A high resolution oceanic GCM developed in LASG/IAP and a planning assimilation system based on it


Liu Hailong
LASG, Institute of Atmospheric Physics, Beijing, 100029
P.R. China

Jianglong Li, Yongqiang Yu, Wei Li and Xuehong Zhang

State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics (LASG),
Institute of Atmospheric Physics (IAP),
Chinese Academy of Sciences (CAS),
Beijing,
China, 100029

lhl@lasg.iap.ac.cn

The application of Argo data to optimal estimation of the model errors covariance matrix of ocean data assimilation system
Liu Yimin
National Climate Center of China,
Beijing, 100081,
P. R. China

Zhang Renhe, Yin Yonghong, Tao Niu
Chinese Academy of Meteorological Science,
Beijing 100081,
P. R. China.

liuyimin63@yahoo.com

Use of Argo observations in the French MERCATOR ocean monitoring and forecasting system
Pierre Bahurel, M. Benkiran, F. Hernandez, G.Larnicol, V. Tomazou
MERCATOR Ocean
8-10 rue Hermes - Parc Technologique du Canal
Ramonville St Agne 31526
FRANCE

Pierre.Bahurel@mercator-ocean.fr

MERCATOR is the French operational ocean monitoring and forecasting system, and one of the core systems of the international GODAE experiment. Based on routine assimilation (SAM reduced order optimal interpolation Mercator Assimilation Scheme) of satellite altimetry and vertical in situ profiles in primitive equations ocean models (OPA primitive equation OGCM), it is currently implemented in three different configurations:

The first one is a "eddy-permitting" configuration implemented on North Atlantic with
43 levels on the vertical and a 1/3° horizontal resolution,

The second one is a full "eddy resolving" configuration implemented on North Atlantic
and Mediterranean Sea with 43 levels on the vertical and a 5-7 km resolution (~1/15°)

The third one is a "low resolution" configuration implemented on Global Ocean with 32
levels on the vertical and a 2° horizontal resolution.

The MERCATOR system has been running in near-real-time since January 2001, with a progressive implementation of these three model-assimilation components. Argo data distributed by the CORIOLIS system have been used since 2001 as a validation data set. In 2003, a major upgrade has been performed on the Mercator assimilation scheme to introduce a multivariate approach and the assimilation of in situ data. This new assimilation system is currently implemented on the first configuration. It uses fully multivariate Empirical Orthogonal Functions (EOFs of T(z), S(z) and the barotropic streamfunction) to assimilate simultaneously altimeter SLA, in situ temperature and salinity profiles (included Argo data) and surface observation such as the Sea Surface Temperature (SST) and Salinity (SSS). As a complementary approach, Mercator developed a data-combination component (called Armor) where satellite altimetry and temperature data, and in situ temperature and salinity vertical profiles, are combined to derive 3D ocean fields.

An overview of MERCATOR and its applications will be given. The use of Argo data by the MERCATOR system will then be illustrated. Some examples of the use of Argo data for validation will be given and results of the assimilation of Argo data jointly with satellite altimetry will be presented.

CORIOLIS data centre: In situ data for operational oceanography

T.CARVAL(1), L. PETIT de la VILLEON (2) CHRISTINE COATANOAN

(1) IFREMER, IDM/ISIB,
Centre de Brest,
BP70,
29280 Plouzané,
France

(2)IFREMER, IDM/SISMER,
Centre de Brest,
BP70,
29280 Plouzané,
France

Thierry.Carval@ifremer.fr
Loic.Petit.De.La.Villeon@ifremer.fr
Christine.Coatanoan@ifremer.fr

The GyroScope project : objectives, overview, and preliminary results

Yves Desaubies and Gyroscope partners
IFREMER
Centre de Brest
BP70,
29280 Plouzané,
France

Yves.Desaubies@ifremer.fr

Analysis of Temperature and Salinity in the N Atlantic.

