Search

Understanding shifts in benthic communities along gradients over time is of crucial importance in estuarine ecology and for conservation. Yet studies considering historical data and both taxonomic and functional aspects are rare. We analysed data of macrobenthic invertebrates of the highly modified Elbe Estuary (Germany) from 1889, 1985, 1986 and 2006 to assess community shifts and considered both taxonomic and functional aspects. The number of species, functional richness (FRic) and functional redundancy (FR) reached minimum values in 1985 and 1986 and were highest and similar in 1889 and 2006. However, beta-diversity analysis showed that species assemblages in 1889 and 2006 greatly differed, almost exclusively owing to species replacement. Most striking was the decline in gastropods after 1889 to levels approaching extinction in the estuary in the other sampling years, most likely due to elevated concentrations of chemical substances, such as tributyl tin. Functional beta diversity was mainly a result of functional nestedness and was exceptionally low in a comparison of data from 1889 to 2006, which indicated nearly complete functional recovery. Null models revealed that function loss in 1985 and 1986 was non-random, consistent with habitat filtering. All functional groups were present on all sampling dates, which suggested that FR might have provided protection from significant ecosystem function loss. Our results indicate that FRic and FR in 1889 could have promoted ecosystem resilience and stability.

The Venice System is one of the best-known and most applied systems to classify waters with respect to salinity. It has often been subject to criticism because the criteria used to derive zone boundaries were not made explicit. Consequently, an alternative multivariate PCA method that aimed at identifying salinity zones by means of the salinity ranges preferred by species was introduced by Bulger et al. (1993). We tested the applicability of both methods using salinity and benthic macroinvertebrate data for the poikilohaline Elbe Estuary (Germany) from 1997 to 2012. This was done by comparing the resultant salinity zone limits from the two approaches with boundaries where significant community changes were found by means of cluster analysis. Only the Venice System polyhaline and limnetic zone boundaries, and the PCA method outer estuary zone limit, showed good agreement with the benthic community pattern. None of the other Venice System or PCA method zone limits reflected the benthic community patterns. Our findings suggest that zone limits can only be well determined from mean salinity at the inner and outer end of the estuary, where the variation of salinity is relatively low. In contrast, in the middle of the estuary variation in salinity is the better predictor of zone boundaries. Thus, application of the Venice System or the PCA method in poikilohaline estuaries, such as the Elbe, is not meaningful and their use should be limited to homoiohaline systems. For poikilohaline systems, we found cluster analysis to be a better tool to identify salinity-zone boundaries.

In the context of the requirements of the European Water Framework Directive (WFD) numerous benthic indices have been proposed recently to assess the ecological quality of marine and transitional waters. In several studies these metrics were applied to coastal and estuarine environments under diverse anthropogenic pressures. Although currently the dumping of dredged material is one of the most prominent human impacts that modify estuaries, the performance of benthic indices to detect effects of dredged sediment relocation has not yet been tested explicitly. Hence we examined a selection of common metrics (species richness, Shannon diversity, AMBI, M-AMBI, W-value, BO2A) at 11 dumping and 11 nearby reference areas in the highly modified Elbe estuary (Germany), where permanent dredging is necessary to maintain the depth of the navigation channel. In order to cover the entire estuary, the study area spanned over the whole salinity gradient from limnic to euhaline. Additionally, we investigated changes in benthic communities due to dredged material placement. All indices, except the W-value, were suitable to differentiate between dumping and reference areas and showed significantly better index values exclusively at reference areas. The applicability of AMBI and M-AMBI was restricted in the limnic stretch due to the more frequent occurrence of freshwater species there. The W-value and BO2A were non-satisfactory in the case of azoic sediment, and in most cases these two indices indicated much better ecological status classifications than the other indices tested. Furthermore, the BO2A had restricted applicability with increasing salinity. At eight of eleven sites the benthic communities differed significantly between dumping and reference areas. Our findings show that the power of conventional benthic indices to detect physical disturbances like the dumping of dredged sediment varies greatly. Having this in mind, we suggest to choose carefully the benthic indices for ecological quality assessments according to the WFD in estuaries in order to avoid misclassifications. Such errors may lead to unnecessarily expensive remediation activities or, in the opposite case, to inactivity although actions were necessary. Furthermore, in order to better meet the WFD requirements we suggest that, regarding frequency and volume, dumping should be adapted as far as possible to the natural processes of sediment movement.