Three interrelated climate phenomena are at the center of the Climate Variability and Predictability (CLIVAR) Atlantic research: tropical Atlantic variability (TAV), the North Atlantic Oscillation (NAO), and the Atlantic meridional overturning circulation (MOC). These phenomena produce a myriad of impacts on society and the environment on seasonal, interannual, and longer time scales through variability manifest as coherent fluctuations in ocean and land temperature, rainfall, and extreme events. Improved understanding of this variability is essential for assessing the likely range of future climate fluctuations and the extent to which they may be predictable, as well as understanding the potential impact of human-induced climate change. CLIVAR is addressing these issues through prioritized and integrated plans for short-term and sustained observations, basin-scale reanalysis, and modeling and theoretical investigations of the coupled Atlantic climate system and its links to remote regions. In this paper, a brief review of the state of understanding of Atlantic climate variability and achievements to date is provided. Considerable discussion is given to future challenges related to building and sustaining observing systems, developing synthesis strategies to support understanding and attribution of observed change, understanding sources of predictability, and developing prediction systems in order to meet the scientific objectives of the CLIVAR Atlantic program.

The tropical oceans have long been recognized as the most important region for large-scale ocean-atmosphere interactions, giving rise to coupled climate variations on several time scales. During the Tropical Ocean Global Atmosphere (TOGA) decade, the focus of much tropical ocean research was on understanding El Nino-related processes and on development of tropical ocean models capable of simulating and predicting El Nino. These studies led to an appreciation of the vital role the ocean plays in providing the memory for predicting El Nino and thus making seasonal climate prediction feasible. With the end of TOGA and the beginning of Climate Variability and Prediction (CLIVAR), the scope of climate variability and predictability studies has expanded from the tropical Pacific and ENSO-centric basis to the global domain. In this paper the progress that has been made in tropical ocean climate studies during the early years of CLIVAR is discussed. The discussion is divided geographically into three tropical ocean basins with an emphasis on the dynamical processes that are most relevant to the coupling between the atmosphere and oceans. For the tropical Pacific, the continuing effort to improve understanding of large- and small-scale dynamics for the purpose of extending the skill of ENSO prediction is assessed. This paper then goes beyond the time and space scales of El Nino and discusses recent research activities on the fundamental issue of the processes maintaining the tropical thermocline. This includes the study of subtropical cells (STCs) and ventilated thermocline processes, which are potentially important to the understanding of the low-frequency modulation of El Nino. For the tropical Atlantic, the dominant oceanic processes that interact with regional atmospheric feedbacks are examined as well as the remote influence from both the Pacific El Nino and extratropical climate fluctuations giving rise to multiple patterns of variability distinguished by season and location: The potential impact of Atlantic thermohaline circulation on tropical Atlantic variability (TAV) is also discussed. For the tropical Indian Ocean, local and remote mechanisms governing low-frequency sea surface temperature variations are examined. After reviewing the recent rapid progress in the understanding of coupled dynamics in the region, this study focuses on the active role of ocean dynamics in a seasonally locked east-west internal mode of variability, known as the Indian Ocean dipole (IOD). Influences of the IOD on climatic conditions in Asia, Australia, East Africa, and Europe are discussed. While the attempt throughout is to give a comprehensive overview of what is known about the role of the tropical oceans in climate, the fact of the matter is that much remains to be understood and explained. The complex nature of the tropical coupled phenomena and the interaction among them argue strongly for coordinated and sustained observations, as well as additional careful modeling investigations in order to further advance the current understanding of the role of tropical oceans in climate.

