Quantifying Model Uncertainty in Environmental Impact Assessment
Integrated environmental assessment is aimed at evaluation of the current and future environmental quality in order to facilitate implementation of sustainable decisions. It clearly requires appraisal of relevant natural and societal processes that affect environmental resource. Due to the ability to imitate real world system reactions to external disturbances without actual perturbation of the system, simulation models are commonly recognized as efficient tools for environmental assessment. Simulation results are used for decision/policy making process which requires not only projected values of ecosystem characteristics, but also evaluation of their uncertainty. Traditionally evaluation of model uncertainty is conducted for a fixed set of parameters and requires a large number of simulation runs which can be time and labour consuming. Investigation of mathematical properties of the model can help to provide realistic assessment of the resulting uncertainty and to suggest the model with lesser uncertainty. The paper examines the importance of realistic evaluation of model uncertainty in environmental assessment. It describes water quality assessment case studies demonstrating when precise estimates of uncertainty become a necessity and when overestimated values can be sufficient.
Keywords: Environmental Assessment, Model Uncertainty, Water Quality
Prof. Marina G. Erechtchoukova
Assistant Professor, Atkinson Faculty of Liberal and Professional Studies, York University