Search Results

You are looking at 1 - 10 of 12 items for

  • Author or Editor: James Luyten x
  • Refine by Access: All Content x
Clear All Modify Search
Gilles Reverdin and James Luyten

Abstract

Drifting buoys were released in the western Indian Ocean from 1979 to 1982. They reveal meanders of the eastward flowing monsoon drift in August–September 1979, 1980 and 1981. Oscillating meridional buoy drifts reach 80 cm s−1 and meridional displacements can exceed 3° for motions at a period close to 25 days. In 1979, this is related to oscillation below the thermocline, sensed by an array of current meters in the western equatorial Indian Ocean. It is likely that the origin of the oscillations is in the surface currents north of the equator.

The currents change in October with the formation of an intense eastward equatorial current in which most of the buoys are entrained. As the buoys drift rapidly towards the eastern Indian Ocean, meridional motions are still present but at shorter periods (12 days) than is observed below the thermocline (25 days). It is possible that is still a manifestation of the same oscillations, but with Doppler shifting and a strong influence of nonlinearities. Waves at 25 days are found in the three oceans. Significant differences in the circulation of relevance for the waves are a more intense seasonal cycle of the currents in the Indian Ocean with predominantly eastward currents near the equator and the presence of the waves in the western portion of the Ocean where a very intense varying circulation is found in the proximity of the Somali Coast during the summer monsoon.

Full access
James Luyten and Henry Stommel

Abstract

The combined effects of buoyancy forcing and wind in a gyre-scale steady ocean circulation are modeled using discrete layers with an interfacial flux, not necessarily vertical. The equations for vorticity conservation of the geostrophic flow in this system are fully nonlinear, involving a Jacobian for the layer thicknesses. These equations are written in a form which can be solved by the method of characteristics. The form of these equations invites the interpretation that the geostrophic baroclinic flow is driven by buoyancy and steered by the wind. Two examples are solved and discussed, a subtropical gyre with heating, and a subpolar gyre with cooling. In each case, there are distinct regimes of flow, depending upon whether the characteristics originate at the eastern or western boundaries of the model. A simple geometrical argument illustrates that the difference between these two regimes, the direct and indirect cells, depends upon the sign of the true vertical velocity relative to the interfacial flux.

Full access
Rui M. Ponte and James Luyten

Abstract

Vertical profiles of horizontal current collected in April and June 1979 in the western Indian Ocean revealed the presence of short vertical scale (150–300 m) deep zonal jets, trapped to within 1° of the equator. Meridional velocity records displayed in general higher temporal and spatial variability and significantly less energy at the larger vertical scales as compared with zonal records. Vertical wavenumber spectra of velocity showed no statistically significant peaks. The conspicuous alternating zonal jets were identified with the slightly more energetic (as compared to background levels) 500–429 stretched meter (sm) wavelengths.

For, most vertical wavenumbers, zonal currents were not significantly coherent at distances greater than 400 km downstream. At shorter zonal separations, phase lags were indistinguishable from zero. On the equator, there is an indication of upward shift of approximately 100 sm over the 500–429 sm band. Also on the equator, zonal current (u) and vertical displacement (ξ) were significantly coherent for the 429 sm wavelength with ξ leading u with increasing depth by approximately /2, For the same wavelength. zonal kinetic energy decayed away from the equator on scales comparable to the theoretical Kelvin wave scaling. Both these results suggested the possible presence of Kelvin waves in the records, at the 429 sm band.

Comparison of these findings with previous studies from the Indian and Pacific oceans indicates different temporal and spatial scales for the deep jets in the two basins.

Full access
Rui M. Ponte and James Luyten

Abstract

Analysis of vertical profiles of absolute horizontal velocity collected in January 1981, February and April 1982 in the equatorial central Pacific revealed two significant narrowband spectral peaks in zonal velocity, centered approximately at vertical wavelengths of 560 and 350 stretched meters (sm). Energy in the 560 sm band roughly doubled between the first and last cruises. Tim-lagged coherence results suggested upward phase propagation at periods of about 4 years. East–west phase fines computed from coherence over zonal separations tilted downward towards the west, implying westward phase propagation and zonal wavelengths on the order of 10 000 km. The peak that was centered at 350 sm occured at the vertical scales of the conspicuous alternating flows in the records, generically called the equatorial deep jets in the past (the same terminology is used here). It showed a more steady character in amplitude and a higher signal-to-noise ratio in comparison with the 560 sm peak. The deep jets were best defined as a finite narrowband process in vertical wavenumber (311–400 sm), accounting for 20% of the total variance present in the broad-band energetic background. At the jets' wavenumber band, latitudinal energy scaling compared reasonably well with Kelvin wave theoretical values and a general tilt of phase lines downward towards the east yielded estimates of 10 000–16 000 km for the zonal wavelengths. Time-lagged coherence calculations revealed evidence for vertical shifting of the jets on interannual time scales. Interpretation of both signals in terms of equatorial waves was ambiguous, because of their relatively long spatial and temporal scales compared to the records. The simplest hypothesis of linear waves in a resting basic state ocean could not be rejected, but more complicated physics cannot be ruled out.

At most wavenumber bands, power levels decayed away from the equator over scales broader than the Kelvin wave scale. Within ½° of the equator, zonal current led (lagged) vertical displacement by π/2 with depth for the 933 sm (140–400 sm) band. The result at the 140–400 sm band agrees with the findings of Eriksen (1981) in the western Pacific, and thus seems to be a general feature of the deep equatorial Pacific fields.

