate discussion of the sampling plan during the community meeting. 

2.         The Arco community meeting was held on May 22, 2004 at the Arco United Methodist Church social hall, which is located in the Arco neighborhood.

A. With the exception of Shea Jones, LCP Chemicals Superfund Site Remedial Project Manager, U. S. Environmental Protection Agency; and Angela Miller, Community Relations Coordinator, U. S. Environmental Protection Agency, all invited presenters attended the community meeting.

B.  David Mellard, Toxicologist, Agency for Toxic Substance and Disease Registry (ATSDR) was asked if the Sampling Plan would produce data needed for him to complete a Public Health Assessment (PHA).  Mr. Mellard stated that the Sampling Plan was insufficient to produce the data needed for a public health assessment.  All in attendance agreed that at a minimum, the Sampling Plan should produce the data needed for a PHA, and resolve questions about potential health risks from LCP Chemicals contamination of the Arco neighborhood.  A sampling grid for each block was recommended by the health professionals present, and Dr. Pegg.

C.  Concerns were expressed by residents about consumption of very large mullet from the small lake in the southwest corner of the Arco neighborhood.  All in attendance agreed that fish from the pond need to be tested for chemical contamination since local fishers consume them.

            D.  The Sampling Plan failed to include significant portions of the Arco neighborhood in both Phase 1 and Phase 2 sampling.  Of particular concern was the Coastal Academy School, which was not included in the Sampling Plan. 

Section 1. Introduction           

3.         Section 1. Introduction - The introduction references a request from the EPA Remedial Project Manager (RPM) for additional testing of the LCP Chemicals Superfund Site (Site) and portions of the nearby residential neighborhood.        

            A. Please provide the GEC a copy of the referenced request from the EPA to the Responsible Parties (RPs) to develop a sampling plan for a portion of the residential areas next to the Site.

            B. The stated purpose of the Draft Sampling Plan, sampling only a portion of the Arco neighborhood, is far different than the understanding the GEC had of the purpose of the Arco neighborhood sampling plan.  It is the understanding of the GEC that the sampling plan purpose is to produce data sufficient in quantity and quality for an Agency for Toxic Substance and Disease Registry (ATSDR) Toxicologist to produce a Public Health Assessment (PHA) with a determination on health risks from potential chemical contamination of the Arco neighborhood during the operational history of the Site.  The purpose of the sampling plan should be clearly stated as producing data of sufficient quantity and quality to produce a PHA with a determination of potential health risks from the Site.

            C. The introduction is correct in identifying the need for additional sampling as part of Operational Unit One (OU1) Remedial Investigation to identify the extent of contamination, both on and off-site, from releases from the Site.  The Sampling Plan should clearly state that the (OU1) Remedial Investigation (RI) is incomplete, particularly for areas outside the current property boundaries of the Site. 

            D.  The property boundaries during the operational history of the Site have not been determined.  The Final Title Search Report, November 17, 1997, (Administrative Record 11.14) only tracks ownership of land owned by LCP Chemicals at the time the company stopped operations.  CERCLA requires the boundaries of the Site during its operational history be defined, which the Title Search failed to do. The sampling plan should refrain from referencing “boundaries” of the Site, which remain to be determined by a completed OU1 RI and proper CERCLA title search.  The GEC is aware of deed references to property owned by Atlantic Refinery Company that is outside currently described Site property boundaries. 

Section 2.1 Purpose of the Study 

4.         Section 2.1 Purpose of the Study – The stated purpose of the study contradicts the purpose stated in the introduction.

            A.  The stated purpose of the study is different than stated in the introduction. “…EPA has determined that additional sampling is necessary to verify the initial results and to confirm the previous conclusions.”  The purpose states the conclusion that the study desires to find, and directs the parties to design a study that will come to a predetermined conclusion.  The purpose of the study should clearly state that the purpose is to complete the Remedial Investigation and produce data of sufficient quantity and quality for the ATSDR toxicologist to produce a PHA that will determine if there are health risks to residents, businesses, and school children attending Coastal Academy School in the Arco neighborhood.

