Laserfiche WebLink
• <br /> 6. Terminology <br /> Previous documents referred to the shallowest deposits as "Above Upper Horizon." <br /> This document appears to introduce the term "Shallow Unit" (an improvement) in . <br /> Paragraph 2.3.1.0.1 (p. 2-10). If these are synonymous and the latter is now preferred, <br /> "Shallow Unit" should be used consistently to avoid confusion. For example, Figure <br /> 2-7 still shows "Above Upper Horizon." <br /> • Please use consistent terminology throughout the report. "Above Upper <br /> Horizon" should be replaced with "Shallow Unit", if both are synonymous and <br /> the latter is preferred. <br /> 7. Lack of statistical approach <br /> The initial sampling phase was designed to determine the location of SWMUs with <br /> contamination above background. During the Phase I RI, a simple random sampling <br /> approach was used. One important factor that determines the number of samples <br /> required is the expected variability and distribution of contaminant concentrations at <br /> each SWMU. Section 3.1.2.3.6 of the Comprehensive RDFS Work Plan (Work Plan) <br /> presents three types of data scenarios and conclusions (based on variability <br /> assumptions) that may be drawn from the Phase I sampling results. Based on <br /> assumptions of variability at each SWMU, the number of samples was calculated and <br /> presented in the Work Plan. After the Phase I RI was completed, the data should have <br /> been evaluated to determine the validity of the assumption, as explained in the Work <br /> Plan, Section 3.1.2.3.7. However, the Phase I RI Report does not contain such an <br /> analysis to determine if assumptions were valid at each site. If assumptions were <br /> determined to be invalid, additional sampling may be needed. <br /> The statistical evaluation of the results is particularly important for those SWMU sites <br /> that are not recommended for further Phase lI investigation and which cover a <br /> relatively large surface area. These SWMU sites are 2 (North Pond), 7, 9, and 30. <br /> SWMU 25 (Boundary Road) should also be included in this category because of its <br /> length. Without a statistical analysis on the Phase I RI results, the recommendation of <br /> no further investigation on these sites is unsubstantiated. <br /> • Statistical evaluation should be performed on the sampling results of each site <br /> and the results presented in the Phase I Report. Recommendations for <br /> sampling or no finther sampling should take into consideration this evaluation. <br /> EPA 14FEB94 310 <br />