PLS and PCR methods in the assessment of coastal water quality

Partial least squares regression analysis (PLS) and principal component regression analysis (PCR) were examined as methodological procedures for assessing the quality of coastal waters in a tourist area. Four variables related to the trophic state of waters, namely nitrate, nitrite, ammonia and phosphate were analyzed. The models resulting from PLS and PCR were very similar. Both defined three groups of water masses characterized by different nutrient loadings. These groups were in accordance with those obtained by numerical classification. The PLS method was selected as the optimal model, on the basis of its lower prediction errors (lower Press and Rmsd values). For management purposes, this statistical model allows mesotrophic conditions, reflecting some nutrient enrichment over background conditions, to be characterized and the successful diagnosis of additional samples within this context.