Learning Bayesian decision analysis by doing: lessons from environmental and natural resources management

The planet we are living on is getting small; each decade the number of peo ple here grows by almost 1 billion. Due to the escalating pressure that man kind puts on natural resources and the environment, there is a pressing nee d to develop management schemes and approaches that acknowledge the pragmat ic character of the problems: We scientists should not just passively obser ve and measure but also need to assist policy makers for better action. Thi s requires the ability to combine, interconnect, link, and analyze jointly information, knowledge, and judgment across scientific disciplines. The met hodological development is blooming and rich. However, the way to applicati ons tends to be long. It is not enough that one has learned and applied a m ethodology; it has also to be comprehended and accepted by many others who often are not all that devoted to methodological challenges; and launched t o responsible institutions. In this paper, we make an overview of lessons l earned from studying, applying, and launching of Bayesian decision analysis -influence diagrams and belief networks in particular-in the field of resou rce and environmental management. A number of case studies from water resou rces and fisheries are used as an illustration. (C) 1999 Elsevier Science B .V. All rights reserved.