Development and test of a statistical model for the ecological assessment of tropical reservoirs based on benthic macroinvertebrates

Reservoirs are heavily modified lentic ecosystems. In spite of their differences from natural lakes, it is important to maintain and improve their chemical and ecological status. In the present study, we tested the value of an assessment tool based on the structure of benthic macroinvertebrate communities, to evaluate the Ecological Potential (EP) of tropical reservoirs. We designed a conceptual assessment scheme based on the Reference Condition Approach, and developed a statistical model based on 28 sites classified as having Maximum Ecological Potential, localized in the reservoir of Serra Azul, Minas Gerais, Brazil. Sixty-two disturbed sites from three reservoirs were used to test the model. A classification system based on three EP classes was found to be the best option, and tracked different levels of total dissolved solids, turbidity, total nitrogen and trophic status. This study confirmed the utility of benthic macroinvertebrates as an indicator group of biological quality in reservoirs, and the statistical model applied was effective in providing a measure of the ecosystem health of the reservoirs. As a further improvement, the level of taxonomic resolution for certain groups such as chironomids could be increased, because knowledge of the species composition may provide a better discrimination of intermediate degradation levels.