Title: Insights on observation residual approaches for model and observation error estimation.

Author: Ricardo Todling (NASA/GMAO)

This presentation gives a brief review on attempts to estimate model error. One particular strategy based on a smoother approach is reviewed. The sequential smoother is shown to estimate additive model model error in the context of a simple chaotic dynamics. Recasting the sequential estimation approach into a corresponding variational formulation reveals details about estimation procedures based on observation residual statistics. Illustration of the limitations of such procedures is presented and a discussion is provided on the consequences of these findings when attempting to estimate and correct for observation error correlations in practical data assimilation applications.

GMAO Head: Steven Pawson Global Modeling and Assimilation Office NASA Goddard Space Flight Center

Curator: Nikki Privé Last Updated: Feb 9 2015