Euthrophication Analysis: A Multivariate Approach

Data from 55 Florida lakes were used to establish quantitative relationships between lake trophic state and watershed characteristics. The lakes were grouped by cluster analysis into 24 colored lakes and 31 relatively clear lakes, and these two groups were further subdivided into classical trophic groups by cluster analysis using even trophic state indicators. Trophic State indices (TSI) were developed from correlation matrices of the trophic indicators by principal component analysis to rank the lakes on a quantitative trophic state basis. TSI values for the 55 lakes were regressed against land use and population characteristics in the lake watersheds, and fertilized cropland and other cultural factors were found important in explaining the trophic state values. Canonical analysis of the individual trophic variables against eutrophication factors gave similar results.