Bathythermograph Data Research Articles

Simulation of heat storages and associated heat budgets in the Pacific Ocean: 1. El Niño‐Southern Oscillation timescale

We use a primitive equation isopycnal model of the Pacific Ocean to simulate and diagnose the anomalous heat balance on El Niño‐Southern Oscillation, ENSO, timescales associated with heat storage changes observed in the expendable bathythermograph (XBT) data set. We focus on the analysis of the total (diabatic plus adiabatic) and diabatic anomalous heat balances in six areas of the tropical and subtropical North Pacific Ocean in the upper 400 m. The diabatic (i.e., from the model conservation temperature equation) and adiabatic (i.e., from the model mass conservation equation) anomalous heat balances add up to the total anomalous heat balance. We computed the adiabatic/diabatic ratios to infer the relative importance of both contributions in different areas and found that they are smaller than 2.0 in only two regions (western equatorial and central North Pacific). The larger ratios (>2) were found along the corridor where adiabatic anomalies propagate westward in the form of Rossby waves and at the eastern equatorial Pacific. For those areas where the adiabatic/diabatic ratio is higher than about 2 the total anomalous heat balance is dominantly between the temporal change of heat and the three‐dimensional divergence of the heat flux. At the central North Pacific area the total anomalous heat balance is between the temporal changes in anomalous heat, the surface heat flux and the vertical advection of heat. Different ENSO events are not always controlled by the same physical processes in the different areas. In many cases these differences are associated with the relative importance of adiabatic to diabatic processes. For instance, the western equatorial Pacific is controlled in general by diabatic processes, while the eastern equatorial Pacific is dominated by adiabatic physics most of the time.

Forced Rossby waves in the southern tropical Indian Ocean

Seasonal and interannual variation of the upper southern tropical Indian Ocean (STIO) is described by harmonic and empirical orthogonal function (EOF) analysis of the depth of the 20°C isotherm (D20) derived from expendable bathythermograph (XBT) data and from an ocean general circulation model (OGCM). The harmonic analysis shows a band of large annual amplitude between 8° and 20°S extending across the STIO with a steady westward propagation in both the XBT and model data. The generation of the annual wave is discussed in terms of Ekman pumping and the westward propagation of long, nondispersive, baroclinic Rossby waves. The Ekman pumping on a large scale over the open ocean strongly modifies waves radiating from the eastern boundary and generates much of the structure in the amplitude and the phase of the annual signal in the STIO. The EOF analysis shows strong interannual variation of the anomaly of D20 on two XBT lines covering the western and eastern sides of the basin. Large interannual anomalies are observed in a band between 6° and 14°S on the western side, with deep D20 in 1988, 1991, and 1994 and shallow D20 in 1990 and 1992. On the eastern side, large interannual anomalies are observed close to the coast of Java, and sign is opposite to that on the western side, suggesting an oscillation in the zonal tilt of the thermocline across the STIO. Those interannual variations of D20 are also well simulated in the model. The generation of the interannual anomalies in the model are also due mainly to the wind stress curl integrated along the Rossby wave trajectories in the STIO.

The mean and annual cycle of upper layer temperature fields in relation to Sverdrup dynamics within the gyres of the Atlantic Ocean

