Phosphorus Concentrations in Stream-Water and Reference Samples - An Assessment of Laboratory Comparability

In 2003, a study was conducted to evaluate the accuracy and precision of 10 laboratories that analyze water-quality samples for phosphorus concentrations in the Catskill Mountain region of New York State. Many environmental studies in this region rely on data from these different laboratories for water-quality analyses, and the data may be used in watershed modeling and management decisions. Therefore, it is important to determine whether the data reported by these laboratories are of comparable accuracy and precision. Each laboratory was sent 12 samples for triplicate analysis for total phosphorus, total dissolved phosphorus, and soluble reactive phosphorus. Eight of these laboratories reported results that met comparability criteria for all samples; the remaining two laboratories met comparability criteria for only about half of the analyses. Neither the analytical method used nor the sample concentration ranges appeared to affect the comparability of results. The laboratories whose results were comparable gave consistently comparable results throughout the concentration range analyzed, and the differences among methods did not diminish comparability. All laboratories had high data precision as indicated by sample triplicate results. In addition, the laboratories consistently reported total phosphorus values greater than total dissolved phosphorus values, and total dissolved phosphorus values greater than soluble reactive phosphorus values, as would be expected. The results of this study emphasize the importance of regular laboratory participation in sample-exchange programs.