Chemical Source Inversion Using Assimilated Constituent Observations

Andrew Tangborn
Joint Center for Earth Systems Technology Field, University of Maryland - Baltimore County

We present a comparison of source inversion for chemical constituent sources using assimilated constituent observations with direct use of the observations. In this model problem, a two-dimensional spectral transport model is combined with a Kalman filter. Inversion is carried out using a Green's function method. Observations are simulated froma "true" state with added Gaussian noise. The forecast state uses the same spectral model, but differs by an unbiased error. Two different observing systems are used, in situ and satellite.

Source inversion is carried out by either direct use of the observations in the Green's function inversion, or by first assimilating the observations and using the analysis as if they were observations. We have conducted 20 twin experiments for each case and find that in the limiting cases of very few localized observations, or an extremely large observation network there is little advantage to carrying out assimilation first. However, in intermediate observation densities, there is a significant increase in the accuracy of the source inversion standard deviation using the Kalman filter algorithm followed by Green's function inversion.

Last Update: September 29, 2005