GLOBAL WATER MASS ANALYSIS

Location: Arts Building 120 LT

Location of Posters: Arts Building 120 LT and 101 LR4

Monday 19 July AM

Presiding Chair: Matthew England

Concurrent Poster Session

WATER MASS FORMATION PROCESSES

P12/W/03-A1 0830

SOME HISTORICAL, THEORETICAL AND APPLIED ASPECTS OF QUANTITATIVE WATER MASS ANALYSIS

Matthias TOMCZAK (School of Earth Sciences, The Flinders University of South Australia, GPOP Box 2100, Adelaide SA 5001,Australia, Email: matthias.tomczak@flinders.edu.au)

The concept of water masses is reviewed from the point of view of quantitative water mass analysis. A theoretical framework is presented which describes the life history of water masses in terms of formation, consolidation, ageing and decay. Water masses are described as physical entities and compared with their atmospheric counterparts (air masses). The classical temperature-salinity diagram is expanded into the mathematical concept of water types in an n-dimensional parameter space. Water types and their standard deviations are introduced as the foundation for quantitative water mass analysis. The relationship between parameter space and physical space is established through the definition of water type density. Mode Waters are discussed as regions in physical space with a minimum in water type density. Some unresolved issues of the structure of the oceanic thermocline are discussed in this context. The definition of water masses is extended to include water masses in the surface mixed layer where air-sea exchange processes continuously modify water mass properties. The paper concludes with a brief discussion of the representation of water masses and their evolution in numerical models.

P12/W/07-A1 0850

SUBDUCTION RATES FOR THE SOUTHERN INDIAN OCEAN THERMOCLINE

J. KARSTENSEN and D. Quadfasel 1) Institut für Meereskunde, University of Hamburg, Germany 1) now at: NBIfAFG, University of Copenhagen, Denmark

Using a kinematic as well as a transient tracer/oxygen approach, the subduction rates for water into the southern Indian Ocean thermocline were calculated and compared. For both techniques, the rates were found to be of the same magnitude with respect to the error margins. Comparing Ekman and Ekman+lateral induced rates in density increments of the winter mixed layer, the side-by-side existence of Indian Central Water and Mode Water is evident. The total volume of subducted water into the permanent thermocline was found to be 32 Sv for the density range 25.3 to 26.9 kg/m_. This is comparable to recent estimates for the North Atlantic (27 Sv) and North Pacific (35 Sv) Ocean. However, the proportion of the lateral volume transport in the Indian Ocean (21 Sv) is twice as large compared to the northern hemisphere oceans (NA 9.5 Sv; NP 10.1 Sv). This is in agreement with the large volume of Mode Water, which can be found in the Indian Ocean. The formation of the Mode Water through mid-latitude convection is located south of the subtropical front, whereas its subduction in the thermocline is lateral, as a combination of the eastward flow field along the front combined with the southern tilting of the front. Relating the rates to tracer distribution in the winter mixed layer depth, enabled us to determine source tracer characteristics of the Mode and Central Water, respectively. The characteristics may be used for further mater mass mixing analysis.

P12/W/22-A1 0910

TWO VARIETIES OF SUBTROPICAL LOWER WATER (SLW) IN THE SOUTH PACIFIC: THEIR ORIGIN AND DISTRIBUTION

Serguei SOKOLOV and Stephen Rintoul (both at CSIRO Marine Research and Antarctic Cooperative Research Centre, GPO Box 1538, Hobart, Australia, emails: Serguei.Sokolov@marine.csiro.au and Steve.Rintoul@marine.csiro.au)

Two varieties of SLW are present throughout the western part of the South Pacific: a cool, fresh variety is situated north of New Zealand, and a warm, salty variety is found north of Fiji and the New Hebrides. New data from a roughly meridional WOCE section P11 along 155E between the Subtropical Front and Louisiade Archipelago at 12S occupied in winter 1993 help refine some details of the SLW origin and distribution in the region. The warm and salty northern variety of SLW is formed in the central South Pacific near the Society Islands where a surface salinity maximum coincides with a maximum in evaporation minus precipitation. The subsurface salinity maximum of the SLW in the western South Pacific is a result of these high salinity surface waters being carried west in the northern arm of the subtropical gyre and under-riding lighter, low salinity water produced by an excess of precipitation over evaporation in the western tropical Pacific. South of 25S on P11 the salinity maximum reaches the sea surface marking the northern boundary of the cooler, fresher, denser southern variety of SLW. The southern boundary of the southern SLW on P11 is marked by the strong near-surface horizontal salinity gradient associated with part of the Subtropical Front at 38S. The formation region of the southern type of SLW - the Central Tasman Sea - also coincides with a maximum excess of evaporation over precipitation. This region also experiences strong cooling by the atmosphere, driving winter convection. Deep mixing during winter increases the oxygen content of the surface layer. The water which feeds the formation zone of southern SLW is the northern variety of SLW. Cooling and evaporation convert about 5 Sv of northern SLW to southern SLW in the Tasman Sea. Ventilated SLW carried eastward by the EAC spreads around the subtropical gyre of the South Pacific and is driven northward beneath the Tropical Convergence zone by Ekman pumping (Morris at al., 1996). The high-oxygen southern SLW under-rides the newly formed northern SLW in the central Pacific. The two super-posed varieties of SLW are carried westward to the Coral Sea in the SEC along the northern flank of the subtropical gyre. Here the two components of SLW are separated by only 40-60 m in depth and become almost indistinguishable within the cyclonic gyre of the Coral Sea. The total net inflow of SLW into the Coral Sea in the neutral density layer between 23.6 and 26.8 is 32 Sv, with 10.1 Sv re-circulating in the Gulf of PNG and entering the Solomon Sea. The remainder of the SLW turns south to feed the EAC.