Repeated hydrographic observations between 1996 and 2001 of the deep water mass distribution on four sections in the western Labrador Sea and northwestern North Atlantic at about 56°N, 53°N, 48°N and 43°N show significant changes in the water mass characteristics. These changes are spreading southward mainly with the Deep Western Boundary Current (DWBC). Shallower convection forms a convective water mass known as upper Labrador Sea Water (ULSW). During periods of deep convection in the Labrador Sea, ULSW was described to be formed in the western boundary current region. In the post deep convection period 1996 to 2001 ULSW was formed in the western and central Labrador Sea and spreads mainly westward towards and along the western boundary. At 53°N ULSW moves southward as a part of the deep Labrador Current, also constituting the upper part of the DWBC. In the early 1990s the deep convection produced a large volume of deep Labrador Sea Water (LSW) which filled intermediate layers of the central region of the Labrador Sea. After these years the convection became weaker, with no apparent LSW renewal in 1996, partial mixing down to 1500m in 1997 and no notable LSW formation between 1998 and 2001. At the southwestern exit of the Labrador Sea at 53°N the deep LSW in 2001 was least in thickness and highest in salinity and temperature compared to the years since 1996. This reflects restratification which resulted in an increase in the density stratification between 1000 and 2000m in the central Labrador Sea as well as year-to-year transformation of the LSW core. LSW passes 43°N off the Grand Banks about 1 to 2 years after it was first seen at 56°N. At the 48°N and 43°N sections the northward flowing North Atlantic Current (NAC), farther offshore than the DWBC, complicates the property distributions. Saltier and warmer LSW recirculates northward with the NAC at 43°N. Between 1996 and 2001 the Gibbs Fracture Zone Water (GFZW) turned colder and fresher. The Denmark Strait Overflow Water (DSOW) showed two periods of cooling and freshening, separated by an abrupt (rapid) increase in temperature and salinity within a year. The arrival time of this increase at the different locations implies a DSOW spreading time that is no more than two years from 56°N to 43°N near the western boundary, or four years from the sill of the Denmark Strait to the Grand Banks.

Europe's relative warmth is maintained by the poleward surface branch of the Atlantic Ocean thermohaline circulation. There is paleoceanographic evidence for significant variability and even shifts between different modes of thermohaline circulation. Coupled ocean-atmosphere climate modelling allows first insight into the relative role of the various drivers of the Atlantic thermohaline circulation variability, i.e. the North Atlantic Oscillation, the tropical Atlantic variability, the ocean basin exchanges, small scale processes like high-latitude convection, overflows and mixing as well as effects of changes in the hydrological circle, the atmospheric CO2-content and solar radiation. The strong need for continous model improvement requires concerted efforts in ocean time series observations and relevant process studies. New instrumentation and methods, both for in situ measurements and remote satellite sensing are becoming available to help on the way forward towards as improved understanding of North Atlantic climate varibility.

This paper is an observational study of small-scale coherent eddies in the Labrador Sea, a region of dense water formation thought to be of considerable importance to the North Atlantic overturning circulation. Numerical studies of deep convection emphasize coherent eddies as a mechanism for the lateral transport of heat, yet their small size has hindered observational progress. A large part of this paper is therefore devoted to developing new methods for identifying and describing coherent eddies in two observational platforms, current meter moorings and satellite altimetry. Details of the current and water mass structure of individual eddy events, as they are swept past by an advecting flow, can then be extracted from the mooring data. A transition is seen during mid-1997, with long-lived boundary current eddies dominating the central Labrador Sea year-round after this time, and convectively formed eddies similar to those seen in deep convection modeling studies apparent prior to this time. The TOPEX / Poseidon altimeter covers the Labrador Sea with a loose "net" of observations, through which coherent eddies can seem to appear and disappear. By concentrating on locating and describing anomalous events in individual altimeter tracks, a portrait of the spatial and temporal variability of the underlying eddy field can be constructed. The altimeter results reveal an annual "pulsation" of energy and of coherent eddies originating during the late fall at a particular location in the boundary current, pinpointing the time and place of the boundary current-type eddy formation. The interannual variability seen at the mooring is reproduced, but the mooring site is found to be within a localized region of greatly enhanced eddy activity. Notably lacking in both the annual cycle and interannual variability is a clear relationship between the eddies or eddy energy and the intensity of wintertime cooling. These eddy observations, as well as hydrographic evidence, suggest an active role for boundary current dynamics in shaping the energetics and water mass properties of the interior region.