Using 28 years of expendable bathythermograph data (1967–1994), we describe the mean and annual cycle of the upper ocean temperature fields in the Atlantic from 30°S to 50°N in the context of the basin‐scale wind‐driven gyres (Sverdrup stream function field), which provide a framework for describing the oceanographic measurements. We examine the circulation field implied by the temperature distributions, which are used as a proxy for the field of mass. Similarities between the temperature and stream function fields increase with depth. In the lower to subthermocline depths of the tropical and equatorial gyres the zonal currents form a closed circulation. A southeastward boundary current is suggested near and below 150 m that provides closure for the tropical gyre, and the equatorial gyre axis is southward of that suggested by the stream function field. Higher in the water column, the North Equatorial Countercurrent (NECC) may be a surface manifestation of the North Equatorial Undercurrent (NEUC), where the latter can be interpreted as the southern limb of the tropical gyre. Because there are large vertical shears in the tropics, the equatorial gyre is not clearly indicated in the vertically integrated temperature field but appears below about 200 m. Here, the South Equatorial Undercurrent (SEUC) can be interpreted as the eastward flowing northern limb of the equatorial gyre and is opposite in direction to the westward flowing South Equatorial Current above. Both the NEUC and SEUC are analagous to currents in the Pacific that are governed by non‐Sverdrup dynamics. Despite the shortcomings of the data, the mean annual cycle appears to be relatively stable, and we have discounted the possibility that in regions where it represents a significant percentage of the total variance, it is changing slowly over the 28 years of record. The wind‐forcing fields, which undergo large meridional movements (5°–6° of latitude) during their annual cycle, with some exceptions, have essentially no counterpart in gyre movements between their seasonal extremes. Most of the variability associated with the annual cycle is confined to the upper 300 m. Greatest variability, where ranges exceed 6°C, occurs in the northwestern Atlantic in late winter and early spring. During this time of year south of the Gulf Stream and below about 100 m, water temperatures exhibit a systematic phase lag with depth. The next largest area of variability, where ranges can also exceed 6°C, resides in the tropical western basin between the equator and 10°N just below 100 m. In the eastern basin, ranges decrease and shoal. Additionally, the phase fields are consistent with the intensification and relaxation of the tropical ridge‐trough system where the NECC disappears in March in the west, but the NECC/NEUC complex is strongest in September.

The mean and variability of ocean circulation past northern New Zealand: Determining the representativeness of hydrographic climatologies

Eastward flow in the Tasman Sea, from the separated East Australia Current, reattaches to the shelf break near North Cape, New Zealand, and then continues alongshore to the southeast as the East Auckland Current. A series of three permanent warm core eddies occurs along the offshore side. The mean transport of the East Auckland Current is about 9 Sv, with an additional 10 Sv or more of circulation in the eddies. An extensive hydrographic data set, archived broad scale expendable bathythermograph (XBT) data, two repeating high‐resolution XBT transects, neutrally buoyant float trajectories, and TOPEX altimetric data are used to estimate the temperature and absolute flow fields and to characterize variability. The aim is to examine the usefulness of time series information and absolute velocity measurements in the interpretation of hydrographic snapshots and climatologies, as well as to describe a region having intrinsic oceanographic interest and complexity. Issues of representativeness of the hydrographic data, of the magnitude and scales of the underlying variability, of the existence of permanent fine‐scale features, and of the appropriateness of deep reference levels are addressed directly. The relatively well sampled hydrographic climatology is shown to contain the equivalent of as many as 10 independent realizations. Temperature errors, relative to the true mean, are typically a few tenths of a degree. Significant seasonal bias is identified in the surface layer, and interannual bias is seen in the position of an eddy near North Cape. The dynamic height field at 1000 dbar relative to 2000 dbar is similar to estimates based on float trajectories and the assumption of geostrophic dynamics. This study underlines the value of time series data in the interpretation of a hydrographic climatology, in quantifying the errors in the estimated mean field as well as determining the magnitude and nature of variability. It also highlights the fact that the mean circulation of the oceans contains significant mesoscale structure, unnoticed in coarsely smoothed climatologies.

