EXECUTIVE SUMMARY

 Recently, improvements in analytical chemistry technology have given scientists the ability to measure lower and lower concentrations of metals and other contaminants in water. Similarly, as the understanding of the environmental effects of contaminants has increased, the acceptable concentrations in water of many of these contaminants have decreased. When monitoring water quality, it is preferable to be able to reliably measure concentrations of contaminants which are one tenth of the acceptable concentrations of those contaminants.

The validity of reporting data with very low concentrations of contaminants depends on collecting representative samples and ensuring sample integrity, by taking all necessary steps to ensure that, once collected, samples do not deteriorate and are not contaminated. Prior to analysis the integrity of water samples must be maintained during collection, transportation and storage, through the implementation of appropriate quality assurance practices. Possible contamination may be detected and measured through quality control samples.

Great care must be taken during sample collection to prevent contamination, and once collected they must be preserved to ensure that they do not deteriorate prior to analysis. Chemical preservation, storage temperatures and holding time all play a role in sample preservation. During sample collection, preservation, handling, transportation and storage, samples are at risk of contamination. The major sources of water sample contamination include: sample bottles and caps; preservatives; filters; sampling, filtering and laboratory equipment; poor sampling, handling and storage practices; and airborne contaminants (e.g., dust, fumes). Contamination can be minimized by ensuring that all persons involved, including those in the field and in the laboratory are properly trained, and by ensuring that instructions for collecting, transporting and storing samples are well thought out and clearly documented.

Water sample contamination can be monitored with the use of blanks, such as trip blanks, field blanks, equipment blanks, and filtration blanks.

Once samples reach the laboratory, a key consideration in the interpretation of analytical results for samples with very low concentrations of contaminants is the detection limit or limiting low concentrations below which a particular analyte cannot be detected. There are two distinct detection limits that may be reported – the Method Detection Limit (MDL) and the Reliable Detection Limit (RDL). The Method Detection Limit is the measured response at which there is a stated probability (usually 95 or 99%) that the analyte is present. The Reliable Detection Limit is the lowest analyte concentration required to be present in the sample to ensure detection; i.e., the analytical response that will exceed the MDL with stated probability (usually 95 or 99%). The detection limit in most common usage and that best approximates an industry-standard is the MDL (99%), but the RDL represents the point at which measured values become believable. Another limit, the Limit of Quantitation, or LOQ, provides a further assured level of confidence that data which exceed it are statistically significant.

An understanding of the meaning and significance of the different detection limits can aid in the interpretation of low level data, such as very low concentrations of metal in water. For example:

• if: result < MDL                 then: analyte not detected
• if: MDL ≤ result < RDL    then: analyte is present but result is not statistically significant
• if: RDL ≤ result < LOQ     then: result is borderline statistically significant at the RDL
• if: result ≥ LOQ                   then: result is statistically significant

A comparison of water quality guidelines with the detection limits which are commercially available in Canada shows that, for most parameters of interest in a metal mining context, it is possible to measure contaminant concentrations as low as 1/10 of the water quality guideline for those contaminants. When monitoring water quality, it is preferable to be able to reliably measure concentrations of contaminants which are 1/10 of the acceptable concentrations of those contaminants. However, for several contaminants, technology is not currently commercially available to be able to reliably measure concentrations at such low levels. These contaminants include: arsenic, cadmium, mercury, selenium, silver and cyanide. The technology is available at some laboratories in Canada to measure all but cadmium and mercury at 1/10 of the lowest guideline.

AETE