The bottom and deep circulation in the Somali Basin are investigated on the basis of hydrographic and direct velocity profiles from three shipboard surveys carried out during the southwest monsoon in 1995 and of velocity time series from the WOCE mooring array ICM7. The inflow of bottom water into the Somali Basin through the Amirante Passage drives a thermohaline circulation, which may be modulated by the monsoon wind forcing. Details of the abyssal circulation have been discussed controversially. Deep velocity records from the mooring array in the northern Somali Basin are dominated by fluctuations with periods of 30–50 days and amplitudes above 5cms−1. Despite this strong variability annual record averages indicate the existence of a deep western boundary current (DWBC) below 3000m at the base of the continental slope south of Socotra Island as part of a cyclonic bottom circulation. The southwestward DWBC transport off Socotra Island is estimated to 2±1.3Sv. The bottom and deep water exchange between the Somali and Arabian Basin north of 7°N is estimated from two cross-basin geostrophic velocity sections referenced by vertically averaged LADCP currents. For the bottom water, an eastward transport into the Arabian Basin of 1.4±0.5Sv and 2.1±0.6Sv was determined in June and August, respectively, while for the deep-water layer above 3500m eastward transports of 3.6±2.3Sv in June and 4.0±2.4Sv in August were obtained.

Sea-surface height data acquired by the TOPEX/POSEIDON satellite over the Arabian Sea from October 1992 to October 1998 are analyzed. Strong seasonal fluctuations are found between 6° and 10°N, which are mainly associated with westward propagating annual Rossby waves radiated from the western side of the Indian subcontinent and that are continuously forced by the action of the wind-stress curl over the central Arabian Sea. An analysis of hydrographic data acquired during August 1993 and during January 1998 at 8°N in the Arabian Sea reveals the existence of first- and second-mode annual Rossby waves. These waves, which can be traced as perturbations in the density fields, have wavelengths of 12×103 and 4.4×103km as well as phase velocities of 0.38 and 0.14m/s, respectively. The waves are associated with a time-dependent meridional overturning cell that sloshes water northward and southward. Between 58° and 68°E in the central Arabian Sea, we found a Rossby-wave induced transport in the upper 500m of about 10Sv southward in August 1993 and northward in January 1998. Below 2000m, there was still a northward transport of 3.2Sv in August 1993 and a southward transport of 4.8Sv in January 1998. A comparison of steric height differences between August 1993 and January 1998 calculated from the observed density fields as well as calculated from the reconstructed density fields using first- and second-mode annual Rossby waves agree quite well with the corresponding sea-surface height differences. Implications resulting from the reflection of annual Rossby waves, like fluctuations of the western boundary currents, are discussed.

The differences in the water mass distributions and transports in the Arabian Sea between the summer monsoon of August 1993 and the winter monsoon of January 1998 are investigated, based on two hydrographic sections along approximately 8°N. At the western end the sections were closed by a northward leg towards the African continent at about 55°E. In the central basin along 8°N the monsoon anomalies of the temperature and density below the surface-mixed layer were dominated by annual Rossby waves propagating westward across the Arabian Sea. In the northwestern part of the basin the annual Rossby waves have much smaller impact, and the density anomalies observed there were mostly associated with the Socotra Gyre. Salinity and oxygen differences along the section reflect local processes such as the spreading of water masses originating in the Bay of Bengal, northward transport of Indian Central Water, or slightly stronger southward spreading of Red Sea Water in August than in January. The anomalous wind conditions of 1997/98 influenced only the upper 50–100m with warmer surface waters in January 1998, and Bay of Bengal Water covered the surface layer of the section in the eastern Arabian Sea. Estimates of the overturning circulation of the Arabian Sea were carried out despite the fact that many uncertainties are involved. For both cruises a vertical overturning cell of about 4–6Sv was determined, with inflow below 2500m and outflow between about 300 and 2500m. In the upper 300–450m a seasonally reversing shallow meridional overturning cell appears to exist in which the Ekman transport is balanced by a geostrophic transport. The heat flux across 8°N is dominated by the Ekman transport, yielding about –0.6PW for August 1993, and 0.24PW for January 1998. These values are comparable to climatological and model derived heat flux estimates. Freshwater fluxes across 8°N also were computed, yielding northward freshwater fluxes of 0.07Sv in January 1998 and 0.43Sv in August 1993. From climatological salinities the stronger freshwater flux in August was found to be caused by the seasonal change of salinity storage in the Arabian Sea north of 8°N. The near-surface circulation follows complex pathways, with generally cyclonic-circulation in January 1998 affected at the eastern side by the Laccadive High, and anticyclonic circulation in August 1993.