Seasonal cycle of the East Australian Current

The seasonal circulation associated with the East Australian Current is examined using a set of steric heights derived from the historical hydrology and expendable bathythermograph data collected in the region. The data are separated into a network of regional bins allowing for known oceanographic and topographic features and a two‐harmonic best fit to the seasonal cycle is obtained in each bin. Maps of the annual‐frequency component of the surface and depth‐integrated steric heights (h and P) show the development and progression of the EAC flow regime through a complete seasonal cycle. The EAC has a strong seasonal cycle from 25°S to 45°S, with strongest southward flow in austral summer. The seasonal cycle in surface flow over the continental shelf is documented by two independent methods, geostrophically, using cross shelf sea level gradients derived from coastal tide gauge data and steric heights at the continental shelf edge, and directly from merchant ship observations. The two estimates are in good agreement. The seasonal cycle in the EAC is more pronounced than in other midlatitude western boundary currents for which data are available. At 28°S, the strength of the total Tasman Sea transport (southward flow) varies between a minimum transport of 7 Sv in winter (July) to a maximum of 16 Sv in summer. The semiannual frequency components of h and P is important near 30°S near the EAC outflow, but not elsewhere. The seasonal cycle of the EAC is not due to strong seasonal variations in Tasman Sea wind stress curl east of the region of interest. Seasonally reversing zonal flows occur offshore north of 25°S, which are apparently locally forced by reversing wind stress curls; but if these flows were fed from the south by the EAC current system, the EAC would have to be weaker in summer, not stronger. The Leeuwin Current Extension along Australia's west and south coasts may pass up the east coast of Australia, providing an important contribution to the enhanced southward flow of the EAC in summer. The vigorous anticyclonic eddies of the EAC also show a marked seasonal cycle, and this is probably an important part of the mechanism for the strong seasonal cycle of the EAC south of 25°S. The location of the strongest anticyclonic eddy in the EAC moves steadily southward throughout the summer season, and the phase of the coastal EAC appears also to move southward, contrary to the expectations of linear theory and to the hypothesis that the Leeuwin Current Extension is the major cause of the seasonal cycle.

Response of global upper ocean temperature to changing solar irradiance

By focusing on time sequences of basin‐average and global‐average upper ocean temperature (i.e., from 40°S to 60°N) we find temperatures responding to changing solar irradiance in three separate frequency bands with periods of >100 years, 18–25 years, and 9–13 years. Moreover, we find them in two different data sets, that is, surface marine weather observations from 1990 to 1991 and bathythermograph (BT) upper ocean temperature profiles from 1955 to 1994. Band‐passing basin‐average temperature records find each frequency component in phase across the Indian, Pacific, and Atlantic Oceans, yielding global‐average records with maximum amplitudes of 0.04°±0.01°K and 0.07°±0.01°K on decadal and interdecadal scales, respectively. These achieve maximum correlation with solar irradiance records (i.e., with maximum amplitude 0.5 W m−2 at the top of the atmosphere) at phase lags ranging from 30° to 50°. From the BT data set, solar signals in global‐average temperature penetrate to 80–160 m, confined to the upper layer above the main pycnocline. Operating a global‐average heat budget for the upper ocean yields sea surface temperature responses of 0.01°–0.03°K and 0.02°–0.05°K on decadal and interdecadal scales, respectively, from the 0.1 W m−2 penetration of solar irradiance to the sea surface. Since this is of the same order as that observed (i.e., 0.04°–0.07°K), we can infer that anomalous heat from changing solar irradiance is stored in the upper layer of the ocean.

Can the Topex/Poseidon altimetry data be used to estimate air‐sea heat flux in the North Atlantic?

We propose a novel approach to directly invert large‐scale anomalous annual net air‐sea heat flux from the Topex/Poseidon altimetry data in the North Atlantic. The major advantage of this new approach over the conventional bulk formula approach is that it does not need those climate parameters used in the conventional bulk formula calculation in estimating the large‐scale anomalous annual net air‐sea heat flux. Comparison with expendable bathythermograph data demonstrates that it is a feasible approach.

