Inverse Box Model Research Articles

Canary Current and North Equatorial Current from an inverse box model

The large‐scale Canary Basin circulation is estimated from a box inverse model applied to hydrographic data from a quasi‐synoptic survey carried out in September 2003. The cruise consisted of 76 full depth CTD and oxygen stations. Circulation is required to nearly conserve mass and anomalies of salinity and heat within layers bounded by neutral surfaces. It permits advective and diffusive exchange between layers and an adjustment of the Ekman transport and the freshwater flux divergences. The Canary Current at the thermocline layer transports a net mass of 4.7 ± 0.8 Sv southward north of the Canary Islands from the African coast to 19°W. It is divided into a northward circulation at a rate of 1.1 ± 0.5 Sv between the African coast and Lanzarote Island and a southward transport of 5.8 ± 0.6 Sv. It transports North Atlantic Central Water and organic matters advected offshore by the filaments protruding from the upwelling system off northwest Africa. At 24°N, the Canary Current feeds the North Equatorial Current that transports a mixture of North and South Atlantic Central Waters westward. In the intermediate layer a southwestward flow of 1.2 ± 1.1 Sv transports Mediterranean Water to the Subtropical Gyre, though the highest salt flux is transported by a meddy. Oxygen distribution and mass transport suggest a northeastward deep flow of a water mass colder than 2.2°C consisting of diluted Antarctic Bottom Water. The heat and freshwater divergences and the average dianeutral velocity and diffusion between the sections and the African coast are negligible.

Correction to “Water flux estimates in the central Mediterranean Sea from an inverse box model”

[1] In the paper “Water flux estimates in the central Mediterranean Sea from an inverse box model” by A. Vetrano et al. (J. Geophys. Res., 109, C01019, doi:10.1029/2003JC001903), typographical errors appeared in the following sentence in paragraph 63 in Appendix A: “There are N-K vectors Vi, I = K + 1, N and N-K vectors Ui, I = K + 1, N such that AVi = 0 and AUi = 0.” The corrected sentence is “There are N-K vectors Vi, I = K + 1, N and N-K vectors Ui, I = K + 1, N such that AVi = 0 and AUi = 0.” [2] In addition, multiple errors occurred in Tables 1 and 5. The corrected Tables 1 and 5 appear below.

Red Sea deep water circulation and ventilation rate deduced from the 3He and 14C tracer fields

3He is injected in the deep ocean at plate boundaries in relation with hydrothermal activity. Its oceanic distribution, which is at a steady-state, shows appreciable vertical and horizontal gradients. Hence, 3He conservation equations may be inverted to determine the flow field and mixing coefficients within the ocean. Here, we use this technique to investigate the deep circulation of the Red Sea, whose deep thermohaline circulation is comparable to that of a miniature world ocean. 3He data, which are a combination of the Geosecs, Meseda and Merou cruises, are inverted using a linear inverse box model. The present study allows us to draw the details of the internal circulation. The latter is characterized by (i) a descending branch in the northernmost part of the basin, which corresponds to the sinking and subsequent north–south movement of dense surface waters and of additional water entrained by the sinking plume, (ii) by an internal counterclockwise recirculating loop with a northward return flow at intermediate depth. In the literature, the rate and modes of renewal of the Red Sea deep waters are poorly constrained, with bulk residence time estimates ranging from a few decades to a few centuries. In our model, the deep water renewal rate is directly dependent on the magnitude of the 3He sources. The global 3He flux is estimated using two independent approaches. The first method is based on the calculation of the mean 3He transfer flux at the air–sea interface. The second approach relies on recent estimates of the global terrestrial 3He flux. Both methods agree within their respective uncertainties. The circulation scheme defined by the inversion is further constrained by simulating the bomb 14C distribution. Both isotopes, one steady-state tracer ( 3He) and one transient tracer ( 14C), lead to reasonable agreement. The renewal time corresponding to the sinking of newly formed deep water in the northern part of the basin is 60 years. However, the global ventilation rate, deduced from the simulation of the decay rate of a numerical tracer, is much faster (26 years). This result shows, as already pointed out by others, that deep water formation is not the only process by which the deep Red Sea is ventilated. Although this mechanism is important in our model, the basin wide deep circulation coupled with efficient vertical exchange between the deep basin and the upper layers appear to be an even more powerful mode of ventilation.

