It is vital to demonstrate the usability and reliability of field-based measurement system results in a Triad-based project. Establishing usability/reliability often takes place in two distinctly different phases.

The first phase consists of a demonstration of methods applicability, if required. A demonstration of methods applicability typically involves a pilot study to evaluate the performance of a proposed set of methods under site-specific operational conditions. A demonstration of methods applicability is warranted if there is either insufficient certainty about site-specific method performance, and/or if there is a need to "tweak" method protocols to make sure they are optimized for site-specific needs. A demonstration of methods applicability can be a stand-alone study, or it can be woven into pre-planned characterization activities. As it names implies, the primary purpose of the first phase is to ensure that all involved are comfortable with the proposed set of methods and their associated operating procedures.

The second phase is during implementation of a field-based measurement system and is synonymous with implementing appropriate QA/QC protocols. It differs from a standard laboratory QA/QC program, however, in that there is typically a heavier emphasis on using information from sample splits to help guide the interpretation of data produced by less traditional methods, with the split being analyzed by both the field-based method and a standard laboratory method. However, the splits are not used as the sole mechanism to establish the performance of the field method. Sufficient in-field method QC is expected to independently establish that the method and the operator are performing within expected limits during field implementation. Splits are generally performed (after sufficient attention to homogenization and sub-sample support) for the purposes of demonstrating the comparability between data sets produced by the field method and data produced by more traditional methods. Demonstrating method comparability provides confidence that decisions made on field-generated data are correct. Split samples may be selected randomly in certain situations. However, splits are usually selected for the purpose of managing some aspect of decision uncertainty (e.g., concern about detection limits, interference effects, or changing relationships between indicator parameters being measured and true parameters of concern), so there will be a specific rationale for choosing samples for split analyses.

When reviewing the results from split analyses, it is important to remember that the purpose is not to first of all establish a quantitative correlation between a field-based technique’s result and a standard analysis result (although that may be an outcome). It is to establish that the field-based method is providing information useful to the decision that needs to be made.