Thermostad Distribution in the North Pacific Subtropical Gyre: The Central Mode Water and the Subtropical Mode Water

The upper thermal structure of the midlatitude North Pacific is studied with the use of all the bathythermograph data compiled in the global ocean temperature and salinity profiles released by the National Oceanographic Data Center. Climatological temperature data are prepared for each 2.5° × 5° (latitude × longitude) rectangle. The upper layer of the subtropical gyre is characterized by two types of thermostads or mode waters: one in the western basin known as North Pacific subtropical mode water (NPSTMW) and the other in the central basin to be named North Pacific central mode water (NPCMW). The NPCMW thermostad lies centered around 160°W between the Kuroshio Extension and the Kuroshio bifurcation front. Its local core temperature ranges from 10° to 13°C with a somewhat zonally elongated pattern, in contrast to the more uniform core temperature at 16°–17°C of the NPSTMW thermostad lying centered at 150°–160°E south of the Kuroshio Extension. The climatological map of the wintertime mixed layer thickness together with sea surface temperature distribution indicates that the NPCMW formation area is likely situated south of the Kuroshio bifurcation front and at 175°E–160°W extending as far south as about 37°N around 170°W. The climatological geostrophic flow field suggests that the NPCMW thermostad is spread from its formation area by the North Pacific Current flowing primarily eastward with relatively small southward velocity, which is contrasted with the NPSTMW thermostad spread within the anticyclonic Kuroshio recirculation region and explains the rather zonally elongated temperature distribution of NPCMW. The NPCMW formation can be fit into the recent estimation of the subduction rate of the main thermocline, coinciding with the region of the moderate subduction rate dominated by the lateral induction term.

North Pacific basin heat content variability determined from long-range acoustic transmissions

ATOC signals transmitted from the Pioneer Seamount acoustic source have been received on hydrophone arrays located throughout the North Pacific Basin since the beginning of 1996. Propagation times can be used to infer average temperature (or heat content) variability because the speed of sound changes with temperature. The time series of resolved-ray travel times show both an annual cycle with amplitudes up to 0.5 s, and other higher frequency fluctuations of comparable magnitude caused by natural oceanic variability. The range-averaged temperature is derived from the low-pass filtered (<1 cpd) resolved-ray travel times by constructing a model for the structure and statistics of oceanic variability, fitting the model to the data using weighted least-squares techniques, and then calculating the range- and depth-averaged quantity including its uncertainty. Because many of the resolved rays reflect off the sea surface, the mixed layer in particular, and the upper ocean in general, must be correctly modeled. The variability of acoustically determined heat content is compared to that determined from the Levitus94 atlas, historical expendable bathythermograph (XBT) data; other concurrent data types are not yet available for direct comparison. [Work supported by the Strategic Environmental Research and Development Program through ARPA.]

Mean annual variation of transport of major currents in the tropical Pacific Ocean

Expendable bathythermograph (XBT) data and the climatological temperature/salinity relationship were used to calculate the mean annual cycle of dynamic height and geostrophic transport of major currents relative to 400 db along five shipping tracks covering a large part of the tropical Pacific Ocean. The data were selected in bands centered on the most frequently repeated XBT tracklines for the period 1967–1988. Long-term bimonthly mean temperature was calculated in 1° latitude bins along the tracks. The transport function (vertically integrated dynamic height) was then calculated using the mean temperature/salinity relationship. The mean annual cycle of transport of the North Equatorial Current (NEC), the North Equatorial Countercurrent (NECC) and the South Equatorial Current (SEC) (south of 2.5°S) were determined between the ridges and troughs of the transport function. The stochastic errors in bimonthly mean transports were 1–2 Sverdrups on the most sampled tracks. Mean transports of the NEC and NECC increase regularly with longitude from east to west. The NECC has a large annual cycle with a transport-maximum during northern fall and winter. Seasonal variations of the NEC are small. Seasonal variations of the SEC are slightly larger, and they have considerably different phase from track to track. The variation of thermal structure associated with the currents is described. The results of this study are compared in detail to the results of earlier studies of the transports. The differences between the studies are larger than the expected stochastic errors in the mean transports due to differences in the definition of boundaries of the currents and to differences in the procedure for calculating the mean annual variation. The results of all the studies are summarised to facilitate future comparisons to ocean general circulation models and other applications.