E. Autret, Fabienne Gaillard
DOPS/IFREMER
Centre de Brest
BP 70
29280 PLOUZANE
FRANCE

Fabienne.Gaillard@ifremer.fr

Near real time velocity estimates using float profiles

Pascal. Le Grand1 and Jean.-Pierre Mazé2

Laboratoire de Physique des Oceans - IFREMER/Centre de Brest
BP-70 29280 Plouzane
France

2Societe HOCER
5 rue Eugene Bourdon, ZI Keragadec
294920 GUIPAVAS
France

plegrand@ifremer.fr

Comparing and combining Argo data with altimeter data

Stephanie Guinehut, Pierre-Yves Le Traon and Gilles Larnicol
CLS, Space Oceanography Division,
8-10 rue Hermes - Parc Technologique du Canal
Ramonville St Agne 31526
FRANCE

pletraon@cls.fr

Determination of North Atlantic circulation by inversion of profiling float data

Carole Grit, Herlé Mercier
LPO/IFREMER
Centre de Brest
BP 70
29280 PLOUZANE
FRANCE

cgrit@ifremer.fr

Characteristics and variability of Subpolar mode water in the North Atlantic, as observed from Argo floats

H. Mercier and T. Hascoet
LPO/IFREMER
Centre de Brest
BP 70
29280 PLOUZANE
FRANCE

Herle/Mercier@ifremer.fr

Using Argo/Gyroscope float data, we describe the SubPolar Mode Waters (SPMW) properties and their seasonal and inter-annual variability, and we investigate the SPMW formation processes. The floats display a substantial data set of high quality salinity and temperature measurements. The space-time sampling allows us to interpolate the data on pressure and potential density levels throughout the mid-latitude North Atlantic Ocean via an objective analysis. Comparison with previous studies shows the consistency of the interpolation method and the ability of the data set to observe the SPMW.

We confirm the presence of several varieties of SPMW in the North East Atlantic, quite unequally distributed and being set apart by the density ranges they occupy. They are distributed around the outer periphery of the North Atlantic Subpolar Gyre in extended patches north of the North Atlantic Current (NAC) and in small patches south and near Newfoundland. South of the NAC density jumps are detected between the different varieties of SPMW. The analysis helps us distinguish recently ventilated SPMW from subducted SPMW. Those varieties can coexist in the same geographical area at different depth. A significant variability has been observed in regard to previous data sets.

Mode waters are formed in late winter when mixed layers reach their deepest. Our observations suggest that the deepening of the mixed layers occurs during a 2 month temporal window and that the deepest mixed layers appear in very restricted areas in the SubPolar Gyre. The local nature (in time and space) of the late winter deep mixed layers may explain the discontinuous aspect of the properties of the different SPMWs.

Operational oceanography in France: the CORIOLIS project

Sylvie Pouliquen, Thierry Carval, Yves Desaubies, Loic Petit de la Villéon, Gérard Loaec
Coriolis,
Ifremer,
BP70,
Plouzané29280,
France

Loic Gourmelen
EPSHOM,
BP426,
29275 Brest Cedex,
France

sylvie.pouliquen@ifremer.fr

Cross-crest pathways of Iceland Scotland Overflow Water in the North Atlantic Ocean.

1Walter Zenk, 1Matthias Lankhorst, 2Bernard le Cann
1Institut fuer Meereskunde an der Universitaet Kiel
Duesternbrooker Weg 20
D – 24105 Kiel
Germany

2 IFREMER, Brest

wzenk@ifm.uni-kiel.de

Monthly mapping of temperature and salinity by objective analysis with Argo floats and TRITON buoys data in the Pacific Ocean

Shigeki Hosoda and Shinya Minato
Japan Marine Science and Technology Center
2-15, Natsushima-cho,
Yokosuka,
Kanagawa, 237-0061,
Japan

hosodas@jamstec.go.jp

Application of a 4DVAR data assimilation system to the Argo data


Qin Jiang
3173-25, Showa-machi, Kanazawa-ku,
Yokohama-city,
Kanazawa, 236-0001
JAPAN

Toshiyuki Awaji (2,3), Nozomi Sugiura (3), Shuhei Masuda(3), Hiromichi Igarashi(3),Kensuke Takeuchi(1), Nobuyuki Shikama(1)

(1) Frontier Observational Research system for Global Change
(2) Kyoto University
(3) Frontier Research system for Global Change

jiang@jamstec.go.jp

ARGO is an international project aiming at monitoring the world ocean in real time by using automatic profiling floats. The measured profiles of temperature and salinity within the subsurface and middle (2000m) layers of the ocean with the Argo floats play an important role in understanding and forecasting climate.

However, since the Argo float is a floating buoy, the obtained Argo data is sparsely distributed both in time and in space. For the effective use, it is necessary to create the 3-dimensional grid dataset from the sparsely observed Argo data.

