Laserfiche WebLink
Red Creek Quarry Baseline Water Investigation Section 5 <br /> <br /> <br />5-13 <br />DRAFT for review purposes only. Use of contents on this sheet is subject to the limitations specified at the end of this document. <br />according to the data validation protocols in the QAPP (BC 2023a, Appendix 3.6). These data are <br />considered to be qualitative based on the type of data validation issue. <br />Metals/metalloids were analyzed for both total and dissolved concentrations within each sample. <br />Total concentrations reflect the entire amount of the constituent present in an unfiltered sample and <br />reflects both the dissolved and solid (e.g., precipitate or sorbed to particulate) phases in the sample. <br />Dissolved concentrations reflect the amount of the constituent occurring as particles passing <br />through an 0.45-micrometer filter. Dissolved concentrations greater than total concentrations can <br />indicate a procedural issue (e.g., filter contamination, or an inadvertent mis-labeling of total and <br />dissolved samples) however, it is frequently a result of the inherent limit of analytical precision. <br />Instances where metals/metalloids dissolved concentrations are greater than total concentrations <br />were evaluated per acceptance criteria (EPA 2013). The acceptance criteria establish acceptable <br />levels of disagreement between total and dissolved concentrations as a function of concentration <br />relative to the method detection limit. Dissolved concentrations may exceed total concentrations in a <br />sample with a relative percent difference (RPD) of less than 20 percent. For example, if the total and <br />dissolved constituent concentrations are very similar and/or close to the method reporting limit, then <br />the difference in concentration between the total and dissolved fractions may be well within the error <br />limits of the analytical method. <br />For the RCQ baseline dataset, the majority of the instances where the RPD between dissolved and <br />total metal/metalloid concentrations was greater than 20 percent was due to detection limits. One <br />instance where metal/metalloid dissolved concentrations are greater than total concentrations was <br />for the sample collected from the Stilling Well during the first quarterly sampling event <br />(August 2022). The difference between metals/metalloid fractions is likely the result of the sequence <br />in which groundwater sample bottles were filled. The process of bailing can contribute to an increase <br />in turbidity during sample collection. The total metals bottle was likely filled with less turbid water <br />than the field filtered dissolved bottle resulting in heterogeneous concentrations of <br />metals/metalloids. Sampling methods were modified after the first quarterly sampling event to fill <br />the metals bottles from the same bailer of water. <br />Field duplicate samples were collected at a rate of one duplicate for surface water and one duplicate <br />for groundwater for each quarterly sampling event. Both parent and duplicate sample data are <br />maintained in the database, but only parent sample concentrations were used for geochemical plot <br />generation or statistical analyses. Duplicate sample results, agreements, and implications for data <br />usability were evaluated as part of the data validation process and discussed in the individual data <br />validation reports (Appendix E). Variations in the analytical results of duplicate samples occurred for <br />different analytes in each of the five quarterly sampling events. Differences in the duplicate sample <br />precision is likely based on the methods for sample collection resulting in heterogeneity of the <br />samples submitted to the laboratory in separate bottles. Differences in the duplicate sample results <br />indicate the range of concentrations for analytes for the surface and groundwater sources. <br />5.4.2 Summary of Analytical Results <br />Overall summary statistics for groundwater and surface water samples were calculated for each <br />constituent using Microsoft Excel software including the number of analyses conducted, number and <br />percent of samples above the comparative water quality standards, number and percent below the <br />detection limit, maximum and median concentrations (Appendix H). Data reported below the <br />analytical detection limit were included in the calculation of the median value by substituting half the <br />detection limit in the median value calculation. Other methods for calculating summary statistics on <br />censored data set (i.e., containing non-detect values) typically require larger numbers of analyses <br />(Helsel and Hirsch 1992)