Marine Ecosystems Research Department - jamstec japan agency ...

Japan Marine Science and Technology Center
Frontier Research System for Global Change
Fig.33 Relationship of functional components of our program.
In the following, we make a brief summary of our
results in the first year of the research period ( years)
with an emphasis on Theme for which the Frontier
Research System for Global Change (FRSGC) is mainly
responsible.
The main research activity regarding Theme during
FY, whose principal investigator (PI) is Dr.
Mitsudera at the International Pacific Research Center
(IPRC) of University of Hawaii, was directed to the
acquisition and quality check of observational and
reanalysis datasets for atmosphere, ocean, and land surfaces
through the international data network. As a result,
many important datasets (Table ) have been archived.
A close examination shows that the BUFR dataset
obtained from NCEP/NCAR is the most suitable input
data for our D-VAR coupled DA experiment and that
the ERA- of ECMWF is better for the assessment of
our experimental results. In addition, a very high resolution
SST dataset (km cloud-free and global) has been
constructed as well as mean sea surface fluxes in the
Pacific by colleagues of some universities.
A high-resolution coupled GCM in Theme was
improved markedly in both computational speed and
climate dynamics on ES by the joint group of ES
Center and FRSGC (PI is Dr. Kurihara of FRSGC).
This improved coupled GCM was used as the forward
model of our D-VAR DA system in Theme . The DA
group of FRSGC has completed the oceanic component
of our D-VAR coupled DA system and applied it to the
state estimation of global ocean circulation. As for the
atmospheric component, the DA system using a dry
model version has been constructed and applied to a sensitivity
experiment. We summarize these results below.
Using a sophisticated ocean GCM (MOM) and the
D variational adjoint method, we have succeeded in
obtaining a dynamically-self consistent reanalysis data
for climatological seasonality of the global ocean, with
finer resolution than the existing systems. A synthesis
of available observational records and the model provided
realistic features of ocean circulations with no
artificial sources/sinks for temperature and salinity
fields, in contrast to the nudging approach often used.
Thus, this new dataset enables us to qualify the water
mass formation and movement as well as the surface
conditions. For example, we made a water mass analysis
of the North Pacific Intermediate Water (NPIW)
that characterizes the subsurface region in the North
Pacific. The sensitivity experiment using the backward
adjoint code revealed that the origin of the NPIW can
be traced back to the Sea of Okhotsk and to the Bering
Sea in the subarctic region and to the subtropical
Kuroshio region further south (Figure ). This is in
155