Philippe Richoux

Biomonitoring through biological traits of benthic macroinvertebrates: how to use species trait databases?

The aim of this paper was to investigate the potential use of biological and ecological traits of macroinvertebrates as indicator systems of quality conditions in freshwater ecosystems. To provide a framework for the trait analysis, a data base was developed; it stored biological information about 472 benthic macroinvertebrate taxa. Twenty-two variables describing biological and ecological traits were resolved into a number of simple attributes. A fuzzy coding procedure was used to describe the link between taxa and variables. Three strategies, using simple descriptive statistics or multivariate ordination techniques, were examined in their abilities to demonstrate some of the characteristics of the environment and to bring out clear information on the rate of changes in habitats. We used the Loire River, France, as a specific example. Like faunal analyses, trait analyses showed that a major change occurred in the community in the upper reaches of the river, but transition zones that were not detected with faunal data alone, were discriminated. They corresponded to differences in combinations of biological and ecological traits of communities and were linked to tributary influences (e.g. the Allier–Loire confluence) or anthropogenic disturbances: dams (Grangent, Villerest) or urban zones (Le Puy, Tours). We concluded that the species trait approach had the potential to evaluate the actual state of ecosystems, to discriminate among different types of human impact, and to develop monitoring tools considering the functional diversity of communities.

Macroinvertebrate Community Structure and Environmental Characteristics along a Large River: Congruity of Patterns for Identification to Species or Family

In developed countries the longitudinal ecological pattern of large rivers is often strongly influenced by water pollution and flow regulation. The relationship between macroinvertebrate communities and this anthropogenic environment was studied along the French Rhône River. Collected specimens were identified to both family and species level and the results compared. Along the 512 km of this 7th-8th-order river, 10 reaches were chosen for study. Macroinvertebrates were collected 3 times during spring and summer 1991, using 2 techniques, artificial substrates and hand nets Environmental variables (16) were measured and associated with the collected faunistic communities, composed of 73 species or 53 families, depending on the level of identification. The faunistic and environmental structures were compared by means of a co-inertia analysis carried out on the 2 matrices. For the environmental variables and for both families and species, the longitudinal changes were preponderant. The correlation between fauna and environment was very good, as shown by the correlation coefficients r of their F1 scores in the co-inertia analyses: 0.839 and 0.814 for families and species, respectively. For both the fauna and the environment, the ecological pattern of the Rhône River changed downstream of the main city, Lyon, and of the main tributary, the Saône River, where the fauna became poorer. Ephemeroptera and Trichoptera, which were well represented in the Upper Rhône, were replaced by molluscs and lentic groups in the Lower Rhône. The environmental changes in the lower course consisted of higher water mineralisation, increased pollution, and greater flow regulation. Similar patterns were observed at both family and species level, with only a slightly more regular longitudinal change in the upper course at species level. Water pollution was not the only reason for an impoverished fauna in the Lower Rhône, as several taxa found in the upper course were also found in a by-passed section of the lower course. Thus the strong regulation of the Lower Rhône is largely responsible for the limited fauna. The family level of identification gave a meaningful picture of the macroinvertebrate longitudinal pattern of a large river. This family pattern was then compared with that observed in other European large rivers such as the Rhine and the Danube.