Prediction under uncertainty as a boundary problem: A general formulation using Iterative Closed Question Modelling. (August 2015)

• Record Type: Journal Article
• Authors: Guillaume, Joseph H.A.; Kummu, Matti; Räsänen, Timo A.; Jakeman, Anthony J.
• Is Part Of: Environmental modelling & software. Volume 70(2015:Aug.)
• Page Start: 97
• Page End: 112
• Publication Date: 2015-08
• Subjects: Uncertainty analysis -- Iterative Closed Question Modelling -- Boundary critique -- Process philosophy; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2015.04.004
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: Making predictions about environmental systems is a challenge due to the high level of uncertainty involved. In this paper we give a general formulation of prediction under uncertainty as a boundary problem. This leads to development of a methodology for making predictions under uncertainty, named Iterative Closed Question Modelling (ICQM). ICQM involves iteratively devising questions and testing the certainty of their answers by creating complete model scenarios (complete taken to include structure, parameters and inputs for each scenario instance). The model scenarios are categorised in terms of which answer they support, and whether they are plausible or not. Using a simple two-parameter flow duration curve model, the paper demonstrates the application of ICQM using eight alternative uncertainty analysis techniques. ICQM provides a useful and generic approach to making predictions under uncertainty, helps to understand how existing techniques address the boundary problem differently and promotes the development of new techniques. Graphical abstract: Highlights: Prediction under uncertainty is formulated as a boundary problem. A modelling methodology to approach uncertainty is proposed. It is broadly applicable, with eight techniques shown.