Inland Waters, Vol 4, No 4 (2014)

Using multivariate analysis and stable isotopes to assess the effects of substrate type on phytobenthos communities

Tatenda Dalu, Nicole Bertine Richoux, Pierre William Froneman
Pages: 397-412


For more than a century, artificial substrates have been employed in phytobenthos studies. In the present study, we compared the phytobenthos community structure in a field experiment over 3 seasons (summer, autumn, and winter) on 3 types of artificial substrates (brick, brown clay tiles, and grey clay tiles) and 3 natural substrates (macrophytes, rocks, and sediment) in a small, temperate system. A combination of multivariate analyses (cluster, multi-response permutation procedure, indicator species [IndVal], and canonical correspondence analysis [CCA]) and stable isotope analysis was used. We identified 96 total phytobenthos taxa. Artificial substrates resulted in different substrate communities, as shown by stable isotope analysis, cluster analysis, and a multi-response permutation procedure, with only those communities growing on grey tiles being similar to natural substrate communities. Overall, artificial substrates exhibited slightly higher species richness compared to natural substrates over the 3 seasons, although there were no significant differences (p > 0.05). Phytobenthos grown on brown tiles, rocks, and bricks showed seasonal variability of the carbon isotope δ13C values using one-way ANOVA (p < 0.05). Phytobenthos community structure did not show great seasonal variation; however, CCA identified water flow, conductivity, ammonium, phosphate, and water depth as important in structuring phytobenthos communities on different substrates. IndVal analysis showed that common phytobenthos taxa were not restricted to a single substrate, but preference was generally high for natural substrate, especially rocks, compared to artificial substrates. Substrate microhabitat type seems to influence the communities within the study areas.