You must have a javascript-enabled browser and javacript and stylesheets must be enabled to use some of the functions on this site.


The ECCO Near Real-Time Ocean Data Assimilation System

Ichiro Fukumori(1) , Benyang Tang(1) , Zhangfan Xing(1) , Dimitris Menemenlis(1) , Ou Wang(1) , and Sophie Ricci(1)

(1) JPL/Caltech, M/S 300-323, Pasadena, CA 91109, United States


Products of the ECCO Sequential Ocean Data Assimilation System ( consists of near-global ocean state estimates from 1993 to present. These results are being utilized in diverse studies of ocean circulation and in geodetic applications. This presentation will describe the estimation system, the skill of its estimates, its latest improvements, and future outlook.

The on-going near real-time estimation system is based on an approximate Kalman filter and Rauch-Tung-Striebel smoother (partitioned, reduced-state, time-asymptotic approximations), and assimilates altimetric sea level and in situ temperature profiles with uncertainty in time-variable wind forcing as model control. The model is based on the MIT general circulation model with a variable 0.3 to 1-deg spatial resolution in a near global domain (73S to 73N). Estimates are statistically self-consistent and are in closer agreement with independent observations than a non-constrained model simulation.

Various improvements to the operational system are being developed. These include expansions to the model control space, such as time-invariant errors (bias) and diabatic sources of model error. A higher resolution system is also being developed to explore estimation of meso-scale variability in an eddy-permitting global circulation model.


Workshop presentation


                 Last modified: 07.10.03