P12/W/02-A1 0930

THE ORIGIN OF WATERS OBSERVED ALONG 137°E

Frederick BINGHAM (UNC-Wilmington, Wilmington, NC 28403 USA; Email: binghamf@uncwil.edu) Toshio Suga and Kimio Hanawa (both at: Department of Geophysics, Tohoku University, Sendai 980-77 Japan; Email: suga@pol.geophys.tohoku.ac.jp)

Using the World Ocean Atlas data set, we examine the origins and flow paths of waters observed along 137°E section in the western North Pacific. The method of Bingham (1999) was used to trace waters from 137°E back through the subtropical gyre to their outcrops. We divide the water masses observed along this section into four zones. (1) There is a zone of waters that connect directly to areas of surface subduction, chiefly the water mass known as North Pacific Tropical Water. Working backward along geostrophic streamlines, we find the locations of the outcrops of these waters. Subducted waters are aged using this technique and found to be between 0.5 and 35 years old by the time they reach 137°E. For this subducted regime, waters on a given isopycnal observed along 137°E increase in age with decreasing latitude, with waters at the southern end of the section being 2-3 times older than waters at the northern end. (2) There is a recirculating zone, where waters do not have any direct contact with the surface. Some of these recirculating waters, particularly the North Pacific Intermediate Water are strongly influenced by surface processes outside the subtropical gyre, but are not subducted by Ekman pumping. (3) There are North Pacific Subtropical and Central Mode Waters, which have direct contact with the surface, but mainly through buoyancy forcing rather than Ekman pumping. And (4) there is a seasonal thermocline, where waters are strongly influenced by surface heating and cooling, and isopycnals can disappear during the winter. We use these generalizations to characterize water mass variability observed in the PR-1 hydrographic section along 137°E.

P12/C/JSP21/E/01-A1 0950

INTERPRETATION FOR THE FORMATION OF THE CFC MAXIMUM IN THE NORTH PACIFIC

Sachiko OGUMA, Yutaka Nagata(both at Marine Information Research Center, Japan Hydrographic Association, Tokyo 104-0061, Japan) Goro Yamanaka(Oceanographic Division, Meteorological Research Institute, Tsukuba 305-0052, Japan) Toshio Suga and Kimio Hanawa(both at Department of Geophysics, Graduate School of Science, Tohoku University, Sendai 980-8578, Japan)

Chlorofluorocarbon (CFC) maximum layers have been found in the broad area of the Subtropical North Pacific. By using the results which contain the distributions of current and temperature and its variation, obtained by an Ocean General Circulation Model (OGCM) developed by Yamanaka (1997), and by giving spatial (depending on surface temperature) and temporal variations of CFC at the sea surface, we calculate CFC distributions in the North Pacific, and compare them with the observed CFC distribution, especially in the cross-section along 30N. The main feature of the CFC distribution including its maximum layer is well reproduced. In order to find the source area of CFC maximum layer water, trajectory analysis is also made. It is shown that the CFC maximum layer water is originated near the bottom of the winter mixed layer near 39N of the central part of the North Pacific, along the line of outcrop of 25.8 sigma_theta isopycnal surface. However, the distribution derived from an assumption that the CFC concentration is simply advected along trajectories by keeping the values in the injected area is much different from that reproduced in our model which includes the effect of eddy-diffusion. Distribution of CFC concentration would depend not only on advection but also on eddy-diffusion. Using a simple box model checks the relative importance of the advection to diffusion. It is suggested that this ratio is different between in the upper half and in the lower half of the CFC maximum layer. The position of the CFC maximum layer may be located between the regimes where diffusion is dominant and where advection is dominant.

P12/E/03-A1 1010

PHYSICAL MECHANISMS OF NORTH PACIFIC INTERMEDIATE WATER FORMATION

Nikolai A. MAXIMENKO (Shirshov Institute of Oceanology, Russian Academy of Sciences, 36, Nahimovskii prospect, Moscow 117851, Russia, email: maximenko@stream.sio.rssi.ru)

Results of isopycnal analysis of mean distribution of seawater properties in the North Western Pacific along with the propagation of seasonal signal point at the western part of the Kuroshio Extension as a formation area of the North Pacific Intermediate Water (NPIW). In addition to critical analysis of conventional theory of dominant role of lateral mixing, some new hypotheses are forwarded. Among them are: cabbeling and frontal convection, density-driven intrusions, geostrophic adjustment both in the confluence zone of subarctic and subtropical watermasses and downstream the current (due to dissipation of its zonal momentum), non-turbulent vertical transfer of zonal momentum (wave radiation and baroclinic interaction, resulting in "form drag"- type effects), interaction between thermocline and deep circulation. In all the cases specific forcing may appear in NPIW layer, making it drifting ageostrophically equatorward. Developed is a scheme of measurements and historical data reanalysis that may verify the importance of this physical mechanisms.