The data assimilation technique, on the other hand, is known as an effective method to construct the physically integrated reanalysis data by incorporating the measured data into the oceanographic model.

By the end of this June, 825 Argo floats have been deployed and more than 38,000 measured profiles of temperature and salinity is now available. In this study, we tried to use the 4DVAR data assimilation system developed by the FRSGC to the observed Argo data, in order to find an appropriate method for making reanalysis Argo data. In the assimilation experiment, a global ocean model with the 1x1° horizontal resolution and the measured Argo data of 2002 were used. By comparisons of the simulated and assimilated ocean states, it is revealed that both fields of temperature and salinity are significantly improved by the Argo data. Whereas, more further examination on the physical performance as well as the operational use of this assimilation system leaves to the future work.

Three-dimensional analyses of salinity and temperature in the Pacific using a variational method with vertical coupled temperature-salinity EOF modes

Masafumi Kamachi, Yosuke Fujii, Shiro Ishizaki, Norihisa Usui, Toshiya Nakano,
Tamaki Yasuda, and Hiroyuki Tsujino
Oceanographic Research Department
Meteorological Research Institute
1-1 Nagamine,
1-2 Tsukuba305-0052
Japan

mkamachi@mri-jma.go.jp

Pacific mixed layer depth from Argo
Shinnosuke Kanegae & Masahisa Kubota
School of Marine Science and Technology,
Tokai University 3-20-1,
Shimizu-orido,
Shizuoka,Shizuoka 424-8610
JAPAN

kanegae@mercury.oi.u-tokai.ac.jp

New Pacific historical dataset for the Argo standard scheme of delayed-mode quality control and its performance of salinity correction

Taiyo Kobayashi
Frontier Observational Research System for Global Change
2-15 Natsushima-cho,
Yokosuka237-0061
Japan

Shinya Minato
Japan Marine Science and Technology Center

taiyok@jamstec.go.jp

The JCOPE ocean forecast system

Yasumasa Miyazawa
FRSGC/JAMSTEC
3173-25 SHowamachi, Kanazawa-ku,
Yokohama City,
Kanazawa 236-0001
Japan

Toshio Yamagata
FRSGC/JAMSTEC and the University of Tokyo

miyazawa@jamstec.go.jp

A simulation of Argo float distribution in the Southern Pacific

Okumura, Tsugukazu (*), Naoto Iwasaka (*, **), Eitarou Oka (**), Kensuke Takeuchi (**)
(*) Tokyo University of Mercantile Marine
Etchujima 2-1-6, Koto-ku, Tokyo 135-8533
(**) Frontier Observational Research System for Global Change
2-15, Natsushima-cho, Yokosuka-shi, 237-6001

iwasaka@ipc.tosho-u.ac.jp

A new type of pycnostad in the subtropical-subarctic transition region of the North Pacific
Hiroko Saito (1), Toshio Suga (1,2), Kimio Hanawa (1), Eitaro Oka (2), Nobuyuki Shikama (2)
(1) Department of Geophysics,
Graduate School of Science,
Tohoku University,
Sendai 980-8578,
Japan

(2) Frontier Observational Research System for Global Change

piko@gw1.pol.geophys.tohoku.ac.jp

Barrier layer in the North Pacific subtropical gyre

Kanako Sato(*1), Toshio Suga(*1)(*2) and Kimio Hanawa(*1)

(1)Department of Geophysics,
Graduate School of Science,
Tohoku University,
Aramaki-Aza-Aoba,
Aoba Ward,
Sendai980-8578
Japan

(*2)Frontier Observational Research System for Global Change, Yokosuka, Japan

satou@pol.geophys.tohoku.ac.jp

Recovery of Argo floats

N.Shikama 1),E.Oka 1), K.Ando 2), H.Nakajima 3), K.Suehiro 4), T.Takashiba 4), K.Mizuno 5), and K.Uehara 5)

1)Frontier Observatioal Research System for Global Change,
JAMSTEC,
2-15, Natsushima,
Yokosuka,
Kanagawa 237-0061
Japan

2)Ocean & Earth Division, Ministry of Education,Culture,Sports,Science and Technology
3)Marine Works Japan LTD
4)Hydrographic and Oceanographic Department, Japan Coast Guard
5)National Research Institute of Far Seas Fisheries, Fisheries Research Agency

nshikama@jamstec.go.jp

We have recovered seven Argo floats in five observational cruises since November 2001. The aim of recovery is to examine the long term trend of CTD sensors of floats and to find out the cause of malfunction of floats such as "energy flu" experienced with APEX floats. The calibration after recovery for three floats show no significant drift in temperature and salinity measurement compared with the calibration before deployment.