The so-called equatorial stacked jets are analyzed with ship-board observations and moored time series from the Atlantic Ocean. The features are identified and isolated by comparing vertical wavenumber spectra at the equator with those a few degrees from the equator. Mode-filtering gives clear views of the jets in meridional sections, the typical extent being +/-1degrees in latitude. The vertical structure can be well described (explaining 82% of the variance) by N-1-stretched cosines, with a Gaussian amplitude tapering in the vertical. The stretched wavelengths are somewhat variable.Fitting jets of a fixed (stretched) wavelength to four moored sensors in the depth range 1300-1900 m, allows one to track the vertical phase of the jets with an rms error of 30degrees-45degrees. The resulting fit from a 20-month moored time series shows long periods of unchanging jet conditions and intermittent times of high variability. There is no significant vertical propagation on these timescales nor a seasonal reversal. Using a composite from many different experiments, interannual variability is visible, however.A possible mechanism for the stacked jets is inertial instability, resulting from background meridional shears at the equator. A condition is that the Ertel potential vorticity becomes zero somewhere, due to meridional asymmetries in the zonal flows. The ship-board observations show that this may be approximately fulfilled by the instantaneous zonal low-mode flows at various depths, resulting from an excess of zonal momentum south of the equator most of the time. Inertial instability should act to redistribute this zonal momentum, and our mooring data show indeed persistent northward momentum flux, but not at the depth levels expected. The momentum transport might suggest that the jets can also flux or mix other properties across the equator.

Observations from cruises in the Arabian Sea and data from satellites are interpreted using different realizations of a multi-level primitive equation model and an eddy-permitting reduced-gravity shallow water model of the Indian Ocean. The focus is on the interannual circulation variability of the Arabian Sea, and especially of the meridional location of the Great Whirl (GW). The results suggest that the variability in the western Arabian Sea is not only due to the interannual variability in the wind field, but that a substantial part is caused by the chaotic nature of the ocean dynamics. Decreasing the friction coefficient from 1000 to 500m2s−1 in a 19° numerical reduced-gravity model, the variance of the GW location increases dramatically, and the mean position moves southward by one degree. In the eddy-permitting experiments analyzed, both mechanisms appear to determine the GW location at the onset of the GW dynamics in late summer.

ocean acoustic tomography has evolved into a multipurpose remote-sensing measurement technique that has been employed in a wide variety of physical settings. In the context of long-term oceanic climate change, acoustic tomography provides integrals through the mesoscale and other high-wavenumber noise over long distances. In addition, tomographic measure-ments can be made without risk of calibration drift; therefore these measurements have the accuracy and precision required for large-scale ocean climate observation. The transbasin acoustic measurements offer a signal-to-noise capability for observing ocean climate variability that is difficult to attain by an ensemble of point measurements. On a regional scale, tomography has been employed for observing regions of active convection, for measuring changes in integrated heat content, for observing the mesoscale with high resolution, for measuring barotropic currents in a unique way, and for directly observing oceanic relative vorticity. The remote-sensing capability has proven effective for measurements under ice in the Arctic (in particular the recent well-documented temperature increase in the Atlantic layer) and in regions such as the Strait of Gibraltar, where conven-tional in-situ methods are problematic. As oceanographic science moves into an era of global-scale observations, the niches for these acoustic techniques appear to be (1) to exploit the unique remote-sensing capabilities for regional programs which are otherwise difficult to carry out,(2) to be a component of process-monitoring efforts in regions where integral heat content or transport data are desired, and (3) to …