Presiding Chair: Matthias Tomczak (FIAMS, Flinders University)

Introduction 1030

P12/W/15-A1 Poster 1030-01

VARIABILITY OF THE NORTH ATLANTIC WATER MASS STRUCTURE AND CIRCULATION

S.A.DOBROLIUBOV (Department of Oceanology, Moscow State University, Vorobievy Gory, Moscow, 117234, Russia, e-mail: dobro@ocean.geogr.msu.su), A.V.Sokov and V.P.Tereschenkov (both at P.P.Shirshov Institute of Oceanography, Nahimovskiy pr., 36, Moscow, 117851, Russia, e-mail: boba@gulev.sio.rssi.ru)

Interdecadal changes of thermohaline characteristics and circulation of the intermediate and deep waters in the North Atlantic are considered. The results are based on the comparison of the Russian 1997 survey of 60N and WOCE AR5 with historical data. Significant changes of intermediate and deep waters T,S properties are found, accompanied by changes of the cross section mass, heat and freshwater fluxes. The earlier determined intensification of the Meridional Overturning Circulation (MOC) in the 80-s is confirmed. The MOC strengthening is accompanied by the rapid decrease of the recirculation flows and the formation of the well-defined two-layer structure of the MOC. The derived changes of the thermohaline structure suggest that at the phase of intensified MOC the preferable direction of the Labrador Sea Water spreading is north-east into the Irminger Basin, while at the period of reduced MOC the LSW flows preferably to the east and south-east. Enhancement of the southward freshwater transport in the Denmark Strait was closely correlated with the intensification of the northward meridional heat transport; the latter varied from 0.28 PWt to 0.38 PWt at 60N.

P12/W/16-A1 Poster 1030-02

INTERDECADAL CHANGES OF MASS AND MERIDIONAL HEAT TRANSPORT IN THE

NORTH ATLANTIC.

A.V.SOKOV, V.P.Tereschenkov (P.P.Shirshov Institute of Oceanography, Nahimovskiy pr., 36, 117851, Moscow, Russia; e- mail: sokov@gulev.sio.rssi.ru) and S.A.Dobroliubov (Department of

Oceanology, Moscow State University, Vorobievy Gory, Moscow, 117234, Russia, e-mail: dobro@ocean.geogr.msu.su)

Estimates of meridional mass and heat transports across the planes of transatlantic hydrographic sections along 24.5N, 36N, 48N and 59N (computed for the three time periods 1957-1962, 1981- 1982 and 1991-1993) are considered. The intensity of the Meridional Overturning Circulation (MOC) didn’t change significantly at 59N during the whole time period. While remarkable changes were detected to the south from 59N. Strengthening of MOC occurred at 48N and 36N in 1981-82 (19Sv and 20Sv respectively) against 14Sv and 12Sv in 1991-93 and 9Sv and 8Sv in 1957-59. Intensification of MOC at 24.5N is less profound. The results demonstrate negative correlation between the variability of MOC and Labrador Sea Water (LSW) production rate. The MOC intensification observed in 1980s corresponds to the increase of the mass transport in the Denmark Strait Overflow waters (DSOW) layer at 59N and is accompanied by the increase of the Antarctic Bottom Water (AABW) transport at 24.5N and its upwelling in the area between 24.5N and 36N. During the weak MOC (in the 1990s) the LSW production reached its maximum, the transport of DSOW reduced and the AABW penetrated further to the north than 36N.

P12/W/12-A1 Poster 1030-03

INTERANNUAL VARIABILITY OF THE UPPER OCEAN CIRCULATION AT 36N IN THE ATLANTIC OCEAN.

V.P.TERESCHENKOV (P.P.Shirshov Institute of Oceanography, Nahimovskiy pr., 36, Moscow, 117851,Russia, e-mail: boba@gulev.rssi.sio.ru) and A.V.Arkhipkin (Department of

Oceanology, Moscow State University, Vorobyivy Gory, Moscow, 117234, Russia, e-mail: arkhip@ocean.geogr.msu.su)

Interannual changes of the ocean steric height at 36N in the Atlantic ocean are considered. The study is based on the temperature and salinity observations in the upper 2000m layer collected by repeated surveys of the cross-ocean hydrographic section during 1972-1984 time period. Variability of the upper layer steric height is inspected in terms of water mass properties variability. Special attention is paid to the evolution of subtropical (18 degree) and subpolar mode waters and Mediterranean waters, regarding their remote relation to the changes of the atmospheric conditions over the sites of their origin. The changes of the upper ocean zonal pressure gradients are discussed in sense of North Atlantic circulation features, such as Gulf Stream, its recirculation jets and Azore Current.

P12/E/01-A1 Poster 1030-04

WATER MASS, TEMPERATURE, SALT AND HEAT EXCHANGE OF ANTARCTIC AND NORTH ATLANTIC BASINS.

ANISIMOV M.V., Ivanov Yu.A., Lebedev K.V. and M.M. Subbotina P.P.Shirshov Institute of oceanology, 117851,Nakhimovsky pr.36, Moscow, Russia, Tel.: (095)124-7729; fax: (095)124-5983;

e-mail: MMS@sio.rssi.ru, MVA@sio.rssi.ru

On the base of ocean general circulation model with using of climatological massive (Levitus, 1994), we calculated temperature, salinity, density, free surface level and current fields in the North Atlantic and Arctic basins. We defined the integral mean annual and mean seasonal characteristics of water mass, salt and heat transports. Also we fulfilled the analysis of inter seasonal redistribution of water mass, salt and heat transports for climatic seasons. Estimates of climate averaged variability of integral transports were made.

P12/E/06-A1 Poster 1030-05

THE SPREADING OF RED SEA OVERFLOW WATERS IN THE INDIAN OCEAN

LISA M BEAL and Amy Ffield (Lamont-Doherty Earth Observatory, Palisades, NY

10964, USA. Email lbeal@ldeo.columbia.edu)

As a result of its remarkably high salinity and low oxygen content, remnants of Red Sea Waters (RSW) have been identified in the southwest Indian Ocean, over 50ƒ south of their source. Here we describe the mean pathway for the spreading of RSW throughout the Indian Ocean. A

comprehensive set of observations is used, taken from the National Oceanographic Data Center archives and from the World Ocean Circulation Experiment Hydrographic Program for the Indian Ocean. The analysis suggests that the large scale salinity and oxygen distribution of the Indian Ocean is stationary at intermediate depth, indicating that there is little monsoon or interannual variability. From property maps the spreading of RSW appears to be favoured along the western boundary, particularly south of the westward flowing South Equatorial Current, where there is a clear tongue of RSW spreading through the Mozambique Channel and into the Agulhas Current. The property distributions imply a predominantly diffusive mixing regime in the ocean interior with a stronger advective component at the western boundary. The evidence for a long-time-mean pathway for RSW into the Agulhas Retroflection region suggests a consistent supply of saline

intermediate waters which may be captured in Agulhas Rings and exported to the South Atlantic. A quantitative estimate of the transport of RSW in the Agulhas Current at 32S suggests that almost all of the Red Sea outflow waters end up in the WBC.

