Using multiple taxonomic groups to index the ecological condition of lakes

Biological indicators of communities typically reflect a common environmental signal reflecting the general condition of the ecosystem, as well as individual signals by indicators differentially sensitive to particular environmental conditions. We describe here a method of integrating and interpreting such indicators from 19 New England lakes for five taxonomic groups (diatoms, benthos, zooplankton, fish, and birds). Our approach provides a systematic standardized way to integrate multiple metrics from different taxonomic groups by addressing four elements crucial to analyzing data from multiple indicators: covariate control, re-scaling of data, standardizing the sign of responses, and dimensional reduction. We evaluated the biological metrics against individual environmental stressors and against multivariate physicochemical metrics characterizing general anthropogenic stress among the lakes. The method detected a response to variation in the gross environmental condition of the lakes that was correlated across taxa and metrics. In addition, a differential response to near shore conditions was demonstrated for fish. The success of the approach in this study lends support to its general application to ecological monitoring involving complex data sets.