Weak concordance between fish and macroinvertebrates in Mediterranean streams.

Stefano Larsen,Giorgio Pace,Laura Mancini,Massimiliano Scalici,Lorenzo Tancioni,Martin Solan

doi:10.1371/journal.pone.0051115

Abstract

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.

Highlights

Human activities have long impaired the natural dynamics of biotic communities in inland waters systems both directly, for example via hydromorphological alteration, pollution, and introduced species, and indirectly via modification of river catchment from agriculture and urbanization [1,2,3,4]
In EU Countries, use of the multi-assemblage approach has become an official policy since the Water Framework Directive (WFD) [22] required the classification of river ecological status using four biotic elements as indicators
The percentage of explained variation for of each principal component is given in parentheses

Summary

Introduction

Human activities have long impaired the natural dynamics of biotic communities in inland waters systems both directly, for example via hydromorphological alteration, pollution, and introduced species, and indirectly via modification of river catchment from agriculture and urbanization [1,2,3,4]. In EU Countries, use of the multi-assemblage approach has become an official policy since the Water Framework Directive (WFD) [22] required the classification of river ecological status using four biotic elements as indicators (diatoms, macrophytes, macroinvertebrates and fish). The rational behind this approach is based in the concept of indicator or surrogate communities, which are expected to be representative of other taxa as well [23,24]

Methods

Results

Discussion

Conclusion