scope:

This introductory section has provided an overview of the objectives of the entire project, which was initiated because there are no standard objective quantitative means of evaluating microcomputer-based hazard response models. There are dozens of such models that have been recently proposed including several sponsored wholly or in part by the U.S. Air Force: ADAM, AFTOX. CHARM. DEGADIS, SLAB. and OB/DG. A few data sets exist for testing these models, but, up until now, the models have not been tested or intercompared with these data on the basis of standard statistical significance tests. The U.S. EPA recently sponsored a related model evaluation project (Reference 1), which had a more limited scope and considered fewer models and datasets.

The scope of this volume is limited to analyses of the three components of model uncertainty. Uncertainties due to data errors are studied from several viewpoints, including (1) analysis of observed data errors at several sites, (2) analysis of variability in derived parameters, and (3) application of Monte Carlo procedures to calculate model sensitivity to input data errors. Uncertainties due to stochastic fluctuations are discussed by first reviewing available theoretical empirical formulas, then suggesting some applied formulas, and finally testing these formulas with fast-response concentration observations at Maplin Sands. The component due to model physics errors is difficult to quantify, but was analyzed by demonstrating how the variance in observed concentrations could be reduced by the application of similarity scaling principles.