P12/E/02-A1 Poster 1030-06

WATERMASS STRUCTURE AND TRANSPORT IN THE INTERMEDIATE LAYERS OF THE SOUTH INDIAN OCEAN

BENNY N. PETER & Vimal Kumar (Department of Physical Oceanography, Cochin University of Science & Technology, Fine Arts Ave, Cochin, India 682016, email: benny@md2.vsnl.net.in)

The watermass assembly in the intermediate layers of the South Indian Ocean and its exchange with the Antarctic Ocean(at 32 degree south) is described in the present study. The hydrographic data procurred from the National Oceanographic Data Cente, washington, USA and National Research Institute of Far Seas Fisheries, Shimizu, Japan are used. Isanosteric Analysis is employed to identify the watermass movement. The distribution of potential temperature, salinity and depth on 140, 120, 100, 80, 60 cl/ton surfaces are presented. The method introduced by Montogomery and Stroup(1962), depicting the transport in the T-S diagram, is adopted to quantify the watermass exchange. The watermasses in the South Indian Ocean is much influenced by the Indonesian Throughflow, even in the intermediate layers. Also, the cross equatorial spreading is weakened towards deeper levels. The intrusion of Antarctic waters are more confined to the central part, while the return flow is taking place near the continential boundaries. The Indonesian Throughflow adds more low saline cool water to the South Indian Ocean. The thermohaline front which seperates the cool, less saline Pacific and saline, warm Indian waters is weakened with respect to depth. In the intermediate layers a net southward transport of 5.84Sv is observed. The major contribution of low saline Antarctic Intermediate Water towards north is confirmed from the flux distribution in the T-S diagram.

P12/E/07-A1 Poster 1030-07

WATER MASSES OF COASTAL REGIONS OF THE FAR-EASTERN (THE JAPAN, OKHOTSK, AND BERING) SEAS

Gennady YURASOV ( Pacific Oceanological Institute, Russian Academy of Sciences, Far-Eastern Branch, Baltiyskaya Str., 43, 690041 Vladivostok, Russia, email: pacific@online.marine.su)

Yury Zuenko (Pacific Fishery and Oceanography Research Institute, Shevchenko Alley, 4, 690600 Vladivostok, Russia).

General principles of water mass analysis based on statistical methods are applied to all listed seas. Materials of the long-term seasonal surveys of coastal regions are examined for this purpose. Using proposed technique a classification of water masses and types of water vertical structure was conducted.

Influenced by the same formation mechanism at all seas the water masses produce an estuaric (at the Okhotsk Sea only), coastal, shelf and subarctic types of the vertical structure. All of them have their attributed stratification degree and characteristics of water masses. In the regions of complete tidal mixing the water masses are homogenous from the sea surface to the bottom. The same named water masses for each of the listed seas are differed each other by values of their parameters (temperature, salinity, depth) and sizes of their extension. During winter season at all three seas the water masses of higher density are being formed. As a result of convection and mixing these waters ventilate deep layers. Particularly this mechanism develops at the Japan Sea where the shelf is of insignificant extent. It is supposed that the higher density waters of the Okhotsk Sea ventilate the layer of lower salinity in the North Pacific Ocean.

P12/W/19-A1 Poster 1030-08

A THREE-LAYER MODEL FOR THE STUDY OF TIDALLY INDUCED DENSITY CURRENTS

Rainer Weigle, Peter BRANDT, and Angelo Rubino (Institut für Meereskunde, Universität Hamburg, Troplowitzstr. 7, D-22529 Hamburg, Germany, Email: brandt@ifm.uni-hamburg.de)

The dynamics of tidally induced internal waves is investigated by using a numerical three-layer model based on the weakly nonlinear, weakly non-hydrostatic Boussinesq equations. Numerical solutions of the three-layer Boussinesq equations are compared with analytical solutions of different equations describing internal solitary waves in a two-layer system. In contrast to the two-layer models, the three-layer model is capable of describing the evolution of sub-surface jets. As a result of the three-layer model these jets develop undulations that finally disintegrate into internal solitary waves. Several characteristics of these waves, like e.g., wave amplitude and distance between the first two solitary waves of a wave train are studied as a function of the initial jet velocity and the stratification. Results of the numerical simulations are compared with results of high resolution in-situ measurements carried out north and south of the Strait of Messina, in the European Mediterranean Sea. Implications of the presented study for a possible inversion of sea surface manifestations of oceanic internal waves into characteristics of the interior ocean are discussed.