Interannual variability of the Kuroshio transport in response to the wind stress field over the North Pacific: Its relation to the path variation south of Japan

We have investigated the interannual variability in the Kuroshio transport, focusing on the baroclinic response of the upper ocean to the long‐term variation of the wind stress field over the North Pacific from 1961 to 1987. First, temporal change of the Kuroshio transport at the western boundary from 17° to 31°N has been estimated on the assumptions that the Kuroshio is a baroclinic flow confined in the upper layer and that the effect of interior wind stress curl on the volume transport at the western boundary is transferred by nondispersive baroclinic Rossby wave. Peak values in the estimated transport appear in its beginning region (17°–21°N) when or just before the Kuroshio settles into a meandering state south of Japan in 1975, 1981, and 1986. These peaks are due to negative wind stress curl anomalies which are induced by the intensified trade winds in the central and eastern tropical North Pacific on and just after an El Niño event and in the western tropical North Pacific months to years before the next El Niño event. Such peaks are not found in the barotropic Sverdrup transport. In downstream regions, i.e., in the East China Sea and south of Japan (north of 23°N), on the contrary, the estimated transports (through baroclinic and barotropic responses) do not have such good correlation to the path variation. Temporal change in thermocline depths estimated from bathythermograph data clearly shows the westward propagations of positive anomalies by nondispersive baroclinic Rossby waves at latitudes of the beginning of the Kuroshio (south of 21°N), which is in phase with peaks of the Kuroshio baroclinic transport at the western boundary. On the other hand, there are only the local propagations of the anomalies in the downstream regions, as previous studies reported. However, it is emphasized that temporal change of the geostrophic transport in the East China Sea (∼28°N) is coincident with that of the baroclinic transport at the beginning of the Kuroshio, not with that at the same latitude. A numerical experiment with a nonlinear reduced‐gravity model confirms that temporal changes of the transport in the East China Sea and south of Japan are due to the advective (nonlinear) effect of the Kuroshio itself, while the linear baroclinic response to the wind stress field is dominant in its beginning region.

Impact of altimeter, thermistor, and expendable bathythermograph data on retrospective analyses of the tropical Pacific Ocean

This study explores the relative impact of three major components of the tropical Pacific Ocean observing system on data assimilation analyses. The extensive observing components that are available during the 2‐year period from October 1992 through September 1994 include sea level derived from TOPEX/POSEIDON altimetry, thermistor data from the Tropical Ocean/Global Atmosphere Tropical Atmosphere‐Ocean (TOGA TAO) mooring array, and expendable bathythermographs (XBTs) from the Volunteer Observing Ship program. In the first part of this study, methods are introduced to assimilate these data into a primitive equation model using optimal interpolation. The resulting velocity and thermal analyses are then compared with a numerical simulation that has no data assimilation. The most innovative aspect of the assimilation procedure is the development of a statistical model that uses satellite altimetry to update the subsurface thermal field. To determine the impact of individual components of the observing system, three additional experiments are conducted. In each experiment, one component is withheld from the assimilation. The resulting analysis is then compared with the missing data set. Our results show (1) To resolve major features of the seasonal cycle, it is necessary to have either the altimeter or the XBT data. In the absence of XBTs, the analysis appears to develop a significant temporal drift in its thermal structure. (2) Intraseasonal variability, such as tropical instability waves, is best resolved using altimetry because of that data set's superior spatial resolution. (3) Thermistor data from the TOGA TAO mooring array are helpful, but not as crucial, in resolving the seasonal cycle within ±8° latitude as the XBTs. The mooring array is too coarsely spaced to allow this analysis system to resolve tropical instability waves.

