Ordination of zooplankton community data to detect pesticide effects in pond enclosures

Trends in community structure of crustacean zooplankton among 17 forest pond enclosures (5 m x 5 m x 1 m deep) were examined and analyzed to determine effects of two candidate forest pesticides. Eight enclosures were treated at two concentrations of an experimental insecticide, tebufenozide; five were treated at two concentrations of a biological herbicide, bialaphos; and four served as controls. Zooplankton community structure was characterized by ordination of species assemblages using principle components analysis (PCA) and correspondence analysis (CA). Two-dimensional plots of the first two axes from PCA and CA were constructed to explore temporal and treatment-related patterns in community structure. The first four axes of both ordination functions were used as multivariate descriptors of community structure and were examined for differences among treatments by ANOVA. Species ordination plots and sorted rotated factor loadings provided objective means of identifying the species important in determining divergence in community structure. These were considered indicator species and were examined for differences among treatments by ANOVA and specified contrasts. The ordination and subsequent analysis revealed trends in community structure and sample differences that indicated clear, concentration-dependent effects of bialaphos and equivocal effects of tebufenozide.