Full access
Kathleen O'Neill and James R. Luyten

Abstract

Vertical profiles of horizontal velocity made along 53°E in the western Indian Ocean, during and after he onset of the southwest monsoon in 1976, show features in zonal velocity of relatively small vertical scale. Persistence of the features over the month-long observation period and over 2½ degrees of longitude indicates long temporal and zonal scales. The vertical structure is common to those profiles close to the equator, with no appreciable variation in amplitude or depth. Between 1°30′N and 3°N the phase of the features reverses. There is no evidence of similar features at 5°N.

The data suggest meridionally trapped equatorial waves. Similarities between the observed phenomena and the linear theory of equatorial waves are striking but quantitative comparisons lead us to question its naïve application here. Equatorial intensification is apparent but it does not scale with the Rossby radius of deformation, indicating that Kelvin waves are not dominant at any vertical scale. The phase change and corresponding amplitudes of zonal velocity do not fit a low-frequency, first meridional-mode long Rossby wave. Higher meridional modes produce further inconsistencies. There is no evidence of vertical propagation of the larger scale features.

Although the small vertical scale variability is ubiquitous in the equatorial regions, no satisfactory theory exists at present. Our conclusion is that although there are similarities between observations and linear theory, there are serious discrepancies which may be related to the proximity of these observations to the slanting coast of East Africa.

Full access
James R. Luyten and Dean H. Roemmich

Abstract

Moored records of zonal velocity in the western equatorial Indian Ocean are dominated by motion of semi-annual period. The motion is coherent over an array spanning more than 1000 km along the equator, 160 km across the equator, and 500 m in the vertical The signal has an amplitude of 0.15 m s−1. It has zonal, vertical and meridional length scales which are long compared to the dimensions of the array, and it shows upward propagation of phase. Behavior is characteristic of an eastward propagating equatorial Kelvin wave and a westward long equatorial Rossby wave. The vertical wavelength is an estimated 5000 m, implying an equatorial trapping scale of 400 km. Zonal wavelengths are ∼8000 km for the equatorial Rossby wave and 24000 km for the equatorial Kelvin wave. A downward energy flux, estimated to be 3 × 1016 erg s−1, most likely represents propagation away from surface forcing by the zonal wind.

Full access
James Luyten, Henry Stommel, and Cau Wunsch

Abstract

A three-layer vorticity diagnostic calculation is made for an area of the northern North Atlantic centered at 55°N, 22.5°W. It suggests the circulation there is not driven by vertical Ekman suction due to the wind stress, but due to nearly horizontal due nearly the sloping mean density surface in that portion of the upper ocean mixed by deep wintertime convection. The computed act annual heat gain is −44 ± 8 W m−2. Such diagnostics define a linear relation between buoyancy gain and Ekman pumping and hence serve as a link; through the ocean, between meteorologist's charts of wind stress and heat gain. The meridional transport is strongly influenced by bottom topography; the upward slope of the bottom toward the north permits a large net northward transport.

Full access
Peter R. Gent and James R. Luyten

Abstract

Vertically propagating linear wave calculations using realistic equatorial buoyancy profiles are presented which show the percentage of the downward surface energy flux that reaches the deep equatorial oceans. The percentages vary widely depending upon the buoyancy profile and the equivalent depth but can be as low as 10% on average for equivalent depths between 1 cm and 1 m if the thermocline is sharp. This means that models with constant or weak thermocline buoyancy profiles, which allow all or most downward surface energy flux to reach the deep ocean, are very unrealistic in this respect. Another conclusion is that the observed, very low-frequency, small vertical-scale deep jets cannot be explained by linear wave theory as caused by surface forcing. It is also shown that a WKB analysis of observations can be misleading even if applied to a single vertically propagating wave in a region that excludes the main thermocline. Implications are that comparing estimates of the equivalent depth from the mixed Rossby-gravity wave dispersion relation and a WKB analysis is of little value because the error bars on both estimates are large, and that WKB estimates of downward vertical energy flux into the deep ocean can also be misleading.

Full access
Arthur Voorhis, James R. Luyten, Gerald Needell, and John Thomson

Abstract

Horizontal currants and dynamic height anomalies measured with a vertical profiler in the central equatorial Pacific during January 1981, February 1982 and April 1982 show significant changes in the upper ocean which are analyzed and discussed in terms of the vertical baroclinic modal structure. Most noticeable was a marked increase in average undercurrent speed and transport from February to April 1982, which was accompanied by a raising of sea level and a deepening of the pycnocline. This was due to an average increase in both first and second baroclinic modes. Over half of the increases of the first mode was due to a Kelvin wave pulse propagating eastward through the survey area at an estimated speed of 300 cm s−1. Evidence is presented which indicates that the wave was forced by a tropical storm in the western Pacific. The increase in the second baroclinic mode is attributed to the large-scale relaxation of the trade winds in the central and western Pacific during the spring of 1982.

Full access
James Luyten, Michael McCartney, Henry Stommel, Robert Dickson, and Ed Gmitrowicz

Abstract

Because the volumetric census of deep and bottom water in the North Atlantic Ocean consists of three isolated linear ridges along which heat and salt flow through the main volumetric mode (and point of intersection), it is possible to deduce the expected ratio of heat flux and ratio of salt fluxes measured in the Denmark Strait overflow off Greenland and in the Antarctic Bottom Water near the equator. The weakly stratified layers of upper North Atlantic Deep Water fall on the nearly linear ridge at temperatures above that of the mode.

There is an incompatibility between observed ratio and deduced ratio. It is predicted that a remeasurement of the flux of Antarctic Bottom Water near the equator will show that the previous determination of 4°N is unrepresentatively low.

Full access