Environmental Modeling Under Uncertainty: Monte Carlo Simulation

Fedra, K. (1983). Environmental Modeling Under Uncertainty: Monte Carlo Simulation. IIASA Research Report. IIASA, Laxenburg, Austria: RR-83-028

Preview

Text RR-83-028.pdf Download (2MB) | Preview

Abstract

The study of environmental systems as ecological and physicochemical as well as socioeconomic entities requires a high degree of simplifying formalism. However, a detailed understanding of a systems function and response to various changes for the explicit purpose of systems management and planning still requires fairly complex hypotheses, or models. Such models can hardly be subjected to rigorous tests without the aid of computers. Systems simulation is a powerful tool when subjecting complex hypotheses to critical tests of their logical structure and their performance over the range of plausible input conditions .

Based on a formalized trial-and-error approach using Monte Carlo methods, this report presents and discusses an approach to simulation modeling under uncertainty. An introduction to the causes and implications of the problem, namely uncertainty, and a short formal presentation of the methodology proposed are followed by some more technical remarks on Monte Carlo simulation. Using three different application examples, the author discusses the role of uncertainty in the formal testing of model structures, in parameter estimation, and in prediction. In the last example, the limits of estimation and, with it, prediction are demonstrated. In a comparison of Monte Carlo simulation and alternative approaches to including and evaluating uncertainty in simulation modeling, the discussion section examines the implications of uncertainty for model application in a broader framework.

Actions (login required)

View Item