Nowadays it is not possible to deny the fact that the human are one more component of river basins and that their activity affects enormously the water quality. To European level, the high density of population placed in the river basins has supposed an increase of the bad river water quality. In the last decades the increase of nutrient loads in the aquatic systems has turned into a priority problem to solve for the competent water agencies.
The management of the fluvial ecosystems is not an easy task. They are complex systems due to the narrow relationship among fluvial and terrestrial ecosystems. Additional to the complexity of these systems, one finds the associate difficulty of the management or control of pollutant inputs (from point and/or non-point sources). For all these reasons the management of the river water quality is a complex task that requires a multidisciplinary approach.
With the aim to reach this approach, different tools have been used, from mathematical models to expert systems and decision support systems. However, most of the efforts have been directed to the resolution of limited complexity problems, doing that many of the environmental complex problems, as river ecosystems management, have not been really treated. For this reason, there is needed the application of helping tools in the decision-making processes and that incorporate wide heuristic and empirical knowledge: expert systems and decision support systems. The ideal management of the river water quality requires an integrated and multidisciplinary approach, which can be reached by an intelligent tool based on the concepts and methods of the human reasoning. The present thesis describes the methodology developed and applied in the building of an Expert System, as well as the development process of this Expert System, as the main module of reasoning of an Environmental Decision Support System. The main objective of the present thesis has been the development of a tool to help water managers in the decision-making processes to improve water quality of altered reaches. In addition, the thesis shows the functioning of the developed tool by means two study cases. The results derived from the developed, implemented and presented Expert System show that these systems can be useful tools to improve the management of fluvial ecosystems.