Principal component proxy tracer analysis

Abstract

We introduce a powerful method for dynamical reconstruction of long-lived tracers such as ozone. It works by correlating the principal components of a matrix representation of the tracer dynamics with a series of sparse measurements. The method is tested on the 500 K isentropic surface using a simulated tracer and with ozone measurements from the Polar Aerosol and Ozone Measurement (POAM) III satellite instrument. The Lyapunov spectrum is measured and used to quantify the lifetime of each principal component. Using a 60 day lead time and five (5) principal components, cross validation of the reconstructed ozone and comparison with ozone sondes return root-mean-square errors of 0.20 ppmv and 0.47 ppmv, respectively.