Environmental Simulation PDF Download

Simulation modeling is arguably the most versatile scientific tool for predicting the future environment. However, the reliability of model-based predictions is limited to the behavior domain defined by the historical data employed for conceptualizing and calibrating the model. Future changes in external inputs and internal structure tend to produce system behavior significantly different from prior predictions. To abate this seeming lack of credibility, it is now customary to qualify model predictions with uncertainty estimates. This dissertation explores the complementary approach of back-casting future scenarios. Centered on the analysis of uncertainty, a methodological framework is developed for the computational evaluation of environmental futures, driven by stakeholder participation as a means for establishing credibility in the model. The analysis reveals possible structural change between the observed past and speculated future scenarios by comparing the ranking of key sources of uncertainty in model outputs. Three sampling-based methods are employed: Regionalized Sensitivity Analysis (RSA), Tree-Structured Density Estimation (TSDE), and Uniform Covering by Probabilistic Rejection (UCPR). RSA and TSDE are tested for identifying and ranking the key factors that influence ecological behavior in Lake Oglethorpe, Georgia, and UCPR, for recovering parameters of a rainfall-runoff model of an experimental watershed near Loch Ard, Scotland. The framework is applied to an integrated assessment of ecological behavior in Lake Lanier, Georgia. Stakeholders' fears and desires for the future state of the reservoir are elicited and encoded for analysis. The results indicate: (i) that the desired future is more reachable, and accompanied by more significant structural change, than the feared future, and (ii) that sediment-water-nutrient interactions, secondary production, and microbial processes play a critical role in the future ecological behavior of the reservoir. Thus, it is possible to: (i) confirm or refute stakeholder concerns for the future environment, (ii) inform priorities for future environmental policy actions, (iii) identify critical gaps in current knowledge, in order to prioritize future scientific research, and (iv) promote adaptive community learning, through the continual mutual feedback between scenario-generation and systematic analysis. By bridging the gap between stakeholder imagination and scientific theory, through computational analysis, the framework provides a promising direction for integrated environmental assessment.