Operation-Oriented Ensemble Data Assimilation of Total Column Ozone

Abstract

We have developed an advanced data assimilation system for a global ozone simulation model with a four-dimensional ensemble Kalman filter (EnKF). Satellite-measured total column ozone data known as EP/TOMS version 8 were assimilated. Two-month data assimilation cycle experiments were performed from January to February 2001 with three different observing density settings: i) full use of EP/TOMS data points, ii) reduced use of the data points at 1/10, and iii) reduced use of the data points at 1/100. For comparison, additional data assimilation experiments were performed with Newtonian relaxation or a nudging technique. In the case of full use of EP/TOMS data points, the EnKF outperformed the nudging technique by approximately 30% in terms of the root mean square difference in the northern hemisphere midlatitudes. Moreover, the reduced use of EP/TOMS data points demonstrated that the EnKF data assimilation was robust and stable, while the nudging technique was vulnerable to the decrease in observations. Overall, the EnKF is a promising data assimilation technique for the total column ozone and is usable for operational applications.