Temporal variation in community–environment relationships and stream classifications in benthic diatoms: Implications for bioassessment

Diatom based biomonitoring tools are proven to be a practical way to indicate stream conditions, but only little is known how stable diatom inferred stream classifications are in time. We studied annual variation in diatom community–environment relationships, community structure and diatom indices (Index of Pollution Sensitivity, IPS and Trophic Diatom Index, TDI) during three consecutive years (2010–2012) in four drainage basins distributed in Finland. We also used a cluster analysis to examine if stream classifications resulted in distinct and temporally stable community types. We found only subtle temporal variation in community–environment relationships, nutrient concentrations (N, P) and conductivity consistently being the main factors structuring communities. According to Mantel tests, diatom communities resembled each other significantly at different years, and the values of IPS and TDI indices remained relatively stable in the basins through time. The stream classification based on diatoms also resulted in temporally stable and statistically distinct community types. We thus suggest that sampling of diatoms, e.g., in every three years seems to be a reliable procedure to assess biological water quality. Generally, to choose the correct metrics to assess water quality is essential, and both biological and physicochemical factors should be considered.