Freshwater Releases Into Estuarine Wetlands Change the Determinants of Benthic Invertebrate Metacommunity Structure

In recent years, the relative importance of the processes driving metacommunity composition has aroused extensive attention and become a powerful approach to identify community patterns and their regulatory mechanisms. We investigated variations in the composition of benthic community in restored wetlands and natural wetlands in the Yellow River Delta (Shandong Province, China). First, spatial structures within each wetland were modeled with Moran eigenvector maps. Next, the variation in community structure among local environmental and spatial variables was partitioned using constrained ordination, and the “elements of metacommunity structure” analysis was used to determine the patterns of best fit for species distributions within metacommunities. Finally, the null model was used to analyze non-random patterns of species co-occurrence. The community structure of benthic invertebrates in restored wetlands and natural wetlands differed significantly. The benthic invertebrate metacommunity structure showed a nested distribution in restored wetlands and a quasi-Clementsian structure in natural wetlands. Pure environmental fractions and pure spatial fractions were critical in regulating benthic invertebrate metacommunities of restored wetlands. In natural wetlands, pure spatial fractions and the interaction between environmental and spatial factors (shared fractions) played a major role in the metacommunity. A species co-occurrence analysis showed that species co-occurred more frequently than expected by chance, demonstrating that biotic interactions were not the main driver of metacommunity structures in both wetland types. Accordingly, the benthic invertebrate metacommunity in estuarine wetlands following freshwater releases was mostly determined by environmental and spatial effects, which resulted in a metacommunity with nested distribution. These results are important for biodiversity protection and ecosystem management of estuarine wetlands in the Yellow River Delta.

A comparative analysis of metacommunity types in the freshwater realm

Most metacommunity studies have taken a direct mechanistic approach, aiming to model the effects of local and regional processes on local communities within a metacommunity. An alternative approach is to focus on emergent patterns at the metacommunity level through applying the elements of metacommunity structure (EMS; Oikos, 97, 2002, 237) analysis. The EMS approach has very rarely been applied in the context of a comparative analysis of metacommunity types of main microbial, plant, and animal groups. Furthermore, to our knowledge, no study has associated metacommunity types with their potential ecological correlates in the freshwater realm. We assembled data for 45 freshwater metacommunities, incorporating biologically highly disparate organismal groups (i.e., bacteria, algae, macrophytes, invertebrates, and fish). We first examined ecological correlates (e.g., matrix properties, beta diversity, and average characteristics of a metacommunity, including body size, trophic group, ecosystem type, life form, and dispersal mode) of the three elements of metacommunity structure (i.e., coherence, turnover, and boundary clumping). Second, based on those three elements, we determined which metacommunity types prevailed in freshwater systems and which ecological correlates best discriminated among the observed metacommunity types. We found that the three elements of metacommunity structure were not strongly related to the ecological correlates, except that turnover was positively related to beta diversity. We observed six metacommunity types. The most common were Clementsian and quasi-nested metacommunity types, whereas Random, quasi-Clementsian, Gleasonian, and quasi-Gleasonian types were less common. These six metacommunity types were best discriminated by beta diversity and the first axis of metacommunity ecological traits, ranging from metacommunities of producer organisms occurring in streams to those of large predatory organisms occurring in lakes. Our results showed that focusing on the emergent properties of multiple metacommunities provides information additional to that obtained in studies examining variation in local community structure within a metacommunity.

Weak concordance between fish and macroinvertebrates in Mediterranean streams

Although anthropogenic degradation of riverine systems stimulated a multi-taxon bioassessment of their ecological integrity in EU countries, specific responses of different taxonomic groups to human pressure are poorly investigated in Mediterranean rivers. Here, we assess if richness and composition of macroinvertebrate and fish assemblages show concordant variation along a gradient of anthropogenic pressure in 31 reaches across 13 wadeable streams in central Italy. Fish and invertebrate taxonomic richness was not correlated across sites. However, Mantel test showed that the two groups were significantly, albeit weakly, correlated even after statistically controlling for the effect of environmental variables and site proximity. Variance partitioning with partial Canonical Correspondence Analysis showed that the assemblages of the two groups were influenced by different set of environmental drivers: invertebrates were influenced by water organic content, channel and substratum features, while fish were related to stream temperature (mirroring elevation) and local land-use. Variance partitioning revealed the importance of biotic interactions between the two groups as a possible mechanisms determining concordance. Although significant, the congruence between the groups was weak, indicating that they should not be used as surrogate of each other for environmental assessments in these Mediterranean catchments. Indeed, both richness and patterns in nestedness (i.e. where depauperate locations host only a subset of taxa found in richer locations) appeared influenced by different environmental drivers suggesting that the observed concordance did not result from a co-loss of taxa along similar environmental gradients. As fish and macroinvertebrates appeared sensitive to different environmental factors, we argue that monitoring programmes should consider a multi-assemblage assessment, as also required by the Water Framework Directive.