Long‐term temperature and circulation changes off eastern Australia

Long‐term changes in the thermocline circulation have been observed in the southwest Pacific Ocean from a combined data set of hydrographic and expendable bathythermograph data. The data were divided up into a network of 5° × 5° bins (with some exceptions to allow for coastal features) and mean values of temperature at several depths were obtained for the four pentads between 1970 and 1989. Differences in temperature between pairs of pentads show features which are coherent over scales as large as the Coral Sea. Since there is a high correlation between the temperature at 450 dbar (T450) with steric height in the Tasman and Coral Seas, we have focused our study on the behavior of T450. The 1985–1989 pentad appears to be cooler at 450 m depth than the preceding 15 years over most of the region of study (by more than 1°C in some bins). Applying an empirical orthogonal function analysis to the data shows that the first two modes explain 47% and 25% of the variance, respectively. The first mode consists of a warming from 1970 to 1976, followed by a steady cooling period to the latter part of the 1980's (below the starting temperature); its spatial structure shows that it is centered just to the west of New Caledonia and has a Coral Sea‐wide scale, reminiscent of a weakening of the Sverdrup wind‐driven gyre. The second mode has a more complex spatial character, which is most pronounced in the EAC region. It is qualitatively similar to that found in a recent study of the effect of blocking the Indonesian Throughflow in a global ocean model. It shows a warming from 1970, which plateaus from 1976 to 1983 and then decreases slightly. There are some statistically significant changes in the T‐S relations for the region, but the associated steric height variations are at the noise level. Futhermore, the long‐term vertical profile of temperature change in the top 450 m is shown to be similar to that associated with the energetic changes due to eddy movements. Using the relationship of steric height with T450, the data suggest that there were quite large changes in the circulation over this period. We estimate that the zonal geostrophic inflow to and outflow from the Coral and Tasman Seas at 162°E weakened by 15 and 10 Sv respectively, in the period between the 1975–1979 and 1985–1989 pentads. This is large compared to a similarly estimated long‐term mean inflow of 27 Sv. However, little evidence of broad‐scale, in‐phase El Niño‐Southern Oscillation‐related changes are found south of about 10°S.

An application of a new depth correction formula to archived XBT data

Historical expendable bathythermograph (XBT) data collected in the Tasma Sea off Eastern Australia (equator-45°S; 140–180°E), once corrected for a systematic depth error, are shown to be not significantly different from a hydrographic data set obtained over the same period. Prior to correction the XBTs exhibit significant differences (of order 0.3–0.4°C from the mean) from 200–700 m depth. This result appears to confirm that the new depth corrections remove the errors in the XBT depth measurements to within the quoted absolute accuracy of the XBTs.

Sea surface topography fields of the tropical Pacific from data assimilation

Time series of maps of monthly tropical Pacific dynamic topography anomalies from 1979 through 1985 were constructed by means of assimilation of tide gauge and expendable bathythermograph (XBT) data into a linear model driven by observed winds. Estimates of error statistics were calculated and compared to actual differences between hindcasts and observations. Four experiments were performed as follows: one with no assimilation, one with assimiation of sea level anomaly data from eight selected island tide gauge stations, one with assimilation of dynamic height anomalies derived from XBT data, and one with both XBT and tide gauge data assimilated. Data from seven additional tide gauge stations were withheld from the assimilation process and used for verification in all four experiments. Statistical objective maps based on data alone were also constructed for comparison purposes. The dynamic response of the model without assimilation was, in general, weaker than the observed response. Assimilation resulted in enhanced signal amplitude in all three assimilation experiments. RMS amplitudes of statistical objective maps were only strong near observing points. In large data‐void regions these maps show amplitudes even weaker than the wind‐driven model without assimilation. With few exceptions the error estimates generated by the Kaiman filter appeared quite reasonable. Since the error processes cannot be assumed to be white or stationary, we could find no straightforward way to test the formal statistical hypothesis that the time series of differences between the filter output and the actual observations were drawn from a population with statistics given by the Kaiman filter estimates. The autocovariance of the innovation sequence, i.e., the sequence of differences between forecasts before assimilation and observations, has long been used as an indicator of how close a filter is to optimality. We found that the best filter we could devise was still short of the goal of producing a white innovation sequence. In this and earlier studies, little sensitivity has been found to the parameters under our direct control. Extensive changes in the assumed error statistics make only marginal differences. The same is true for long time and space scale behavior of different models with richer physics and finer resolution. Better data assimilation results will probably require relaxation of the assumptions of stationarity and serial independence of the errors. Formulation of such detailed noise models will require longer time series, with the attendant problems of matching very different data sets.

