Benthic diatom communities of coastal marine environments

This project examined the effects of ecological factors and anthropogenic contaminants on benthic diatom communities near Casey Station, in the Windmill Islands, Antarctica. Preliminary sampling indicated significant compositional differences occurred between diatom communities in bays immediately adjacent to Casey Station and those more distant. Subsequently, a more detailed appraisal of spatial variability in benthic diatom communities was undertaken. Significant differences in community compositi9n were demonstrated for all scales examined, however, increased distances between samples typically corresponded to increased dissimilarity. The influence of environmental factors on spatial variability was assessed using direct gradient ordination techniques. Grainsize, particularly the mud content, accounted for the majority of variation in diatom abundances explained by the measured environmental variables. Differences in light availability, water depth and grain-size explained 60% of the variation in community composition observed between locations. The remaining 40% of the variation in diatom community composition remains unexplained; potential causes include freshwater input, and chemical contamination. A preliminary analysis of temporal variability in benthic diatom communities from the Windmill Islands is presented. Seasonal, short-term (lOO's ofyears) and longterm (1000's ofyears) changes in community composition were examined at several locations. Temporal variability of the diatom communities within Brown Bay was also assessed, in conjunction with metal and Pb data. Within one core, a shift in community composition was detected subsequent to the onset of chemical contamination. Some changes in species abundance exceeded the natural variability observed in control cores, and thus were attributable to chemical contamination. Relationships between diatom abundances and concentrations of different metals within Brown Bay were examined using direct gradient analyses. Approximately 50% of the variation in the diatom community composition was related to metal concentrations. Further clarification of the interactions between metals, sediment grain-size and diatom community composition would be required prior to the development of predictive models. Direct gradient analyses also demonstrated a weak correlation between diatom community composition and hydrocarbon distribution. Manipulative field experiments were used to demonstrate that the composition of diatom communities can be influenc.ed by both metal and hydrocarbon contaminants at concentrations comparable to levels produced by station activities. This supports the conclusion that anthropogenic contaminants are responsible for some compositional differences observed in diatom communities, both within Brown Bay and between Brown Bay and reference locations.