Representing uncertainty in silvicultural decisions: an application of theDempster-Shafer theory of evidence

Forest management decisions often must be made using sparse data and expert judgment. The representation of this knowledge in traditional approaches t o decision analysis implies a precise value for probabilities or, in the ca se of Bayesian analysis, a precisely specified joint distribution for unkno wn parameters. The precision of this specification does not depend on the s trength or weakness of the evidence on which it is based. This often leads to exaggerated precision in the results of decision analyses, and obscures the importance of imperfect information. Here, I suggest an alternative bas ed on the Dempster-Shafer theory of evidence, which differs from convention al approaches in allowing the allocation of belief to subsets of the possib le outcomes, or, in the case of a continuous set of possibilities, to inter vals. The Dempster-Shafer theory incorporates Bayesian analysis as a specia l case; a critical difference lies in the representation of ignorance or un certainty. I present examples of silvicultural decision-making using belief functions for the case of no data, sparse data, and adaptive management un der increasing data availability. An approach based on the Dempster-Shafer principles can yield not only indications of optimal policies, but also val uable information about the level of certainty in decision-making. (C) 2001 Elsevier Science B.V. All rights reserved.