Regresar

Multivariate approach in the exploration of the environmental tolerance of macroinvertebrate families in the Mediterranean rivers for the GUADALMED project

Abstract:

Macroinvertebrates and selected physico-chemical characteristics (temperature, pH, basin area, stream order, altitude, discharge, conductivity, ammonium, nitrites, nitrates, phosphates) were studied in 157 sites distributed over 12 Mediterranean fluvial basins located in the Iberian Peninsula and in the island of Majorca. A total of 127 taxa were recorded, mostly identified to the taxonomic resolution of family. Correspondence analysis (CA) and canonical correspondence analysis (CCA) were used to explore the tendencies of variation of the abundances of different families and their relationships with physico-chemical variables. The selected environmental variables explained more than half of the pure variance of the data matrix of families on the three first canonical axes. Salinity and temperature were the main variables explaining the distribution of families. These variables formed a natural gradient which was negatively correlated with altitude and discharge. The most halo-thermophilous families, according to this ordination, were Thiaridae and Ephydridae, followed by crustaceans, Odonata, Coleoptera and Heteroptera (OCH). Stoneflies were the group of families favouring narrow ranges of the lower values of temperature and salinity, followed by mayflies and caddisflies. A second environmental gradient was chiefly correlated with discharge. However, the gradient was also related with temporality of the aquatic habitat. In this second gradient it was difficult to separate the natural regime from anthropogenic disturbances. Organic pollution was associated with the third emerging gradient. Asellidae, Syrphidae, Chironomidae and Physidae were the macroinvertebrate families most related to levels of organic enrichment. Generally, the richness of families whithin any taxonomic group, showed very strong negative correlations with the organic pollution-associated gradient. The IASPT biotic index showed a stronger negative correlation with the gradient salinity-temperature than with the pollution gradient. Therefore, biotic assessments using this index must by done cautiously. In this weakness underlies the general tendency to overestimate the indicator value of good quality attributed to the families in the group EPT compared to that of the orders OCH. The IBMWP biotic index was, in great part, immune to the above bias thanks to the major contribution of family richness in the calculation of the index. Family richness was not correlated to the gradient salinity-temperature. © Asociación Española de Limnología, Madrid. Spain.