Two APEX floats recovered in June 2003 had experienced to touch the sea bottom several times before recovery. The rubber plug was tightly fit to the bottom of the case of these two APEX floats, however, a small quantity of sand or muddy sand, 13.1g and 2.1g respectively, was found inside the bladder case. The sand seemed to get into the case through eight small holes located on the upper part of the case which are used for water discharge from the case. An estimate of quantity of sand based on the ballasting piston position of the APEX floats gives a reasonable result.

The recovered APEX showing the symptom of energy flu is now being carefully examined at Webb Research to find out the cause of battery drop .

A user-friendly method of choosing launching points to deploy ARGO floats drifting with certain average spacing

Kakuta Shinya and Sasaki Yasunori
Japan Marine Science and Technology Center
Natsushima-cho 2-15,
Yokosuka 237-0061
Kanagawa
Japan

kakutas@jamstec.go.jp

An easy-to-use method of determining launching points to deploy floats drifting with certain average spacing is proposed. Counting the number of drifting floats within a certain radius from the last launching point, this method determines the next launching point successively. It is easy to adopt this method even on cruise tracks curved or broken. In the areas where floats converge or diverge, the above-mentioned radius should be extended or reduced proportionally to the rate of convergence or divergence.

Objective is to deploy drifting floats with average spacing of about 300 km. It may be supposed that, ideally, floats are located on square grids with intervals of 300 km. However, it is not practical to realize such even spacing owing both to limitation of ship time and to drift of floats. To accomplish the objective, thus, "average spacing of about 300 km" should be defined. This definition is discussed in the following section.

Counting the number of drifting floats within a certain radius from the last launching point, this method determines the next launching point successively. For practical use, we propose to define "average spacing of about 300 km" as follows. Suppose that a ship is now launching a float at Point A, the center of a circle with 3L/2 of radius. Here, L is desired average spacing, for example, 300 km for Argo. The area of the circle is 9piL2/4, nearly equal to 7L2. Thus, the number of floats in the circle is desired to be 7. Suppose that only 5 floats, including one just now launched at Point A, are drifting in the circle. Thus, another 2 floats are desired to be launched in the circle. Launching points of those next 2 floats and Point A are desired to be apart from each other. Thus, the launching interval AB is determined by dividing the radius by 2. When a float has been launched at Point B, the succeeding launching point is similarly determined with Point B centered.

In the above description, the area of the circle 9PiL2/4 is approximated as 7L2, for simplicity of explanation. However, it is not necessary to round 9p/4 to an integer number. More generally speaking, the interval AB can be determined by dividing the radius by (9Pi/4-5). Since this method determines distance between Points A and B, the cruise track is allowed to be curved or broken. This is an advantage in practice because cruise tracks are not always straight.

Of course, intervals among floats within the circle may evolve to take any value during their life span; namely, some floats may get closer while others apart. Such detailed motion is not predictable because of oceanic eddies. As a result, it is also unable to predict where float-sparse portions will appear in smaller scales than the radius in the future, for example, one and a half year later. Thus, it is meaningless to worry about future local convergence or divergence of drifting floats in such small scales in advance.

A role of eddies in formation and transport of North Pacific Subtropical Mode Water

Hiroki Uehara
Department of Geophysics,
School of Science,
Tohoku University,
Sendai 980-8578,
Japan

uhiroki@gw1.pol.geophys.tohoku.ac.jp

Evidence for cyclonic eddies formation in the Antarctic Divergence off Wilkes Land

Shuki Ushio
Arctic Environment Research Center
National Institute of Polar Research
9-10, Kaga 1-chome, Itabashi-ku,
Tokyo 173-8515, JAPAN

Yasushi Fukamachi and Masaaki Wakatsuchi
Institute of Low Temperature Science,
Hokkaido University

ushio@pmg.nipr.ac.jp

Short-term variation of surface mixed layer temperature in a warm core ring in the sea east of Japan

