Imprecise Probability and Water Resources Decisions

by W. F. Caselton, Univ of British Columbia, Vancouver, BC, Canada,
W. Luo, Univ of British Columbia, Vancouver, BC, Canada,



Document Type: Proceeding Paper

Part of: Water Management in the '90s: A Time for Innovation

Abstract: In water resources engineering weak information would correspond to circumstances when the record is extremely short relative to the mean rate of occurrence of important design events, and when the subjective information has been derived from a very limited experience of such events. The term imprecise probability has been coined to describe a concept of uncertainty which attempts to recognize these neglected uncertainties due to weak inputs. A number of different imprecise probability implementation schemes, which still involve conventional probability to some degree, have been proposed. The two imprecise probability methodologies discussed in this paper have the considerable advantage of retaining a close link with the establihsed Bayesian framework. These are Bayesian Robustness, and Dempster-Shafer methods.

Subject Headings: Probability | Water resources | Uncertainty principles | Management methods | Information management | Bayesian analysis | Resource management | Hydraulic engineering

Services: Buy this book/Buy this article

 

Return to search