The semienclosed western Mediterranean Sea has proven to be a useful location to evaluate surface heat flux estimates. In the past the directly measured average oceanic heat transport from the Atlantic into the Mediterranean Sea through the Strait of Gibraltar of similar to 5.2 +/- 1.3 W m(-2) has been compared to estimates of the average heat flux across the surface of the Mediterranean Sea. On long timescales both should closely balance each other. By using a monthly temperature climatology of the western Mediterranean Sea we offer the possibility to extend the comparison to the seasonal timescale. This gives additional information with which different surface heat flux data sets can be evaluated. The seasonal heat content changes of the western Mediterranean and the advective exchange of heat through the Straits of Gibraltar and Sicily are estimated on the basis of a new extensive hydrographic data set and of published values for the volume transports. To demonstrate the method, a limited number of surface heat flux data sets are compared with the oceanographically calculated counterpart. The comparison reveals that some heat fluxes do not only agree well for the long-term averages but also for the seasonal timescale, whereas others show larger deviations. The remaining rms discrepancies of +/-10.2 W m(-2) for the best heat flux data set are smaller than the uncertainty of the oceanographic estimate and of a reasonable magnitude compared to the uncertainty of the long-term average of similar to 5 W m(-2).

The Charlie Gibbs Fracture Zone (CGFZ), a passage of 3600 m sill depth through the Mid‐Atlantic Ridge near 52°N, is a known gateway for the passage of deep waters from the Northeast Atlantic into the western basin. During a shipboard survey of August 1997 deep current profiling yielded eastward deep flow through the passage while geostrophy calculated against an intermediate reference level resulted in westward relative deep transport. The reason was an unusual and deep‐reaching northward excursion of the North Atlantic Current (NAC). Inspection of historical data showed that such interference of the NAC with the CGFZ regime occured occasionally in the past. Relocation of surface circulation patterns by decadal ocean‐climate anomalies may thus be of significance also for the deep circulation.

The mean horizontal flow field of the tropical Atlantic Ocean is described between 20°N and 20°S from observations and literature results for three layers of the upper ocean, Tropical Surface Water, Central Water, and Antarctic Intermediate Water. Compared to the subtropical gyres the tropical circulation shows several zonal current and countercurrent bands of smaller meridional and vertical extent. The wind-driven Ekman layer in the upper tens of meters of the ocean masks at some places the flow structure of the Tropical Surface Water layer as is the case for the Angola Gyre in the eastern tropical South Atlantic. Although there are regions with a strong seasonal cycle of the Tropical Surface Water circulation, such as the North Equatorial Countercurrent, large regions of the tropics do not show a significant seasonal cycle. In the Central Water layer below, the eastward North and South Equatorial undercurrents appear imbedded in the westward-flowing South Equatorial Current. The Antarcic Intermediate Water layer contains several zonal current bands south of 3°N, but only weak flow exists north of 3°N. The sparse available data suggest that the Equatorial Intermediate Current as well as the Southern and Northern Intermediate Countercurrents extend zonally across the entire equatorial basin. Due to the convergence of northern and southern water masses, the western tropical Atlantic north of the equator is an important site for the mixture of water masses, but more work is needed to better understand the role of the various zonal under- and countercurrents in cross-equatorial water mass transfer.