Takashi Yoshida and Mizuho Hoshimoto
Office of Marine Prediction, Japan Meteorological Agency
Otemachi 1-3-4, Chiyodaku,
Tokyo 100-8122
Japan

tyoshida@met.kishou.go.jp

Real-time data assimilation products with Argo data
Kumi Yoshita, Tsurane Kuragano, Satoshi Sugimoto, Yasuhiro Minamoto
Office of Marine Prediction
Climate and Marine Department,
Japan Meteorological Agency
1-3-4 Ote-machi,
Chiyoda-ku,
Tokyo100-8122
Japan

kyoshita@met.kishou.go.jp

Error analysis with Argo data : On the ability of an OGCM to simulate the temperature and salinity in the western Pacific

You-Soon Chang, Homan Lee, Jang-Won Seo, and Yong-Hoon Youn
Marine Meteorology and Earthquake Research Lab.
Meteorological Research Institute,
Seoul, 156-720,
Korea

you@metri.re.kr

The vertical profiles of temperature and salinity derived from an Argo data in the western Pacific have been analyzed by the comparing with climatological data (Levitus94, WOA98) and OGCM output. The OGCM used in this study is GFDL MOM3.0 covers the global ocean with a 1° by 1° and 29 level resolution, respectively. Six areas in the western Pacific are selected near the western boundary currents, which influence the heat budget of the warm pool and play a key role in the global climate change.

The vertical profiles of time mean temperature and salinity averaged over one year have high correlations between three kinds of data (Argo, climatology and model output). The temporal variability of ARGO data is much larger than those of two other data, and also warm bias(0.5-1°) is appeared compared to climatological data. Salinity minimum depth simulated by the OGCM is close to the observed one, but the simulated salinity is generally higher (0.2 psu) than the observed value. And the stratification of model is too week compared to that of observation.

Results from the improved OGCM based on this study could lead to high-order descriptions of the western boundary currents in Pacific.

Mean flow and variability at the Upper Portion of the East Sea Proper Water in the southwestern East Sea with APEX Floats

Homan Lee, Tae-Hee Kim, Jang-Won Seo, and Yong-Hoon Youn
Meteorological Research Institute
Korea Meteorological Administration
460-18 Shindaebang-dong Dongjak-gu,
Seoul 156-720
Korea

homann@metri.re.kr

During 2001 to 2003, 16 APEX floats are used to understand the currents the upper portion of the East/Japan Sea (EJS) proper water (at about 800 m underwater) in the southwestern East Sea, deployed in the EJS by Meteorological Research Institute (METRI) of Korea Meteorological Administration (KMA) and Korea Ocean Research & Development Institute (KORDI) as part of the Array for Real-time Geostrophic Oceanography (Argo) program.

The flow penetrates into the Ulleung basin (UB) through two paths: an extension of the southward flowing the North Korean Cold Current along the east coast of Korea and between Ulleung Island and Dok island called the Ulleung Interplain Gap. Inflows at 800 m are observed range 0.2 to 4.29 cm/sec and the variability in the north in the UB is stronger than that in the south. The eddy kinetic energy is found a few cm2 s-2. In the UB, cyclonic flows from 0.3 - 1.6 cm/sec are observed with the bottom topography.

An advanced method to estimate deep currents from Profiling Floats

Jong Jin Park
241-318, Seoul National University,
San 56-1, Shilim-Dong,
Gwanak-Gu, Seoul, 151-742
South Korea

Kuh Kim
School of Earth and Environmental Sciences, Seoul National University, Seoul, Korea
Brian A. King
Southampton Oceanography Centre, Southampton, UK
Stephen C. Riser
School of Oceanography, University of Washington, Seattle, USA

jpark@ocean.snu.ac.kr

Data blending of Argo drifter profiles with other sources of data in the East (Japan) Sea

Young Jae Ro, Eung Kim, and Yong Hoon Youn*
Dept. of Oceanography,
Chungnam National University,
Taejon,
Korea,

*Marine Meteorology and Earthquake Research Lab., METRI, KMA, Korea

royoungj@cnu.ac.kr

Pacific Islands GOOS - The SEREAD project

Cristelle Pratt
Manager - Ocean and Islands Programme
SOPAC South Pacific Applied Geoscience Commission
Private Mail Bag, GPO,
Suva,
Fiji Islands

Cristelle@sopac.org

3D, EOF-based spatial analysis of Gyroscope observations in the North Atlantic Ocean

