Routine Global Ocean Data Assimilation System of the Consortium
for Estimating the Circulation and Climate of the Ocean (ECCO)

Ichiro Fukumori

Jet Propulsion Laboratory


An ocean data assimilation system producing regular analyses is established so as to monitor global ocean circulation and to better understand processes underlying the ocean's seasonal-to-interannual changes. The effort is part of a consortium for "Estimating the Circulation and Climate of the Ocean (ECCO)." Products of the analysis system are regularly updated and are available via a Live Access Server at An overview of this system is presented and some of its applications are described.

The assimilation system is based on a near-global primitive equation model of high resolution (1-deg telescoping to 0.3-deg in the tropics with 10m near surface layers). Measurements from satellite altimetry (TOPEX/POSEIDON) and in situ hydrography (XBTs and climatology) are assimilated. The assimilation separately employs an approximate Kalman filter/smoother and the adjoint method. The former utilizes a novel approximation of partitioning model error estimates into a series of independent reduced rank systems and evaluating them separately. Such approximation dramatically reduces the computational requirements of Kalman filtering and smoothing, rendering high resolution global eddy-resolving data assimilation computationally feasible.

The ECCO analyses are characterized by their physical consistencies. Temporal evolution of data assimilated estimates are physically inconsistent when model state increments are not identified by explicit processes. The smoother can be recognized as such inversion of model state increments. Smoothed state and control estimates satisfy fundamental conservation statements (e.g., closed energy and tracer budgets) that permit mechanisms of ocean circulation to be investigated.

Products from the ECCO system are being employed to study heat balances and seasonal-to-interannual changes in tropical water masses. Results of the assimilation system are also employed in studies of biogeochemical cycles, earth rotation, and crustal deformation. The data assimilated solutions are generally found to be more accurate than those of the unconstrained model, permitting better estimation of the nature and effects of ocean circulation.