Water flux estimates in the central Mediterranean Sea from an inverse box model

Seasonal hydrological surveys and long‐term current measurements carried out during a wide investigation of the central Mediterranean, connecting the eastern and western parts of the basin with the Tyrrhenian Sea, are used to estimate the absolute geostrophic flow through the region applying an inverse methodology. The inverse box model prescribes conservation of water properties (mass, heat, and salt) inside closed boxes of water, as well as the imposition of mean fluxes by direct current measurements. The model result is a mean circulation of the region during four different periods, along with simultaneous evaluation of water fluxes through the three boundary sections and the associated uncertainties. The overall situation, besides confirming previous qualitative knowledge on the mean circulation at all levels, provides a novel quantitative pattern of the water mass exchange in a region that is crucial for the assessment of a more realistic budget for the whole basin. The positive results obtained indicate that the inverse method may be considered an effective powerful tool for successful applications at regional scale.

Water Mass Conversion, Fluxes, and Mixing in the Scotia Sea Diagnosed by an Inverse Model

Abstract An inverse box model of the Scotia Sea is constructed using hydrographic, tracer, and velocity data collected along the rim of the basin during the Antarctic Large-Scale Box Analysis and the Role of the Scotia Sea (ALBATROSS) cruise. The model provides an estimate of the time-mean three-dimensional circulation as the Antarctic Circumpolar Current (ACC) crosses the region. It concurrently solves for geostrophic and wind-driven Ekman transports across the boundaries of the basin, air–sea-driven diapycnal fluxes, and “interior” diapycnal fluxes below the ocean surface. An increase is diagnosed in the ACC volume transport from 143 ± 13 Sv (Sv ≡ 106 m3 s−1) at Drake Passage to 149 ± 16 Sv on leaving the Scotia Sea, supplied by the import of 5.9 ± 1.7 Sv of Weddell Sea Deep Water (WSDW) over the South Scotia Ridge. There is a lateral redistribution of the transport, primarily in response to a topographically induced branching of the 70–80 Sv polar front (PF) jet and an increase in the transport associa...

Upper-layer circulation in the eastern Equatorial and South Atlantic Ocean in January–March 1995

The upper-layer circulation in the eastern basin of the South Atlantic was studied from hydrographic and direct velocity measurements along WOCE lines A11, A13 and A14. A13 and A14 provide quasi-meridional samplings of the equatorial, subequatorial and subtropical circulation regimes. A13 was carried out along the African coast at about 600 km offshore from it, and A14 along the nominal longitude 9°W. A11 intersects the Cape Basin between 46°S in the west and 30°S in the east. Transport estimates were derived from direct velocity measurements and a box inverse model. In the equatorial eastern Atlantic, the Equatorial Undercurrent (EUC) transport decreases from 25×10 6 m 3 s −1 at 9°W to 13×10 6 m 3 s −1 at 5°E. Re-circulations of the EUC into the northern and equatorial branches of the South Equatorial Current (SEC) are evidenced at 5°E and quantified. In the tropical Atlantic, we estimate 7.5×10 6 and 4.2×10 6 m 3 s −1 for the transports at 9°W of the South Equatorial Undercurrent (SEUC) and South Equatorial Countercurrent (SECC), respectively. Both the SEUC and SECC extend vertically down to intermediate depths and contribute to the northern limb of the Angola Gyre. The Angola Current transport is estimated to be 16±5×10 6 m 3 s −1 for σ 1<32.1 at 13°S. South of the Angola Gyre, the transports show an apparent cyclonic circulation, developed mostly at the intermediate level. The water mass properties suggest that it is, at least partially, a re-circulation of the Benguela Current. Further south and for the subtropical gyre, we estimate 10±5×10 6 m 3 s −1 for the transport of the South Atlantic Current across 9°W and 28±4×10 6 m 3 s −1 for the transport of the Benguela Current at 10°E for σ 1<32.1.

