Premium
A Multimetric Macroinvertebrate Index for the Implementation of the European Water Framework Directive in French Guiana, East Amazonia
Author(s) -
Dedieu N.,
Clavier S.,
Vigouroux R.,
Cerdan P.,
Céréghino R.
Publication year - 2016
Publication title -
river research and applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.679
H-Index - 94
eISSN - 1535-1467
pISSN - 1535-1459
DOI - 10.1002/rra.2874
Subject(s) - water framework directive , species richness , amazon rainforest , diversity index , context (archaeology) , geography , metric (unit) , ecology , index (typography) , environmental science , environmental resource management , water quality , computer science , biology , operations management , archaeology , world wide web , economics
Neotropical, overseas regions of Europe are subjected to the same water policy objectives as the continental ones but were overlooked during recent developments of bioindicators that fulfil the Water Framework Directive guidelines. We designed a macroinvertebrate‐based multimetric index [ Indice Biotique Macroinvertébrés de Guyane (IBMG)] to assess ecological health in remote headwater‐small streams of French Guiana, Europe's only overseas region of continental South America. Invertebrates were sampled at 95 sites including reference and impacted river reaches, following a standardized protocol. Among the 102 biological metrics calculated from site‐specific data, we selected metrics exhibiting the best trade‐off between high discrimination efficiency, low specificity, low redundancy and high stability under reference conditions. The IBMG is composed of two taxonomic richness‐based metrics, two abundance‐based metrics, one trait‐related metric and a diversity index (Shannon's entropy). Each metric was weighted by its discrimination efficiency. Using a test data set, we found that the IBMG was sensitive to the range of disturbances in French Guiana. Finally, comparing the IBMG with other indices developed in other neotropical countries reveals that, for several reasons, multimetric indices developed in the neotropics may perform well in the context of the data sets used to generate them but would certainly fail to be robust when used elsewhere. Copyright © 2015 John Wiley & Sons, Ltd.