Bayesian networks as a participatory modelling tool for groundwater protection

Participatory approaches help to better control and accelerate the integration, to make the decision-making process more transparent and comparable across transboundary river basins and scales, and to increase confidence in an integrated model-based planning process. Integration and participation can be significantly enhanced by using decision support systems (DSS) to assist the planning process, as they provide tools and platforms for collecting data from many sources, integrating models of different nature (physical, socioeconomical, and decisional), evaluating the effects of different planning alternatives, and in some cases, negotiating them. Models are an essential component of DSSs because they provide the system representation based on which the planning process is carried on. This chapter describes the use of Bayesian Networks (BNs)-a relatively recent modeling technique that is encountering wide diffusion in the environmental modeling community-in groundwater protection problems. The advantages of the BNs are that they are graphical, focus dialogue, integrate different types of data, and are interactive, trans- and interdisciplinary, and can quantify difficult cases. The graphical nature of BNs facilitates formal discussion of the structure of the proposed model. Also, the ability of a BN to describe the uncertain relationships among variables is ideal to describe the relationship between events that may not be well understood and intrinsically uncertain.