Oceanic nutrient and oxygen transports and bounds on export production during the World Ocean Circulation Experiment

Large‐scale oceanic transports of nutrient and oxygen are estimated from selected hydrographic sections from the World Ocean Circulation Experiment spanning the world ocean. A so‐called geostrophic inverse box model is used to calculate the velocity field across the transoceanic sections. The circulation is required, a priori, to conserve mass, salt, top‐to‐bottom silicate, and subsurface heat and PO( = 170P + O2). The resulting estimate of the time mean circulation is used to compute horizontal and vertical nutrient transports and their residual sources and sinks associated with biogeochemical processes. Locally, the remineralization rate is, in general, below our uncertainty level, with magnitudes consistent with in situ measurements (0 ± 0.1mol N m−2 yr−1 to 0.7 ± 0.25 mol N m−2 yr−1 below about 100 m). Because of correlations between errors, the export production becomes significant when integrated globally, with 390 ± 240 kmol Si s−1 and 57 ± 40 kmol N s−1(420 ± 290 kmol C s−1) between 47°N and 30°S and at 2000 m. Examination of N* provides estimates for nitrate fixation and denitrification, either consistent (North Atlantic) or contradictory (Indian) with previous findings. Similarily, oxygen utilization rates and net air‐sea exchanges are provided. A net uptake of oxygen at high latitudes and an outgassing at low latitudes are found. Examination of the pseudotracer O* suggests an important role of dissolved organic matter in the oxygen budget. Oceanic oxygen and O* are found in global balance within uncertainties.

Zonal Fluxes in the Deep Water Layers of the Western South Atlantic Ocean*

Zonal transports of North Atlantic Deep Water (NADW) in the South Atlantic are determined. For this purpose the circulation of intermediate and deep water masses is established on the basis of hydrographic sections from the World Ocean Circulation Experiment (WOCE) and some pre-WOCE sections, using temperature, salinity, nutrients, and anthropogenic tracers. Multiple linear regression is applied to infer missing parameters in the bottle dataset. A linear box-inverse model is used for a set of closed boxes given by sections and continental boundaries. After performing a detailed analysis of water mass distribution, 11 layers are prescribed. Neutral density surfaces are selected as layer interfaces, thus improving the description of water mass distribution in the transition between the subtropical and subpolar latitudes. Constraints for the inverse model include integral meridional salt and phosphorus transports, overall salt and silica conservation, and transports from moored current meter observations. Inferred transport numbers for the mean meridional thermohaline overturning are given. Persistent zonal NADW transport bands are found in the western South Atlantic, in particular eastward flow of relatively new NADW between 20° and 25°S and westward flow of older NADW to the north of this latitude range. The axis of the eastward transport band corresponds to the core of property distributions in this region, suggesting Wüstian flow. Part of the eastward flow appears to cross the Mid-Atlantic Ridge at the Rio de Janeiro Fracture Zone. Results are compared qualitatively with deep float observations and results from general circulation models.

Impact of the Gravity Field and Steady‐State Ocean Circulation Explorer (GOCE) mission on ocean circulation estimates 2. Volume and heat fluxes across hydrographic sections of unequally spaced stations

Accurate geoids are expected to improve our knowledge of the dynamic sea surface height (SSH) as a mirror of the dynamic state of the oceans. The dedicated Gravity Field and Steady‐State Ocean Circulation Explorer (GOCE) mission will lead to a highly accurate geoid model with a resolution of degree and order 200. We examine the impact of this mission on the assessment of large‐scale oceanic mass and heat transports via its expected error characteristics. We do so applying a linear box inverse model and a nonlinear section inverse model to hydrographic data and to (synthetic) SSH data. The results are compared to those obtained when substituting the error estimates of the Gravity Recovery and Climate Experiment (GRACE) mission and the present‐day Earth Gravitational Model 1996. For the box inverse model we find an average reduction in transport uncertainties of about 9% for GRACE geoid error covariances and about 17% for GOCE over the “hydrography‐only” solution. In both GRACE and GOCE these average percentage improvements are significantly increased when model error is excluded. Summarizing our results and those of the companion parts of this study, we conclude that the GRACE mission reduces the marine geoid uncertainties such that altimetry may become useful for the study of the steady state ocean circulation. The GOCE mission will improve the accuracy of the circulation estimates significantly on the large scales and introduce higher accuracy on shorter wavelengths as well. Furthermore, it will enable us to study individual ocean currents.

Box inverse models, altimetry and the geoid: Problems with the omission error

[1] When one combines satellite altimetry and a geoid model to improve estimates of the ocean general circulation from hydrographic data with a box inverse model, there arises a problem of different resolution and representation of the data types involved. Here we show how this problem can lead to an artificial leakage of the error estimates of short-scale (high degree) spherical harmonic functions into long wavelength (low wave number) Fourier functions. A similar paradox effect can be seen in an idealized box inverse model constrained by additional sea-surface topography data of low, medium, and high resolution: When more information is added in the form of additional smaller scales, the error of a transport estimate eventually increases. Consequently, including the large geoid omission errors associated with smaller scales in a box inverse model of the Southern Ocean increases the posterior errors of transport estimates over those of a model that does not include the geoid omission error. We do not claim that including or excluding the geoid omission error is correct. Instead, we juxtapose two different ways of estimating the geoid errors to demonstrate the effect that the omission error might have on the long, supposedly well-known, scales. How (or if) to properly account for the geoid omission error must be the topic of further research. A proper treatment of the geoid model errors is demanded when one evaluates the errors of absolute sea-surface topography data.

