RESUMEN
A fluorometric screening method was used to estimate total polycyclic aromatic hydrocarbon (t-PAH) concentrations in sediments collected from the St. Louis River Area of Concern (AOC) in northeastern Minnesota. Sediments were collected as part of a Regional Environmental Monitoring and Assessment Program (R-EMAP) study to assess sediment quality in the AOC. The screening method was calibrated using a PAH surrogate standard consisting of eight PAHs commonly found in the St. Louis River system, at their approximate proportions. Estimated PAH concentrations were compared to GC/MS measured 'true' PAH concentrations to evaluate the overall predictive power of the screening method. Regression analysis of log transformed estimated versus true PAH concentration yielded an r2 of 0.72 (n = 86). In addition, the rates of false positive and false negative predictions associated with the screening method were determined relative to different sediment effects concentrations (SECs) for total PAHs. In general, the rate of false positive predictions was shown to increase as the SEC criteria value decreased, while false negative rates remained consistently low (below 7%). Methodological recommendations which led to a three-fold reduction in false negatives, and the improved prediction of both high and low PAH samples, are presented.