IWRA Proceedings

AbstractWater management, as any other knowledge intensive activity, to be efficiently applied in everyday practice, needs knowledge. Knowledge is usually possessed only by a narrow group of specialists (experts in the area, e.g. river basin dispatchers) who know when, how, and what must be done in order to provide proper water supply, or to cope with a dramatic consequences of floods. This knowledge, as it happens with experts everywhere, may not be available whenever it is necessary from various reasons: • experts need not be always achievable when necessary, • experts can suffer from common human problems, or suddenly their knowledge can be lost because of their mortality, or retirement, • experts can differ in their opinions how to solve a particular situation, etc. All of these and other reasons are in favour of the opinion that various knowledge management solutions using proper knowledge-based tools, which have already proven their usefulness in other areas, could be very beneficial also in water management. From the engineering point of view, the experts reasoning about problems and decision making about suitable solutions is understood as manipulating with specific data, mathematical models of the real situation, simulations, etc. In water management domain, we can think of mathematical modelling of hydrological data, mathematical models for optimization and control of reservoirs seen from the hydrological aspects, models of optimized water management in dry periods, rainfall-runoff terms for outflow forecasting, or elementary models of pollution effusing in rivers. In case of inaccessibility, incompleteness, or incorrectness of data as well as in other situations with high degree of uncertainty, experts still are able to make decisions, while all the classical approaches fail; neither algorithmic solution nor exact formulae can be used. The obstacles related to the lack of data can be solved by employing suitable approaches from the area of soft computing. However, these approaches are not always fully adequate. Then the space for the utilization of the knowledge-based technologies and for knowledge management opens. The paper deals with a number of our recent attempts to employ knowledge-based technologies, and especially some knowledge management approaches in the area of river basin management. In a number of cases at the Czech watersheds this happens to be knowledge rich but often data poor area. We are convinced that in many cases better exploitation of knowledge possessed by human experts in the area can successfully substitute lack of data and help in achieving a more effective management of relevant water resources.