On the export of Antarctic Bottom Water from the Weddell Sea

Abstract A survey of the current field over the South Scotia Ridge, obtained with a lowered Acoustic Doppler Current Profiler (LADCP), is presented. There is a pattern of northward (southward) flow on the western (eastern) side of each of four deep passages in the ridge, which is supported by tracer measurements. The net full-depth LADCP-referenced geostrophic transport over the ridge is 22±7 Sv (1 Sv=10 6 m 3 s −1 ) northward, with the jets on either side of the passages transporting 5–10 Sv in alternating directions. The corresponding Weddell Sea Deep Water (WSDW) transport over the ridge is 6.7±1.7 Sv. This is a factor of 4 larger than the only previous estimate in the literature, and suggests that a significant proportion of the Antarctic Bottom Water (AABW) invading the world ocean abyss escapes the Weddell Sea via the Scotia Sea. The net full-depth and WSDW transports over the ridge are modified to 7±6 and 4.7±0.7 Sv, respectively, by a box inverse model of the western Weddell Gyre. The model incorporates the WOCE A23 crossing of the central part of the gyre and a set of five constraints synthesizing our previous oceanographic knowledge of the region. It diagnoses that 9.7±3.7 Sv of AABW are formed in the Weddell Sea, and that comparable amounts are exported over the South Scotia Ridge (∼48%) and further east (∼52%) assuming that no AABW enters the Weddell Gyre from the Indian Ocean. The WSDW fraction with neutral density γ n >28.31 kg m −3 transported over the ridge upwells in the Scotia Sea at a rate of 6×10 −6 m s −1 , an order of magnitude larger than many basin-scale estimates of deep upwelling in the literature. In contrast, the Weddell Sea Bottom Water exported to the eastern Weddell Gyre entrains upward at a rate of 8×10 −7 m s −1 , more typical of other open-ocean regions. When their different ventilation histories are considered, the comparable transports and disparate upwelling rates of the AABW exported over the South Scotia Ridge and farther east may be crucial to our understanding of teleconnections between the Weddell Sea and the global ocean.

Circulation, Renewal, and Modification of Antarctic Mode and Intermediate Water*

Nine hydrographic sections are combined in an inverse box model of the Southern Ocean south of ;128S. The inverse model has two novel features: the inclusion of independent diapycnal flux unknowns for each property and the explicit inclusion of air‐sea fluxes (heat, freshwater, and momentum) and the water mass transformation they drive. Transformation of 34 3 106 m3 s21 of Antarctic Surface Water by air‐sea buoyancy fluxes, and cooling and freshening where Subantarctic Mode Water outcrops, renews cold, fresh Antarctic Intermediate Water of the southeast Pacific and southwest Atlantic. Relatively cold, fresh mode and intermediate water enter the subtropical gyres, are modified by air‐sea fluxes and interior mixing, and return poleward as warmer, saltier mode and intermediate water. While the zonally integrated meridional transport in these layers is small, the gross exchange is approximately 80 3 106 m3 s21. The air‐sea transformation of Antarctic surface water to intermediate water is compensated in the Southern Ocean by an interior diapycnal flux of 32 3 106 m3 s21 of intermediate water to upper deep water. The small property differences between slightly warmer, saltier intermediate water and cold, fresh Antarctic Surface Water results in a poleward transfer of heat and salt across the Polar Front zone. Mode and intermediate water are crucial participants in the North Atlantic Deep Water overturning and Indonesian Throughflow circulation cells. The North Atlantic Deep Water overturning is closed by cold, fresh intermediate water that is modified to warm, salty varieties by air‐sea fluxes and interior mixing in the Atlantic and southwest Indian Oceans. The Indonesian Throughflow is part of a circum-Australia circulation. In the Indian Ocean, surface water is converted to denser thermocline and mode water by air‐sea fluxes and interior mixing, excess mode water flows eastward south of Australia, and air‐sea fluxes convert mode water to thermocline water in the Pacific.

