Human impacts in Antartic marine soft-sediment assemblages: correlations between multivariate biological patterns and environmental variables at Casey Station

Correlations between the spatial distribution of soft-sediment assemblages and environmental variables were evaluated at a number of control and potentially impacted locations (waste dumps, a sewage outfall and a wharf) at Casey Station, East Antarctica. Patterns of assemblage composition were compared with patterns of environmental variables using univariate and multivariate techniques. The utility of these methods, however, is uncertain in areas of significant habitat heterogeneity (at scales <10m), such as at Casey Station. A number of environmental variables were measured to determine their potential influence on assemblage structure including sediment heavy metal concentrations, total organic carbon (TOC), grain size and depth. Soft-sediment assemblages clustered strongly into two groups and this pattern was found to be most highly correlated with concentrations of heavy metals of anthropogenic origin: cadmium, copper, lead, tin and zinc. The relationship, however, is complex and is likely to be further influenced by variables affecting the bioavailability of metals such as grain size, TOC and sediment anoxia. Impacted locations were characterised by fewer taxa, lower diversity and lower species richness. One of the control locations was found to have naturally high levels of cadmium, nickel and zinc and had assemblages very similar to those at the potentially impacted, polluted locations. One of the potentially impacted locations had coarse sediments and low levels of heavy metals but had assemblages very similar to other impacted locations. Problems of sampling design for human impacts detection in real world situations are discussed in relation to this study. This study provides evidence that these multivariate statistical methods are useful in heterogeneous environments and across pollution gradients where pollutants have the potential to act as a primary cause of spatial variation in assemblage structure.