P12/W/20-A1 Poster 1030-09

GROWTH MECHANISM OF TOPOGRAPHIC INTERNAL WAVES GENERATED BY AN OSCILLATORY FLOW

Tomohiro NAKAMURA and Toshiyuki Awaji (Department of Geophysics, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan, Email: nakamura@kugi.kyoto-u.ac.jp)

A new amplification mechanism for topographic internal waves generated by tidal flow is presented to reveal the unknown growth processes of the internal waves in the Kuril Straits (about 47N in the North Pacific) reported by Nakamura et al. (1998). Accordong to their investigation, the nature of the wave properties is determined by the nondimensional parameter (k U0)/sigma_f, where k is the curvature of topoguraphy, U0 is the flow speed amplitude, and sigma_f is the frequency of tidal flow, through which the topographic internal waves can be classified into the three wave types; (1)unsteady lee waves (k U0/sigma_f >> 1), (2)mixed tidal lee waves (k U0/sigma_f \sim 1), and (3)internal tides (k U0/sigma_f << 1). Notably, their numerical model showed that the growth processes of the first two waves were quite different from those predicted by conventional internal wave theories, thus suggesting the presence of another criterion for wave amplification. For this issue, our theoretical investigation based on the ray tracing of individual waves generated at various time reveals the following interesting facts. Unsteady lee waves are always amplified when the maximum frequency is sufficiently smaller than the buoyancy frequency (i.e., sigma_f << k U0 << N), because their phase speeds and amplitudes are equal and proportionate to the tidal flow speed at their generation, respectively. Fast mixed tidal lee waves are also effectively amplified as well as unsteady lee waves, when the rotation effect is significant (i.e., f \sim sigma_f \sim k U0 << N). Accordingly, amplification of unsteady lee waves and fast mixed tidal lee waves can occur even if the conditions indicated by previous theories (i.e., the critical slope and the critical Froude number) are not satisfied. Since our theoretical model covers generation and amplification processes of topographic internal waves in a broader parameter range than before, it may contribute to a better understanding of the Boundary Mixing processes.

P12/W/06-A1 Poster 1030-10

WATER MASS, TEMPERATURE, SALT AND HEAT EXCHANGE OF ANTARCTIC AND NORTH ATLANTIC BASINS.

Anisimov* M.V., IVANOV* Yu.A., Lebedev* K.V. and SUBBOTINA* M.M. *P.P.Shirshov Institute of Oceanology

On the base of ocean general circulation model with using of climatological massive (Levitus,1994) we calculated temperature, salinity, density, free surface level and current fields in the North Atlantic Antarctic basins. We defined the integral mean annual and mean seasonal characteristics of water mass, salt and heat transports. Also we fullfiled the analysis of interseasonal redistribution of water mass, salt and heat transports for climatic seasons. Estimates of climate avereged variability of integral transports were made.

Presiding Chair: Bernadette Sloyan

WATER MASS CLIMATOLOGY

P12/E/04-A1 1110

WORLD OCEAN THERMAL STRUCTURE

Peter C. CHU, Chenwu Fan, and Hui Liu

We used gradient criteria to establish a global ocean mixed layer depth (MLD), thermocline depth (THD), and thermocline strength (THS) data sets from the National Oceanographic Data Center (NODC) 4.5 million temperature profiles. The annual cycles of MLD, THD, and THS of the world ocean are described based on climatological monthly-mean MLD fields. The interannual variabilities of MLD, THD, and THS are obtained from zonally averaged monthly mean fields as well as the latitudinally averaged (10 deg N and 10 deg S) monthly mean fields. The ocean interannual thermal variability is closely related to El Nino, North Atlantic Oscillation (NAO), and Great Salinity Anomaly in the Arctic regions.

P12/W/11-A1 1130

WATER MASS CHANGES IN THE SCOTIA SEA

David P. STEVENS (School of Mathematics, University of East Anglia, Norwich NR4 7TJ, email D.Stevens@uea.ac.uk) Karen J. Heywood (School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, email k.heywood@uea.ac.uk)

In March-April 1999 two South Atlantic WOCE sections, A21 (Drake Passage) and (part of) A23 (35 degrees west), will be reoccupied as a component of ALBATROSS (Antarctic Large-scale Box Analysis and The Role Of the Scotia Sea). WOCE section A21 was occupied in February 1990 and section A23 was occupied in March 1995. Early results from ALBATROSS will be presented. In particular we will focus on an analysis of differences between ALBATROSS and the earlier WOCE sections. Both the WOCE and ALBATROSS data include a full suite of chemical tracers (such as CFCs and nutrients) as well as traditional hydrography. Assessments will be made of the role of the observed water masses in climate variability.

P12/W/10-A1 1150

WATER MASS CHANGES IN THE NORTH AND SOUTH PACIFIC OCEANS BETWEEN THE 1960's AND 1985-94, IMPLICATIONS FOR CLIMATE CHANGE

Annie P.S. WONG (IASOS, University of Tasmania, GPO Box 252-77, Hobart, 7001, Australia), Nathaniel L. Bindoff (Antarctic CRC, GPO Box 252-80, Hobart, 7001, Australia, N.Bindoff@utas.edu.au) and John A. Church (Antarctic CRC and CSIRO Marine Research, GPO Box 1538, Hobart, 7001, john.church@marine.csiro.au).

Comparisons have been made along five modern hydrographic sections with historical data to investigate water mass changes in the North and South Pacific Oceans. The five modern hydrographic sections were sampled in the decade 1985-94, while the historical data were mostly from the 1960s.

Below the seasonal mixed layer, statistically significant temporal differences in temperature and salinity have been detected in the water masses that occur in the top 2000 dbar of the water column. These differences in water mass property changes are assumed to result from changes in sea surface conditions at the formation regions.

