Da. Cancilla et Xc. Fang, EVALUATION AND QUALITY-CONTROL OF ENVIRONMENTAL ANALYTICAL DATA FROM THE NIAGARA RIVER USING MULTIPLE CHEMOMETRIC METHODS, Journal of Great Lakes research, 22(2), 1996, pp. 241-253
The use of artificial neural networks (ANN), principal component analy
sis (PCA) and universal process modelling (UPM) to identify the source
of water samples based on the variation of chemical data from these s
amples has been investigated. Chromatographic data sets generated from
three locations on the Niagara River were used in this research. The
concentrations of target organic compounds were chromatographically de
termined and used as classification features. Chromatographic variatio
n between three sampling sites was determined over a one-year period a
nd included 149 separate samples. Variation within sampling sites was
evaluated otter a seven-year period. ANN and UPM techniques correctly
identified the source of 95% of the water samples based on minor diffe
rences in the chromatographic data. PCA and UPM gave direct visualizat
ion of differences within chemical data sets. PCA and UPM were also fo
und to be useful tools for the detection of chromatographic outliers f
rom within sampling sires. The correlation between target compounds an
d surrogates are discussed. The results show that these methods are us
eful for the determination of the variation of target organic compound
s over time both within and between sampling sires. The potential of t
hese systems for monitoring analytical quality control based on entire
data sets is also presented.