The Southern Ocean Limb of the Global Deep Overturning Circulation*

Nine hydrographic sections are combined in an inverse box model of the Southern Ocean south of ∼12°S. The inclusion of independent diapycnal flux unknowns for each property and air–sea (heat, freshwater, and momentum) fluxes make it possible to estimate the three-dimensional deep water circulation. The authors find a vigorous 50 × 106 m3 s−1 deep overturning circulation that is dominated by an equatorward flow of Lower Circumpolar Deep Water and Antarctic Bottom Water and poleward flow of upper deep water including Indian and Pacific Deep Water below 1500 dbar. In the subtropical Indian and Pacific Oceans the deep overturning cell is essentially isolated from the thermocline and intermediate waters of the subtropical gyre. The southward flowing upper deep water shoals south of the Antarctic Circumpolar Current, where air–sea fluxes convert outcropping upper deep water to Antarctic surface water and drive a net air–sea transformation of 34 × 106 m3 s−1 to lighter intermediate water. It is the outcropping of upper deep water and transformation by air–sea fluxes that connects the deep and intermediate circulation cells. The significant poleward transport of relatively light (i.e., above all topography at the latitude of Drake Passage) upper deep water, as required here to balance lateral and diapycnal divergence and air–sea exchange, provides observational evidence that advection by standing and transient eddies carries significant meridional transport in the Southern Ocean.

Estimates of Area-Averaged Diapycnal Fluxes from Basin-Scale Budgets

Abstract Estimates of area-averaged diapycnal fluxes for the southern oceans are derived from basin-scale budgets of mass, heat, and salt using a box inverse model. The diapycnal fluxes are found to be significant terms in the isopycnal budgets of mass, heat, and salt. Dense water entering the subtropical Indian and Pacific basins from the south is returned below the thermocline as less dense deep water. In the Southern Ocean deep water is converted to denser deep and bottom water. Water properties at intermediate depth are substantially modified by diapycnal fluxes of heat and salt, but the modification of intermediate water is not solely driven by interior mixing. The inferred fluxes help explain the changes in temperature–salinity curves observed across each basin, and they are consistent with our understanding of the overall three-dimensional circulation of the Southern Ocean. The fact that area-averaged diapycnal fluxes can be determined from basin-scale budgets using a suitably designed inverse mode...

Steady models of Arctic shelf‐basin exchange

Steady state exchanges between the Arctic shelves and the central basins are estimated using an inverse box model. Estimates based only on mass, heat, and salt conservation are highly uncertain and cannot determine how the shelf‐basin exchange is partitioned between dense overflows and surface layer exchange. Inclusion of δ18O conservation greatly reduces the uncertainty. With the δ18O constraints, dense water fluxes from the shelves of 0.2±0.1 Sv in the Canadian Basin and 1.4±0.3 Sv in the Eurasian Basin with a small (O(0.1 Sv)) surface layer transport from shelves to basins best satisfy the requirements of steady state. Up to 3.8 Sv of shelfbreak upwelling is consistent within error bars with steady state, but none is required. Water leaving the shelves is replaced primarily from the Bering Strait, the Barents Sea inflow, and runoff and not from the basin surface layer; proper representation of these external sources for the shelves is important for capturing the full range of plausible steady solutions. The steady state solutions identify the balances which, if upset, could lead to Arctic change.

Mass, heat, and salt transport in the southeastern Pacific: A Circumpolar Current inverse model

A box inverse model for the southeastern Pacific is used to combine hydrographic measurements with synoptic information from floats, acoustic Doppler surface velocity measurements, altimeter‐derived surface geostrophic velocities, and wind measurements. The southeastern Pacific is marked by substantial changes in eddy kinetic energy along the path of the Antarctic Circumpolar Current (ACC), associated with the Eltanin and Udintsev Fracture Zones in the East Pacific Rise. The results of the inverse model show that the heat and salt carried by the ACC are essentially unchanged as the current passes through regions of high eddy variability. Thus, within the error bars of the available measurements the ACC appears not to exchange significant water properties with its surroundings in the southeastern Pacific Ocean. (The system is geographically variable, and elsewhere in the Southern Ocean, meridional exchanges may be substantial.) Meridional flow through the north side of the inverse model domain is consistent with an overturning circulation that carries surface intermediate water northward and returns deep water southward into the ACC. Diapycnal fluxes identified in the inverse solution indicate that water mass transformation occurs predominantly in the outcropping upper layers of the ocean.

