Diatom assemblages from Adirondack lakes (New York, USA) and the development of inference models for retrospective environmental assessment

Abstract

Detrended canonical coreespondence analysis (DCCA) was used to examine the relationships between diatom species distributions and environmental variables from 62 drainage lakes in the Adirondack region, New York (USA). The contribution of lakewater pH, Alm (monomeric Al), NH4, maximum depth, Mg, and DOC (dissolved organic carbon) were statistically significant in explaining the patterns of variation in the diatom species composition. Twenty-three and sixteen diatom taxa were identified as potential indicator species for pH and Alm, respectively (i.e. a taxon with a strong statistical relationship to the environmental variable of interest, a well defined optimum, and a narrow tolerance to the variable of interest). Using weighted-averaging regression and calibration, predictive models were developed to infer lakewater pH (r2=0.91), Alm (r2=0.83), DOC (dissolved organic carbon) (r2=0.64), and ANC (acid neutralizing capacity; r2=0.90). These variables are of key importance in understanding watershed acidification processes. These predictive models have been used in the PIRLA-II (Paleoecological Investigation of Recent Lake Acidification-II) project to answer policy-related questions concerning acidification, recovery, and fisheries loss.