5.3.5 Collection of Field Replicate ISM Samples

Replicate ISM samples (triplicates or more) should be taken to quantify uncertainty in the estimate of the mean concentration within the DU.In the field, replicate incremental samples (three or more) should be taken to ensure reliable estimates of the mean concentration within the DU. The number of replicates and frequency of taking replicate incremental samples should be specified in the SAP and comply with project DQOs.

To statistically evaluate sampling precision for each DU, additional completely separate replicate ISM samples are collected. The increments are collected in simple random, systematic random, or random within grid locations within the DU that are different from those used for the initial ISM sample. ISM field replicates are made of the same number of increments collected in the initial ISM sample and collected using the same sampling pattern from within the same DU. The replicate samples are prepared and analyzed in the same manner as the initial sample. Three replicate samples (i.e., the initial ISM sample plus two additional samples) should be considered the minimum. In some cases, more replicates may be necessary to reduce data variability and/or to calculate a 95% UCL of the mean that is closer to the actual mean of the DU. Section discusses the statistical basis and evaluation of replicate ISM samples.

When sampling in a systematic random sampling pattern, the increments for an ISM replicate sample are generally collected along the same approximate directional lines established through the DU for the initial ISM sample. Increment locations for ISM replicate samples differ from each other by the selection of different random starting locations on the first line/row of the DU and continuing to sample at this different random interval throughout the DU for each replicate (see Figure 5-5). Thus, the increments for ISM replicates should not be collected from the same locations or colocated with those used for the initial ISM sample. When using the random sampling within grid pattern, replicates are constructed from increments taken from different, randomly selected locations within each gridded area. With simple random sampling, three sets of random locations across the DU are selected and increments collected for each set are used to create the replicates. Replicate ISM samples should be submitted to the laboratory as “blind” samples, meaning the laboratory does not know they are replicate samples of the initial ISM samples.

If only one DU is being investigated, a minimum of three replicate samples should be collected to provide a measure of variability. For sites with multiple similar DUs, “batch” type replicates may be a consideration; for example, three replicates in one DU could be used to provide an estimate of variability that is extrapolated to a number of similar DUs (similar to how labs use batch replicates for determining lab analysis precision). Each site and/or project is unique in terms of numbers of DUs and how similar these DUs are, so decisions on numbers of replicates are unique to each site and should be addressed clearly in the SAP. For the batch type of replicates to apply, each DU in the “batch” should have a similar CSM, including the same soil type, site use/history, contaminant deposition, etc. If considered, this batch approach must be discussed, clearly documented, and agreed to by all parties involved during the systematic planning process (see Section 4.4.2).

Section 7 discusses how replicate ISM sample data are used to assess sampling error and make decisions.