Seasonal heat storage in the North Pacific: 1976–1989

Volunteer observing ship expendable bathythermograph data obtained during the period 1976–1989, from 30°N to 40°N in the North Pacific, were used to compute the rate of change of heat storage to a fixed upper ocean temperature surface. The variability of the storage rate in the seasonal timescale, computed on a 5° latitude by 5° longitude spatial scale, is compared to the net surface heat flux computed from available surface marine weather data to the same spatial and temporal resolution. Averaged across the entire basin, the difference between the average monthly heat storage rate and the average monthly heat flux is 3.76 Wm−1. The average basin‐wide absolute heat storage rate agrees to within 7.5% of the average absolute heat flux for the whole basin. An empirical orthogonal decomposition of the spatial patterns of the difference between the heat storage rate and the net heat flux reveals no obvious trends in the heat flux computation or possible physical processes responsible for the difference. Instead, the eddies shed by the warm boundary current, the Kuroshio, is probably responsible for the major part of the difference patterns in locations where the difference values are maximum. The most important results of this study are that the heat storage rate computed to a fixed isotherm matches the net heat flux extremely well at the chosen locations and across the whole basin; and the heat storage rate computation is sensitive to the isotherm choice and to the space scale involved.

Upper layer temperature structure of the western tropical Atlantic

Mean monthly topographies of the 20°C and 10°C isothermal surfaces are used to describe the vertical displacements of the upper and lower thermocline in the western tropical Atlantic. The isotherm topographies are generated from expendable bathythermograph data collected between 1966 and 1993. The topographies confirm, and extend closer to the coast, earlier findings that demonstrate large spatial and temporal variability in the region. For example, the ridge and trough systems observed previously in the interior are shown, and their extension to the western boundary is described. In particular, it is shown that the ridge associated with the North Equatorial Countercurrent (NECC) extends from the interior northwestward along the western boundary, reaching farther north along the boundary in the upper thermocline than in the lower thermocline. South of the equator the northwestern corner of the countercurrent trough is apparent on the lower surface but not on the upper. The annual and semiannual harmonics of the vertical isotherm displacements account on the average for about 60% of total variance on both surfaces. The horizontal structure of the first harmonic amplitude is similar for both surfaces, showing maximum amplitude along the axis of the NECC ridge. Minimum amplitudes are observed to the north along the axis of the countercurrent trough. These distributions are similar to the pattern of the first‐harmonic amplitude of the wind stress curl, supporting earlier studies of curl forcing of near‐surface current features.

Ocean dynamics in the region between Australia and Indonesia and its influence on the variation of sea surface temperature in a global general circulation model

Variation of sea surface temperature (SST) in the region between Australia and Indonesia is of crucial importance to Australian winter rainfall. To gain insight into how the SST is maintained and changed, the present study provides a comprehensive description of ocean dynamics in the region, using results from a seasonal ocean general circulation model run by Semtner and Chervin [1988,1992]. We also investigate the heat budget of the surface layer (0–50 m). Indonesian topography is heavily simplified in this model, but some simple results which probably do not depend strongly on topographic detail are obtained. The currents in the model are verified by comparison to observed annual mean and seasonal thermal structures and circulations determined from the expendable bathythermograph data collected in the region since 1983. We found a distinctive difference in the model circulations and heat budgets in two regions: one between the coast of Indonesia and 13°S (called the northern region) and the other between the northwest coast of Australia and 13°S (called the southern region). Net surface heat fluxes in the model's northern region are low compared to those in most climatologies, possibly because the model does not allow for strong tidal mixing within the Indonesian archipelago. The annual mean vertical velocity in the upper 500 m shows strong upwelling in the northern region and weak downwelling in the southern region. The annual variation in the vertical motion near the surface (<50 m) is driven by local wind stress in both regions. Within the depth range of the thermocline, vertical motion is dominated by a remotely forced semiannual signal in the northern region and a local response to wind stress in the southern region. The surface heat flux and advection in the annual mean heat budget are very small, at least an order of magnitude smaller than those in the seasonal budget. In the southern region the model may represent the ocean heat budget quite well: here the primary control on the seasonal variation on SST is the surface heat flux. However, about half of the flux is balanced by advection in the southern region. This suggests that ocean circulation could also play a role in the generation of the SST anomalies north of Australia, which are critical for Australian climate variation as documented by Nicholls (1989). In the northern region, cold advection due to upwelling is largely cancelled by warm advection from within the Indonesian archipelago. This may explain the fact that observed sea surface temperature depressions south of Indonesia are quite small compared to those in other eastern‐boundary upwelling regions. However, model heat fluxes in the northern region may be in error owing to the absence of tidal mixing. In the model the largest terms show that heat transport by Indonesian throughflow and Ekman current is the main heating process and this process is largely balanced by cooling due to upwelling near the coast of Indonesia, and by release of heat to the atmosphere near the northwest coast of Australia.