            B.  Please provide a copy of the “EPA determination” referenced in the sampling plan stating that additional sampling is necessary to verify the initial results and to confirm the previous conclusions.

            C.  Repeatedly and throughout the Sampling Plan, the conclusions of the study are stated even before the study takes place.  Clearly and inequitably, the study’s purpose is to, “…verify the initial results and to confirm the previous conclusions.”   

Section 2.5 Principal Study Question and Decision Statement 

5.         Section 2.5 Principal Study Question and Decision Statement – This section appears to present a third purpose of the sampling plan, which is “… to determine whether the Site boundaries should extend beyond the LCP property bounds.”

            A.  The purpose of the sampling plan is not clear.  What is clear is that the sampling plan is NOT designed to produce data of sufficient quantity and quality for an ATSDR toxicologist to produce a PHA.

            B.  The purpose of the sampling plan appears to be, among other things, to test the Site to determine if sampling should take place off-site. The Purpose and Principal Study Question and Decision Statement should clearly state that the purpose of the sampling plan is to produce data for a PHA that will determine if there are health risks in the Arco neighborhood from Site contamination. 

Section 3.2 Proposed Sampling Locations and Depth 

6.         Section 3.2 Proposed Sampling Locations and Depth – The number and location of sampling points are grossly biased to the point of incompetence.  

            A. Even if Phase 1 and Phase 2 sampling were to occur concurrently, there would not be sufficient data to complete a PHA.  There are numerous problems and biases with the selected sampling points.

            B. Of the 14 samples in Phase 1 sampling, 5 are located on the Site, 6 on the railroad tracks, 1 in the woods, and 2 in the Arco neighborhood.  Of the 2 samples in the Arco neighborhood, one appears to be located on at the Georgia-Pacific Credit Union.  One sample in the Arco neighborhood will be insufficient to produce a PHA.

            C. The Phase 1 sampling is sufficiently biased to produce the desired result of the study, which is to, “…verify the initial results and to confirm the previous conclusions.”  Completion of Phase 2 sampling will not change conclusions of Phase 1 sampling.  In Phase 2, there are only 7 samples total.  Of the seven samples, only 4 are located in the Arco neighborhood.  Combining both Phase 1 and Phase 2 samples from the Arco neighborhood would leave a total of 5 samples from the Arco neighborhood, which comprises ~ 60 acres.  Five samples is not sufficient to produce a PHA in a neighborhood that covers ~60 acres.

            D.  At a minimum, each block in the Arco neighborhood should be divided into eight sampling grids.  There are 11 blocks in the Arco neighborhood, which would include the Coastal Academy School.  There might need to be even more grids on each block, if the ATSDR toxicologist needs a greater sample density to make a determination about potential health risks from the Site.

            E.  The Sampling Plan has located at least 13 samples in locations where they can be influenced by activities associated with railroad track construction, operation, and maintenance.  At a minimum, elevated levels of PAHs would be expected in railroad areas.  Placement of over half the sampling points in areas likely to be influenced by railroad activities indicates little thought was put into the sampling plan, or the sampling plan was designed with intended bias in mind.   To minimize bias in the sampling results, sampling locations should be located away from railroads (PHAs), buildings (lead based paint), and historical parking areas (leaded gas).  Locating 8 grids on each block and taking samples from interior locations would tend to be the least bias sampling plan and produce the most objective results.  The ATSDR toxicologist should be consulted to confirm that sampling locations are most representatives of actual exposure scenarios in the Arco area. 

Section 3.3 Sampling Constraints 

7. Section 3.3 Sampling Constraints – The sampling plan appears to argue that it is insufficient in data quantity and quality to produce a PHA. 

A. The authors state, “It should be noted that the sampling design is non-probabilistic in nature and does not lend itself to rigorous statistical evaluations methods.”  The study design clearly is insufficient to produce the data for a PHA that will utilize rigorous statistical evaluation methods to determine potential health risks.            

Sincerely, 

Frank Lea, President