www.ecco-group.org/abstracts/woce_abstract_3.txt

The ECCO 1 degree global WOCE Synthesis.  The Estimated Mean State

A. Koehl, Y. Lu, P. Heimbach, B. Cornuelle, D. Stammer and C. Wunsch

for the ECCO Consortium.

The global time-dependent WOCE data synthesis made by the ECCO
Consortium over the 10-year period 1992 - 2001 aims at estimating 
the time-dependent ocean state that is, within error bars, consistent
with the global data sets and the dynamics imbedded in the numerical 
model.  We compare the estimated state with the WOCE observations that 
went into the synthesis. Overall, the agreement with the data is 
close to consistent with the prior error statistics for 
the data and model. The remaining misfit and a comparison
with the reference state provides information about the quality of
both the state and about and model errors.  The state estimation
procedure is able to correct for many of the traditional shortcomings
of the flow field. However, as adjustments are made through changing the
surface boundary conditions and initial conditions only, not all deficits
can be properly represented.  Mean meridional heat transports,
meridional overturning and volume transport through passages are 
slightly reduced in comparison with the unconstrained reference run.


1	The ECCO 1 degree global WOCE Synthesis. The Estimated Mean State
2
3	A. Koehl, Y. Lu, P. Heimbach, B. Cornuelle, D. Stammer and C. Wunsch
4
5	for the ECCO Consortium.
6
7	The global time-dependent WOCE data synthesis made by the ECCO
8	Consortium over the 10-year period 1992 - 2001 aims at estimating
9	the time-dependent ocean state that is, within error bars, consistent
10	with the global data sets and the dynamics imbedded in the numerical
11	model. We compare the estimated state with the WOCE observations that
12	went into the synthesis. Overall, the agreement with the data is
13	close to consistent with the prior error statistics for
14	the data and model. The remaining misfit and a comparison
15	with the reference state provides information about the quality of
16	both the state and about and model errors. The state estimation
17	procedure is able to correct for many of the traditional shortcomings
18	of the flow field. However, as adjustments are made through changing the
19	surface boundary conditions and initial conditions only, not all deficits
20	can be properly represented. Mean meridional heat transports,
21	meridional overturning and volume transport through passages are
22	slightly reduced in comparison with the unconstrained reference run.
23
24