摘要:

A numerical model of oscillatory rough‐turbulent boundary layer flow, featuring closure via the turbulent energy equation, is used to examine the principal features of wave‐current interaction above the seabed. Results are presented from case studies carried out with water depth of 10 m, bed roughness of 0.5 cm and wave period of 8 s. The steady surface current in the absence of waves is nominally 100 cm/s, and waves having near‐bottom velocity amplitudes of 50, 100, and 150 cm/s are superimposed on this current at angles ϕ = 0, π/4, and π/2. Velocity, turbulent energy, shear stress, and eddy viscosity distributions are compared for the steady current alone, for waves alone, and for waves superimposed on the current, and the interaction in the wave‐current boundary layer is examined. Cycle‐averaged vertical profiles of horizontal velocity illustrate the extent to which the mean flow is retarded by wave‐current interaction. For the general case in which the waves are superimposed at an arbitrary angle ϕ, the mean current veers to form an angle with the wave direction greater than that of the initial steady current. Finally, the enhancement of the bed shear stress and the increase in oscillatory boundary layer thickness associated with wave‐current interaction ar

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00491

年代:1988

数据来源: WILEY

摘要:

The interaction of a geostrophic current with a transverse submarine canyon is considered with an analytical, two‐level model of a homogeneous fluid on anfplane. The dynamical hypothesis is that the flow in the canyon is forced by the pressure gradient due to a sloping free surface. The interaction of flow in the canyon with the overlying fluid is through the vertical velocity forced by divergence in the canyon. The mathematical problem considers a flat‐bottom ocean with an infinitely long canyon having a rectangular cross section and uniform depth. The dynamics are rotationally modified, shallow‐water equations for each level. The initial time behavior of the model is considered with a power series solution which shows that flow in the canyon responds quickly, in about 0.1 of the inertia period, to an imposed surface pressure gradient. The time‐dependent model is solved by means of a Laplace transform in time in three spatial domains: over the canyon and on either side. For a canyon of arbitrary width, the flow over the canyon is composed of standing waves, while radiating waves exist on either side of the canyon. In the absence of friction, the oscillations continue indefinitely, since there are standing gravity modes in the canyon. In steady state, there is no net flow of water across the canyon, so such submarine features should act as barriers to barotropic geostrophic flow. The limit of a narrow canyon is considered. Arguments are presented to show that a canyon is narrow if its width is less than half of the smaller of the radius of deformation or the width scale of the overlying current. For a forcing current with a width about equal to the radius of deformation, the oscillations in the canyon have periods of about 0.1 up to 1. times the inertia period. Such oscillations have been observed in most canyons for which current meter observation

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00509

年代:1988

数据来源: WILEY

摘要:

Current observations obtained from the northern Bering shelf region in winter 1984–1985 indicate that midwinter circulation patterns differed from those during the rest of the year. Annual mean circulation in this region is dominated by vigorous northward currents (from 10 cm/s in Shpanberg Strait to more than 50 cm/s in Bering Strait). These northward currents are accompanied by a weak northwestward coastal flow south of St. Lawrence Island. In February–April 1985, however, flow was weakly (0–5 cm/s) southward through Shpanberg Strait and southeastward south of St. Lawrence Island while remaining northward through Anadyr and Bering straits. Comparison with current data obtained from previous years suggests that midwinter reversal of flow in Shpanberg Strait is a recurrent winter feature. It generally occurs at about the same time that the winter ice cover undergoes its major advance, and it is associated with strong northerly winds. Northward flow through Anadyr Strait persists, conversely, through the winter and is the northward extension of a western intensified northward flow over the Bering Shelf. Its source is Bering Sea water farther south, rather than water from the northern central shelf. The resultant “trapping” of shelf water on the northern central shelf, coupled with intensive regional ice formation during the same period as the flow reversal, led in 1985 to significant (about 1.5‰) brine enrichment. This salinity increase agrees qualitatively with historical data, which suggest that the northern central Bering shelf is the site both of sluggish winter circulation and of elevated salinities due to brine enrichment from ic

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00516

年代:1988

数据来源: WILEY

摘要:

The sea ice margin northwest of Spitsbergen is schematically represented by a stationary wedge maintained by opposing ice and warm Atlantic water advection. Ice melt and cooling of the surface water generate a mixed layer with properties approximating those found elsewhere in the Arctic basin. The heat energy required to melt the advancing ice is roughly comparable to the rate of heat loss of the West Spitsbergen Current and suggests that direct heat exchange with the atmosphere is less important. Only a small fraction of the nearly fresh water from ice melt exits through eastern Fram Strait, and most of it appears to return to the Arctic basin. Whether or not it contributes to the formation and maintenance of the Arctic mixed layer depends on the largely unknown circulation pattern north of Fram Strait. An example of the large interannual variation of the mean wind stress field is shown to illustrate the likelihood of an equally large variance of the ice and surface water circulation.