Mass and heat budgets in the East Australian current: A direct approach

We present a technique which uses the large set of expendable bathythermograph data that are available in the waters off Eastern Australia to correct sampling errors in the mass transport function calculated from the much more limited number of deep hydrology stations. This technique utilizes the close correlation between temperature at a single depth and subsurface steric height, found in these waters, to perform the correction. After corrections are applied, it is found that a mass transport function for the top 2000 dbar can be used to estimate the net geostrophic outflows from the Tasman and Coral seas; it balances Ekman inflows to within −0.4±2.0 Sv (1 Sv = 1.0×106 m3 s−1). The mass transport closure allows a much improved determination of the distribution of individual transport components into the region to be obtained, including a robust estimate of the net southward transport through the whole width of the Tasman Sea (including the East Australian Current (EAC)) of 9.4 Sv, through a section at 28°S between the coast and 171°E. A more direct determination of this Tasman Sea transport, which uses annual mean alongshore currents on the shelf to infer mass transport on the slope, provides an independent confirmation. Mass budgets between density surfaces are also examined; for the deeper layers we find a closure of the mass budget within 10% of the net inflow (a maximum misclosure of 0.9 Sv), suggesting little diapycnal mixing. The geostrophic inflow upwells as it flows south in the EAC at a rate of about 3 Sv at a 250 m depth. The net transport balance along the path also allows the first determination of the heat transport associated with the mean mass transport into the Tasman and Coral seas to be obtained (we have not considered any contributions from eddy heat transport). There is a small net inflow of heat transport to the region of 0.13×1015 W, with more heat entering the Tasman Sea than the Coral Sea. The net upward heat flux of 24 W m−2 (heat loss to the atmosphere) is consistent with values given in climatologies obtained from direct surface flux calculations. The results from a Sverdrup‐Munk model agree broadly with the the observed pattern of the mass transport function, particularly the location of the boundary currents, although the model transports are generally higher.

Improvement in the shallow-water prediction of sonobuoy detection ranges for ASW aircraft using on-site acoustic measurements

Sonobuoy detection ranges predicted at shore stations for use by antisubmarine warfare (ASW) aircraft are based on historical data and on expendable bathythermograph (XBT) data that are typically at least 6 to 12 h old. The accuracy of such predictions is questionable as shown by measurements and analysis. Prediction accuracy needs to be improved to increase the effectiveness and lower the costs of antisubmarine search and barriers using sonobuoys dropped from ASW aircraft. These predictions can be improved during the mission by measuring the on-site sound propagation loss between a specially modified air-dropped electronic sound source, similar in concept to a long-lived MK 84 SUS sound projector, and the acoustic receivers in the air-dropped sonobuoys. The derived propagation loss, used in conjunction with the XBT-based predictions, can be used to more effectively select sonobuoy depths and patterns. The existing MK 84 SUS projector could be modified to be neutrally buoyant and provide 140 ping seconds of pre-programmed transmissions over a 6 h life. The transmissions can be shaped for low probability of intercept or mimicry. Similar applications to improve sonar range prediction for surface ships are also addressed.