The construction of large reservoirs in South America and particularly in Brazil has intensified within the last 50 years. Built up primarily for hydroelectricity production, these artificial ecosystems now serve purposes such as: water storage for public use; fisheries and aquaculture; recreation; tourism, and irrigation. These artificial ecosystems were also built up with the purpose of enhancing the regional development. These activities produce multiple impacts among which are eutrophication, a serious problems with various ecological, economic, and social consequences. Basic studies on reservoirs have identified their main ecological characteristics and described some of their fundamental mechanisms of functioning. This information cover: factors involved in reservoir complexity, such as spatial scale, vertical and horizontal heterogeneity, temporal variation at several time scales and, in some cases, watershed/reservoir relationships, and the impact of non-point and point sources of nutrients. Numerous studies have been dedicated to describing aquatic biota, its relationship with environmental factors, and the impacts of degradation/pollution/ eutrophication on biodiversity. Research priorities have been established mainly by limnologists in response to needs identified by environmental impact assessment. Due to the pressure of multiple uses, population growth, and economic factors, aquatic scientists now face a number of questions being posed by managers and engineers. These questions relate to a wide range of practical problems, the solutions to which depend on accumulated data related to the structural complexities and function mechanisms previously referred to. These problems can be divided into three main groups: (i) eutrophication processes and their characteristics, and water quality control; (ii) impact of fisheries, aquaculture, and exotic species introduction on the biota and the water quality; (iii) sustainable development of reservoirs and optimization management of their multiple uses. Predictive capabilities required in the face of these problems should be based on modeling development and intensive use of data-bases resulting from long-term studies on reservoirs. The predictions thus made possible would involve economic factors and the costs of recovery of eutrophic reservoirs. The experience of reservoir research and management in South America and in Brazil is, probably part of a world wide tendency on reservoir research and development with the aim to optimize multiple uses. Based on the cooperation of limnologists and engineers, ecosystem management models must eventually be applied at the watershed / reservoir level, where demands on the part of managers and decision makers will certainly promote a predictive, integrated, and adaptive approach, based on fundamental research. [ABSTRACT FROM AUTHOR]