Simon Ruiz1,D. Gomis2 Jordi Font1
Institut de Ciencies del Mar CMIMA-CSIC
Passeig Maritim 37-49
Barcelona 08003
Spain

1Grup díOceanografia Interdisciplinar, IMEDEA (CSIC-UIB), Mallorca, Spain;
2Grup díOceanografia FÌsica, ICM-CSIC, Barcelona, Spain

jfont@icm.csic.es

The impact of assimilating Argo and altimeter data in the FOAM system

Matt Martin, Mike Bell and Adrian Hines
Met Office,
London Road.
Bracknell,
Berkshire,RG12 2SZ
United Kingdom

Matthhew.Martin@metoffice.com

Comparison of float profiles with an assimilating model (FOAM) in the North Atlantic.

Brian A King and Louise M Duncan
Southampton Oceanography Centre
Empress Dock
Southampton SO14 3ZH
United Kingdom

bak@soc.soton.ac.uk

Float profiles in the North Atlantic have been compared with output from a prognostic ocean GCM (OCCAM) and a GCM assimilating ocean data (FOAM). Significant differences have been found between each model and the reality represented by the float data. FOAM is run with a 300 km length scale for the assimilation of subsurface model-observation differences. The comparison suggests that the correct correlation scale for model errors has significant regional variation, and is often very much shorter. This has implications for the float density required to make a FOAM-type assimilation exercise fully effective.

UK ARGO Data Processing and Quality Control System

R.J. McCreadie & L.J. Rickards
British Oceanographic Data Centre - BODC
Joseph Proudman Building
6 Brownlow Street
Liverpool
L3 5DA
UK

G.J. Dawson
UK Hydrographic Office - UKHO

rebl@bodc.ac.uk

Delayed-mode QC of Argo float salinity in the Southern Ocean

Rebecca McCreadie, Juan Brown and Lesley Rickards,
British Oceanographic Data Centre - BODC
Joseph Proudman Building
6 Brownlow Street
Liverpool
L3 5DA
UK

Brian King
Southampton Oceanography Centre,
Southampton,
UK

rebl@bodc.ac.uk

UK Argo Progress and Plans

Jon Turton
UK Met Office,
London Road.
Bracknell,
Berkshire,RG12 2SZ
United Kingdom

jon.turton@metoffice.com

Global heat and freshwater storage measured by Argo

John Gilson and Dean Roemmich
Scripps Institution of Oceanography,
La Jolla CA 92093,
USA

jgilson@ucsd.edu

A year of oxygen measurements from Argo floats

Nordeen Larson
Sea-Bird Electronics, Inc.
1808 136th Place NE
Bellevue,
Washington 98005-2398
USA

norge@seabird.com

Decadal changes in upper ocean salinity in the North Pacific as inferred from Argo floats and historical observations

Ren Li and Stephen C. Riser
School of Oceanography
Seattle, Washington98195
USA

liren@ocean.washington.edu

New estimates of the heat budget in the tropical Atlantic, first results.


Claudia Schmid,
NOAA/AOML
4301 Rickenbacker Causeway
Miami, FL 33149
USA

S. L. Garzoli (NOAA/AOML), R. Lumpkin (CIMAS, University Miami), and Q. Yao (CIMAS, University Miami)

Claudia.Schmid@noaa.gov

A comparison of temperature profiles from profiling floats and XBTs.

Claudia Schmid
NOAA, AOML,
4301 Rickenbacker Causeway
Miami, FL 33149
USA

Claudia.Schmid@noaa.gov

Monitoring the oceans south of Africa - the need for Argo

I.J.Ansorge1, S Speich1,2, C.J.Reason1, J.R.E. Lutjeharms1

1Oceanography Department, University of Cape Town, Rondebosch,7701, South Africa

2LPO/UBO, University of Brest, Brest CEDEX, France.

ansorge@ocean.uct.ac.za

Mid-depth circulation in the North Pacific Ocean directly measured by Argo float

Koichi Ishikawa and Yasuki Tataksuki

Climate and Marine Department, Japan Meteorological Agency

k-ishikawa@met.kishou.go.jp

Intermediate circulation in the NorthWestern North Pacificderived from subsurface float movements (1998-2002)

Takanori Iwao

Hokodate Marine Observatory

t_iwao@met.kishou.go.jp