Detecting anthropogenic stress in an ecosystem: 2. Macrofauna in a sewage gradient

Previous work with the meiofauna in the Ria Formosa Lagoon has shown that it is possible to use the community structure as an indicator of recent or ongoing contamination events. However, the meiofauna require a substantial degree of taxonomic expertise to identify to the species level, whereas the macrofauna are much more readily identified. In a parallel study, the macrofauna at the same sites as the previous work were identified and evaluated with univariate (diversity, evenness measures) and multivariate statistical methods (PCA). Unlike in the previous study, the diversity decreased as the grain surface area increased and the sediments became muddier. However, there was no change in the evenness of the macrofauna across the same sites. The number of species also reflected the change from sandy (11-17 per site) to muddy (only 1 species present) sediments. PCA identified the grain surface area as the major controlling factor in explaining the variance, although there was a nonstochastic effect with positive values of principal component (PC) 1 only present in samples with a low TOC (<2%) or <10% of grains less than 63 mum diameter. These organisms were suspension feeders as opposed to deposit feeders that loaded negatively on PC1. The second PC characterized a "contamination" gradient with species that have been identified as indicators of pollution. However, these sites did not correlate with the known sewage gradients of the region and suggest the community was responding to another type of contamination gradient. In conclusion, while the macrofauna were easier to identify and quantify, they do not lend themselves as well as the meiofauna to anthropogenic stress identification.