Of the observed temporal differences, the ones with largest scale are from the shallow salinity maxima of North Pacific Subtropical Water (NPSTW) and South Pacific Subtropical Water (SPSTW), and the intermediate salinity minima of North Pacific Intermediate Water (NPIW) and Antarctic Intermediate Water (AAIW). The two salinity maxima, NPSTW and SPSTW, also show signs of salinity increase, while the two salinity minima, NPIW and AAIW, have become warmer and fresher. Since NPSTW and SPSTW originate under the high evaporative cells of the central North and South Pacific, and NPIW and AAIW acquire their properties near the polar gyres, these changes in the ocean interior imply an increase in evaporation over the mid-latitudes, and an increase in precipitation over the high-latitudes regions in both hemispheres. Taken together these results imply a strengthening of the hydrological cycle over the North and South Pacific Oceans. Output from coupled climate model for increasing atmospheric CO_2 level show that the model ocean responds with a warming of the water column. Superimposed on this background warming trend is a decrease in salinity in the salinity minima of the Pacific (AAIW and NPIW) which corresponds to near-surface freshening where their respective isopycnals outcrop. Hence the freshening signature that has been detected in NPIW and AAIW from observational data is qualitatively consistent with the results from this climate model for increasing atmospheric CO_2.

Monday 19 July PM

WATER MASSES IN MODELS

P12/W/23-A1 1400

WATER MASS ANALYSIS IN OCEAN MODELLING

Matthew H. ENGLAND, Centre for Environmental Modelling and Prediction (CEMAP), The University of New South Wales, Sydney NSW 2052, Australia M.England@unsw.EDU.AU

Water mass analysis forms a vital component of ocean model assessment. I will present an overview of the fidelity of models in the context of the global scale water masses, summarising the key processes and parameterisations required for large-scale simulations of water mass renewal. I will also outline how ocean modellers try to verify water mass formation processes in global ocean simulations. Proxy model diagnostics of ventilation rates, such as "age" and age distributions, will also be discussed in the context of ocean model assessment.

P12/W/05-A1 1420

USE OF PASSIVE TRACERS TO ANALYSE WATER MASSES IN CONTROL AND TRANSIENT CO2 SIMULATIONS

Siobhan P. O'FARRELL, CSIRO Atmospheric Research Aspendale, 3195, Australia. email : Siobhan.O'Farrell@dar.csiro.au

Due to the high frequency component of atmospheric forcing, the ocean response when coupled to an atmospheric model differs from that seen with imposed surface boundary conditions. In particular, the rates of convection, depth of mixed layer, and meridional overturning rates are altered at high latitudes under coupled conditions. A passive tracer has been used in the ocean model to explore differences between spin up and coupled experiments as an analogue to some of the chemical species that are regularly measured on ocean hydrographic sections. There are two versions of the ocean model, a version with standard horizontal diffusion and a second with the Gent and McWilliams eddy mixing. Simulations with both ocean versions have included tracers for 50 years in the coupled ocean atmosphere model and in ocean-only runs. The tracer output is discussed in terms of water masses for individual ocean basins.

Additionally, there are 50 years of tracer data from a transient CO2 experiment with the coupled model. The tracer was released at a time when the CO2 level is held constant in the atmosphere at double the value of the control simulation, after the ocean adjustment to the new surface conditions has slowed. A further set of passive tracer data is now being gathered in a transient simulation, as CO2 is increased from 1880 levels to those projected for the end of the next century. Both these tracer data sets will be used to examine how different water masses in the ocean are responding to the climate change signal.

P12/W/13-A1 1440

RENEWAL AND MODIFICATION OF ANTARCTIC INTERMEDIATE WATERS

Bernadette SLOYAN, Alfred Wegener Institute for Polar and Marine Research, Postfach 12 01 61, D-27515 Bremerhaven, GERMANY

Subantarctic Mode water (SAMW) and Antarctic Intermediate water (AAIW) are characterized by a dissolved oxygen maximum and salinity minimum north of the Antarctic Polar Front zone (APF). These water masses move eastward with the Antarctic Circumpolar Current (ACC) and northward into the adjacent Atlantic, Indian and Pacific subtropical Oceans. Along this circulation path significant property changes occur to both water masses. Many mixing mechanisms and their principal location, which results in the observed property changes, have been suggested.

Estimates of interior and air-sea diapycnal fluxes for the southern oceans, across neutral surfaces identifying SAMW and AAIW, are derived from basin-scale budgets of mass, heat and salt using a box inverse model. The diapycnal fluxes quantify mixing occurring in the subtropical oceans and across the APF. The inferred fluxes help us explain the property changes in SAMW and AAIW observed across each ocean basin.

P12/W/04-A1 1500

ZONAL FLUXES IN THE DEEP LAYERS OF THE SOUTH ATLANTIC

Michael VANICEK (Department of Marine Physics, Institut fuer Meereskunde Kiel, Duesternbrooker Weg 20, 24105 Kiel, Germany, email: mvanicek@ifm.uni-kiel.de) Gerold Siedler (Department of Marine Physics, Institut fuer Meereskunde Kiel, Duesternbrooker Weg 20, 24105 Kiel, Germany, email: gsiedler@ifm.uni-kiel.de)

The circulation of the North Atlantic Deep Water (NADW) in the South Atlantic is determined from hydrographic, nutrient, and tracer data from WOCE and other high quality pre-WOCE sections using a linear box-inverse model. Multiple linear regression, which makes use of the correlation between different parameters, is applied to infer the missing parameters in the bottle data set. This interpolation technique also enables us to include the nutrient and tracer measurements in the inverse model with a spatial resolution of the corresponding CTD data. The data define a set of 126 closed boxes on which conservation requirements are imposed. A detailed water mass analysis is performed, incorporating the tracer information from the whole South Atlantic, to determine the vertical boundaries of these boxes. As a result the water column is divided into 11 layers which are defined by neutral densities. Constraints for the inverse model are an integral meridional salt and phosphorus transport, the overall salt and silica conservation, as well as flow conditions inferred from moored current observations. The results are analysed with an emphasis on the zonal spreading of the NADW. A clear meridional separation in the direction of the zonal NADW transports can be observed. Quantitative estimates are given.

