Metamodels for Ozone: Comparison of Three Estimation Techniques

Abstract

A metamodel for ozone is a mathematical relationship between the inputs and outputs of an air quality modeling experiment, permitting calculation of outputs for scenarios of interest without having to run the model again. In this study we compare three metamodel estimation techniques applied to an 18 year long CMAQ simulation covering the Northeastern US (NEUS). The estimation methods considered here include projection onto latent structures, stochastic kriging and a combination of principal components and stochastic kriging.

Although this article has been reviewed by the EPA and approved for publication, it does not necessarily reflect EPA’s policies or views.

This work was made possible by Coordinating Research Council contract A-89.