Do Box Inverse Models Work?

Abstract The performance of a box inverse model is tested using output from a near-eddy-resolving numerical model. Conservation equations are written in isopycnal layers for three properties: mass, heat, and salt anomaly. If the equations are free of error and the vertical exchange of properties between layers is negligible or known, the reference level velocity structure is quite accurately reproduced despite the underdetermined nature of the problem. If the interlayer fluxes of properties are not negligible and they are ignored, the solution for the reference level velocities is poor. If the interlayer fluxes of properties are included as additional unknowns in the inversion, they can be accurately estimated provided the column weights are chosen appropriately. Column weights that minimize the ratio of largest to smallest singular value (the “condition number”) result in the best solutions for interfacial fluxes, and generally also for lateral fluxes. This choice of column weights also makes the inversi...

Production of dense overflow waters feeding the North Atlantic across the Greenland-Scotland Ridge. Part 2: An inverse model

In Part 1 of this paper ( Deep-Sea Research I, 43, 769–806) an alternative large-scale circulation scheme for the Nordic Seas and the Arctic Ocean was developed, the motivation being to explain the formation and subsequent circulation of the waters that feed the dense overflows across the Greenland-Scotland Ridge into the North Atlantic. The new scheme emphasizes the water changes following the direct advective paths of the warm Atlantic inflow to dense overflows at shallow and intermediate depths. In this paper the new circulation scheme is evaluated quantitatively in terms of an inverse box model. For each water mass, conservation statements are written for mass, heat and salt, and an optimal solution is sought that deviates the least from the observed currents and air-sea fluxes, and yet conserves mass, heat and salt. It is found that the proposed circulation scheme is consistent with the conservation statements, and with the water masses, currents and air-sea fluxes in the region. Estimates of volume transport are in this way assigned to the various branches of the proposed circulation scheme. The inverse model is also used to test various hypotheses, and in particular, it is found that the major site of dense-water formation is the Norwegian Atlantic Current east of the Greenland and Iceland Seas. It is estimated that the intensity of this direct pathway exceeds that of the open-ocean gyres 6:1. Lastly, it is found that the dense water mass formation process within the Norwegian Atlantic Current does not induce a significant seasonal signal in the volume transport. This is consistent with observations of the flow field across the Greenland-Scotland Ridge, which so far have not shown indications that the dense overflows are associated with a seasonal signal.

The water masses and circulation at 10°N in the Pacific

The circulation and distribution of water masses near the southern boundary of the North Pacific Basin are described, based on a recent hydrographic survey made at 10°N. A circulation scheme is found, using both the tracer data and a box inverse model. To ensure the best possible realization of the mean state, repeat survey data are used in the boundary current. Historical data are used to check the representativeness of the transport across the onetime section outside of the boundary current. The upper 400 m are dominated by the North Pacific tropical Sverdrup cell, where the net interior flow is to the north, compensated by an equatorward low-latitude boundary flow: the Mindanao Current. The tropical cell is highly baroclinic in the sense that 18 Sv (1 Sv = 1 × 10 6 m 3 s −1) of tropical surface water flows northwards and is returned southward entirely within the shallow tropical thermocline: half in a broad interior flow and half in the southward boundary current. The strongly baroclinic structure of the tropical cell allows for the efficient transport of North Pacific subtropical water masses across the gyre boundary. The tropical cell is responsible for nearly all of the 0.7 ± 0.5 PW of heat transported northwards across the section. There is essentially zero freshwater divergence over the North Pacific north of 10°N. Critical to the salt balance of the North Pacific basin is a poleward flux of salty subthermocline water which ultimately derives from the South Pacific subtropical gyre. The South Pacific water found along 10°N has temperatures and salinities characteristic of the 13°C Thermostad found at the equator, suggestive of a link via the eastern Pacific. At depth, the analysis reveals a deep meridional overturning in the North Pacific consisting of the import across 10°N of 8 Sv of bottom waters of Antarctic origin and their subsequent conversion and export as North Pacific Deep Water. Dominating the mid-depth property distribution and circulation is a deep anticyclonic cell which horizontally cycles 8 Sv of deep water through the North Pacific basin. We find that there is little interaction between the North Pacific wind-driven circulation and the abyssal volume, which contrasts with the Atlantic Ocean, in which there is strong communication between the cold and warm water volumes.