A 12-month mooring record (May 1994–June 1995), together with accompanying PALACE float data, is used to describe an annual cycle of deep convection and restratification in the Labrador Sea. The mooring is located at 56.75°N, 52.5°W, near the former site of Ocean Weather Station Bravo, in water of 3500 m depth. This is a pilot experiment for climate monitoring, and also for studies of deep-convection dynamics. Mooring measurements include temperature (T), salinity (S), horizontal and vertical velocity, and acoustic measurement of surface winds. The floats made weekly temperature–salinity profiles between their drift level (near 1500 m) and the surface. With moderately strong cooling to the atmosphere (300 W m−2 averaged from November to March), wintertime convection penetrated from the surface to about 1750 m, overcoming the stabilizing effect of upper-ocean low-salinity water. The water column restratifies rapidly after brief vertical homogenization (in potential density, salinity, and potential temperature). Both the rapid restratification and the energetic high-frequency variations of T and S observed at the mooring are suggestive of a convection depth that varies greatly with location. Lateral variations in T and S exist down to very small scales, and these remnants of convection decay (with e-folding time 170 day) after convection ceases. Lateral variability at the scale of 100 km is verified by PALACE profiles. The Eulerian mooring effectively samples the convection in a mesoscale region of ocean as eddies sweep past it; the Lagrangian PALACE floats are complementary in sampling the geography of deep convection more widely. This laterally variable convection leaves the water column with significant vertical gradients most of the year. Convection followed by lateral mixing gives vertical salinity profiles the (misleading) appearance that a one-dimensional diffusive process is fluxing freshwater downward. During spring, summer, and fall the salinity, temperature, and buoyancy rise steadily with time throughout most of the water column. This is likely the result of mixing with the encircling boundary currents, compensating for the escape of Labrador Sea Water from the region. Low-salinity water mixes into the gyre only near the surface. The water-column heat balance is in satisfactory agreement with meteorological assimilation models. Directly observed subsurface calorimetry may be the more reliable indication of the annual-mean air–sea heat flux. Acoustic instrumentation on the mooring gave a surprisingly good time series of the vector surface wind. The three-dimensional velocity field consists of convective plumes of width 200 to 1000 m, vertical velocities of 2 to 8 cm s−1, and Rossby numbers of order unity, embedded in stronger (20 cm s−1) lateral currents associated with mesoscale eddies. Horizontal currents with timescales of several days to several months are strongly barotropic. They are suddenly energized as convection reaches great depth in early March, and develop toward a barotropic state, as also seen in models of convectively driven geostrophic turbulence in a weakly stratified, high-latitude ocean. Currents decay through the summer and autumn, apart from some persistent isolated eddies. These coherent, isolated, cold anticyclones carry cores of pure convected water long after the end of winter. Boundary currents nearby interact with the Labrador Sea gyre and provide an additional source of eddies in the interior Labrador Sea. An earlier study of the pulsation of the boundary currents is supported by observations of sudden ejection of floats from the central gyre into the boundary currents (and sudden ingestion of boundary current floats into the gyre interior), in what may be a mechanism for exchange between Labrador Sea Water and the World Ocean.

In the Gulf of Lions, observations of deep convection have been sporadically carried out over the past three decades, showing significant interannual variability of convection activity. As long time series of meteorological observations of the region are available from coastal stations, heat flux time series for the Gulf of Lions for the individual winters from 1969 to 1994 are derived by calibrating these observations against direct measurements obtained over the convection site. These heat fluxes are also compared against heat fluxes obtained by the French PERIDOT weather model for the winter of 1991/92. A Kraus–Turner one-dimensional mixed layer model is initialized by climatological mean temperature and salinity profiles and then driven by the heat flux time series of the individual years. Resulting convection depths are in satisfactory agreement with existing observational evidence, showing the dominance of interannual variability of local forcing on convection variability. The interannual variability of convection depth causes interannual variations in deep-water properties, and these are also compared with the hydrographic database.

In the autumn of 1996 the field component of an experiment designed to observe water mass transformation began in the Labrador Sea. Intense observations of ocean convection were taken in the following two winters. The purpose of the experiment was, by a combination of meteorological and oceanographic field observations, laboratory studies, theory, and modeling, to improve understanding of the convective process in the ocean and its representation in models. The dataset that has been gathered far exceeds previous efforts to observe the convective process anywhere in the ocean, both in its scope and range of techniques deployed. Combined with a comprehensive set of meteorological and air-sea flux measurements, it is giving unprecedented insights into the dynamics and thermodynamics of a closely coupled, semienclosed system known to have direct influence on the processes that control global climate.