Approximation and Inversion of a Complex Meteorological System via Local Linear Filters

Abstract

Dynamical models involving systems of numerous differential equations are commonly used to describe meteorological behavior. Approximations to such systems are often desired, particularly for inversion problems when the model?s adjoint is unavailable. A method for approximating a general, nonlinear system is explored here. The method is particularly useful for inverse problems. An application to a complex multivariate dynamic model for southern California air quality is given, and the method is shown to provide satisfactory estimates of vehicle emission-rate inputs using ozone concentration estimates for the Los Angeles basin.