P12/E/05-A1 1520

MEAN CIRCULATION AND TURBULENCE IN THE EASTERN NORTH ATLANTIC, FROM A REGIONAL NUMERICAL MODEL.

Thierry PENDUFF (Laboratoire des Ecoulements Geophysiques et Industriels, BP53 38041 Grenoble Cedex 9, France, email Thierry.) Penduff@hmg.inpg.fr Alain COLIN DE VERDIERE (Laboratoire de Physique des Oceans, Universite de Bretagne Occidentale, BP 809, 29280 Brest cedex, FRANCE, email acolindv@univ-brest.fr) Bernard BARNIER (Laboratoire des Ecoulements Geophysiques et Industriels, BP53 38041 Grenoble Cedex 9, France, email Bernard.Barnier@hmg.inpg.fr)

An eastern North Atlantic regional configuration of SPEM model is used to study the general circulation in the basin, the mesoscale turbulence, and the interactions between these two scales. Contrary to previous studies, and to ensure consistency between lateral forcing and inner dynamics, the velocity field along the wide open boundaries is largely determined by the model itself. This approach can therefore be considered as a prognostic alternative to inversions; after stabilization, the general circulation is qualitatively and quantitatively close to available data. It is shown that in the North Atlantic Current, baroclinic instability generates eddies that in turn strengthen the mean flow, and that these processes are intensified on the polar side of the fronts; avaliable data seems to confirm this features.

P12/W/08-A1 1600

INFLUENCES ON NEAR-SURFACE VELOCITY FIELDS DUE TO STRONG VERTICAL MIXING IN THE INDONESIAN SEAS

Masatora IIDA, Toshiyuki Awaji, Yoichi Ishikawa, Kazunori Akitomo, Teiji In, Nobumasa Komori, Takaki Hatayama, Tomohiro Nakamura (Department of Geophysics, Graduate School of Science, Kyoto University, Kitashirakawa-Oiwake, Sakyo, Kyoto 606-8502, Japan, email: iida@kugi.kyoto-u.ac.jp), and Bo Qiu (Department of Oceanography, University of Hawaii at Manoa, 1000 Pope Rd., Honolulu, Hawaii 96822, U.S.A., email: bo@iniki.soest.hawaii.edu)

Impacts of strong vertical mixing caused by tidal mixing in the Indonesian Seas are investigated with emphasis on how much this regional mixing has influences on the near-surface velocity fields over the Pacific and Indian Oceans by using an extended reduced gravity layer model, which is basically similar to that of McCreary and Kundo (1989) but is widely extended. Our reduced gravity layer model has dynamically two and thermodynamically three active layers. The first dynamical layer has two thermodynamical layers; the upper one is called the 'mixed layer' which includes a bulk mixed layer model based on Kraus-Turner thermodynamics, and the lower one is called the 'fossil layer'. Within this layer, no vertical temperature gradients exist but horizontal gradients do. This means that density can vary within the layer. In addition, our layer model can deal with vertical heat / momentum exchanges due to entrainment / detrainment.

Using this layer model, we have performed two numerical experiments with and without strong vertical mixing in the Indonesian Seas. Comparison between both results revealed many interesting facts about the influences of the regional vertical mixing on the near-surface velocity fields of the equatorial Pacific and Indian Oceans. For example, the South Equatorial Current in the Pacific Ocean is much intensified, leading to the significant increase in transport of the South acific water into the Indonesian Seas. This is very consistent with the observational result revealed by Ffield and Gordon (1992). Our detailed analysis showed that such an increase in the volume transport of the South Pacific Water flowing into the Indonesian Seas is associated mainly with the occurrence of upwelling Kelvin waves in the western equatorial Pacific due to the effect of strong vertical mixing in the Indonesian Seas and their propagation along the equator.

P12/W/21-A1 1620

THE EFFECT OF THE SUBTROPICS ON THE TROPICAL SALINITY FIELD

Masami NONAKA (Institute of Low Temperature Science, Hokkaido University, Sapporo 060-1809, Japan, Email:nona@lowtem.hokudai.ac.jp) Kensuke Takeuchi (Institute of Low Temperature Science, Hokkaido University, Sapporo 060-1809, Japan, Email:takeuchi@lowtem.hokudai.ac.jp)

In higher latitudes, more saline water compared with that in the tropics is supplied through the sea surface and is advected to the tropics by the ocean circulation connecting the subtropics and the tropics. Using a ocean general circulation model (GCM) which has realistic topography and is forced by the climatological data, we examine the effect of water directly transported to the tropics on the tropical salinity fields and the reason for the discrepancy between the salinity and tritium fields' properties: Tritium indicates an interior route to the equator in the Northern Hemisphere, but high salinity water does not indicate it. The results of our GCM show that the water subducted into the subsurface from the high sea surface salinity (SSS) regions in the subtropics is advected to the tropics, forming a high salinity tongue. In the Southern Hemisphere, the high salinity tongue extends directly to the equatorial region, but in the Northern Hemisphere it extends to the western boundary. In the Northern Hemisphere, water subducted from the low SSS region in the eastern part of the subtropics and water whose salinity is lowered by mixing with the eastern low salinity water extend to the equatorial region through the interior ocean, forming a low salinity tongue. The ocean circulation in our GCM follows an interior route from the North Pacific to the equator, consistent with the subsurface tritium field, and, at the same time, reproduces the properties of the subsurface tropical salinity field. Because the northern interior route transports only low salinity water, the northern high salinity tongue does not follow an interior route and is discrepant from the tritium fields. The discrepancy is caused by the difference of the sea surface distribution: High SSS exists in the western boundary exchange window in the Northern Hemisphere, while sea surface tritium has its maximum near the eastern boundary within the interior exchange window. Here, water subducted from the western boundary (interior) exchange window is transported to the equatorial region through the western boundary region (the interior ocean) in the subsurface layer.

P12/W/18-A1 1640

WATER MASS TRANSFORMATIONS IN THE NORTH PACIFIC OCEAN

Dr Maxim YAREMCHUK, International Pacific Research Center

A dynamically consistent state of the North Pacific ocean has been obtained by the variational assimilation of climatological data into a non-linear steady state circulation model. The assimilated data include Da Silva heat, freshwater and momentum fluxes at the ocean surface, WOCE hydrology, five year (1992-1997) mean Topex-Poseidon altimetry and MEDS surface drifter trajectories, averaged over the period 1990-1996. Advective balance residuals in the optimized fields of potential temperature and salinity are treated as transformation rates of waterproperties caused by local diffusive and mixing processes. Statistical analysis of these "elementary"transitions in the temperature-salinity space provides their decomposition into thetransformation rates between the basic water masses of the North Pacific ocean. The analysis shows in particular, that North Pacific Intermediate Water (NPIW) plays a central role in the water mass balance of the region. Isopicnal flow patterns indentify intensive NPIW divergencies east of Southern Kuril Islands and Hokkaido. A much weaker possible source of NPIW is locatedin the Alaska Gyre. Zonal mean circulation of the optimal state exhibits 8.5 Sv of the deepwaters inflowing from the south which are then involved into the meridional upwelling of 14 Sv between 30-50N. A weaker (5 Sv) overturning cell exhists north of 50N which is driven by the downwelling of the north Bering Sea waters.

REGIONAL WATER MASS STUDIES

P12/W/09-A1 1700

ORIGIN OF DEEP AND BOTTOM WATER IN THE SOUTHWEST PACIFIC

Serguei SOKOLOV and Stephen Rintoul (both at CSIRO Marine Research and Antarctic Cooperative Research Centre, GPO Box 1538, Hobart, Australia, emails: Serguei.Sokolov@marine.csiro.au and Steve.Rintoul@marine.csiro.au)

The majority of the hydrographic stations occupied in the Tasman/Coral Sea have been shallow (<1500 m deep). As a result, there has been little discussion of the deep circulation. Recent high quality deep hydrography,including WOCE P11, helps refine some details of the deep water circulation in the region. According to P11 data, there are two major sources of the deep water in the Coral and Solomon Seas: the "southern" source supplied from the East Australian Basin and the "eastern" source entering the Coral Sea from the east. About 3 Sv of deep water is carried into the Coral Sea by a deep western boundary current in the Cato I. Trough from the Tasman Sea, and about 6 Sv (relative to the bottom) of deep and bottom water recirculate in the Coral Sea across P11. The deep and bottom water entering the Coral Sea from the east are derived from the CDW spreading north from the ACC as a western deep boundary current along the Tonga-Kermadec Ridge, through the Samoa Passage, and finally across the equator to the North Pacific. While the densest water in the boundary current is prevented from spreading to the west by topography, lighter CDW spreads west in the East Mariana Basin and also supplies deep water to the eastern Coral Sea and the Solomon Sea. The deep waters in the Solomon Sea are horizontally homogeneous throughout the basin, and the theta-S curves found for the New Britain Trench and near the northern entrance to the Solomon Sea at the Vitiaz Strait, St. Georges Channel and the Solomon Strait are almost identical to those found in the South Solomon Trench. This is a clear indication that the deep waters in the Solomon Sea enter from the east. The CDW entering from the east is cooler and fresher than water entering the Coral Sea from the East Australian Basin. The distribution of silica in the South Pacific also supports this view. CDW entering the Pacific from the Southern Ocean is initially low in silica (<90, see e.g. Schmitz, 1996), while North Pacific DW can have values as high as 180. The low silica in the Coral Sea Basin is an indicator of the "southern"source of deep water supplied from the Tasman Sea, while the higher values in the eastern part of the Coral Sea reflect deep water inflow derived from the "eastern"source.

P12/W/17-A1 1720

STRAIT OF SICILY WATER MASSES

Alex WARN-VARNAS (Naval Research Laboratory at Stennis, USA,email varnas@nrlssc.navy.mil), Allan Robinson, Jurgen Sellschopp, Wayne Leslie, Pat Haley, Carl Lozano, and Steve Piacsek

We have derived a water mass model for the Strait of Sicily, based on 1994 and 1995 cruise data. The model consists of seven water masses, suggested by the measured shapes of the vertical temperature and salinity profiles. The core of the Atlantic water is distributed below the surface as a shallow layer, in a depth range of 40 to 100 meters, with a salinity minimum. Upper and surface layers above and a mixed region below cap it. At the bottom Levantine water is present with a transition region above. Between the mixed and transition region there is, on occasion, a fresher water layer. The structure and statistics of the water masses is analyzed in terms of their temperature, salinity, and depth parameters. Objective analysis of the temperature, salinity, and depth parameters is performed in latitude and longitude. The water masses are tracked in terms of their parameter signatures. Changes in temperature and salinity distributions are interpreted. 2-D ellipses that represent the water masses, in terms of mean and standard deviation, are derived in a space of temperature, salinity, and depth. Their axes are the standard deviations of parameter space ranges.The areas of the ellipses are compared against the temperature and salinity distributions. The water mass composition ratios are computed and analyzed. Hypotheses and mechanisms for the origin and mixing of water masses are suggested. The temporal and spatial variability of the volume flows associated with the major water masses is studied via high horizontal resolution, 5km, and 64 level ocean model using 6-hour forcing and an embedded mixed layer.