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00527

年代:1988

数据来源: WILEY

摘要:

The generation mechanism of internal waves by tidal flow over Stellwagen Bank in Massachusetts Bay is qualitatively clarified. First, in order to demonstrate the continuous generation process of internal waves, a numerical simulation is carried out for the hydrographic condition during a field study by Chereskin (1983). The resulting internal waveforms agree well with the acoustic images obtained at several stages of the tidal flow. Next, the generation mechanism for these internal waves is analyzed via a physical interpretation of the calculated result. This analysis is made by use of characteristics which describe the propagation of the first‐ and second‐mode internal waves. It is shown that each mode internal wave with an upstream phase propagation is efficiently amplified while being carried downstream toward the location where the maximum tidal flow becomes critical: with the decrease of the tidal flow, the internal wave thus amplified begins to propagate upstream. In previous studies, these internal waves were interpreted as quasi‐steady lee waves. The present analysis, however, provides a satisfactory explanation for the time‐dependent features that are shown in numerical simulation. This indicates that the internal waves over the bank are transient waves formed in response to a time‐varying tidal flow, and not quasi‐stead

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00533

年代:1988

数据来源: WILEY

摘要:

An in‐time spectral method is used to solve the complete nonlinear problem of tide propagation simultaneously over oceanic and coastal areas. For each constituent of the tidal spectrum, the corresponding elliptic modal problem is written under a variational form and solved by using a finite element technique. The model is applied to the computation of the dominant M2wave over the northeast Atlantic. The numerous in situ data over this domain and the large extent of the European continental shelf allow analysis of the gain of accuracy obtained by different components characterizing the model. To get a mean accuracy of less than 2 cm and 2° for the amplitudes and phases of the sea surface elevation, it is found that (1) the use of a refined mesh over the shallow areas and of a quasi‐linearized version of the quadratic Chezy friction law is essential; (2) loading effects computed by means of a Green integral have to be taken into account; and (3) accurate boundary conditions need to be prescribed at the entrances of dynamically active shallow water areas if they are not included in the investigated area (or it is necessary to extend the model over the whole neighboring coastal basin if these boundary conditions are not available). From the M2velocity field solution, bottom friction coefficients are derived for the secondary astronomical constituents of the tidal spectrum, and accurate charts of the S2and N2tides are produced over the modeled

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00543

年代:1988

数据来源: WILEY

摘要:

The true available potential energy (APE) of an isolated feature (e.g. a warm‐core ring) in a basin is the difference in the potential plus internal energy between the observed and reference state. The reference state is that in which the same fluid is redistributed adiabatically to a stably stratified state and the isopycnals are on constant geopotential surfaces. For an isolated feature in an infinitely wide basin, it would seem reasonable to assume that the far‐field density structure is a good approximation of the reference state. This approximation to the reference state will be examined in this paper. Four different methods of calculating the available potential of an isolated feature are compared. These are (1) the gravitational available potential energy using the reference state (APEG), (2) the gravitational available potential energy using the far‐field density structure as an approximation for the reference state, (3) the Boussinesq available potential energy using the reference state (APEB), and (4) the Boussinesq available potential energy using the farfield density structure instead of the reference state. The gravitational available potential energy using the reference state (APEG) is the true available potential energy for an incompressible fluid. Therefore the different approximations of the available potential energy can be compared to the APEGof an isolated feature in an infinitely wide basin. It is found that the gravitational available potential energy using the far‐field density structure has an offset which depends on the density structure of both the feature and far field. Therefore the APEGusing the far‐field density structure is not a good method for calculating the APEGof an isolated feature in an infinitely wide basin, and the reference state must be calculated. We examine the limitations on the Boussinesq approximation for the APE. It is found that the isopycnal displacement from its reference state must be smaller than a length scale based on the reference state buoyancy frequencyNr2. Also, the maximum separation of isopycnals in the observed state must be less than twice the average separation of the isopycnals. With these conditions satisfied, the APEBapproximates the APEGof an isolated feature. The Boussinesq APE using the far‐field density structure instead of the reference state gives the asymptotic limit of the APEGof an isolated feature in an infinitely wide basin. A simple stratified model of an isolated feature is used to demonstrate the difference between these different available potential energy estimates. Finally, the APE estimates using some data from a survey of a Mediterranean salt lens a

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00556

年代:1988

数据来源: WILEY

摘要:

It is demonstrated that the characteristic temperature‐salinity relationship shown by thermocline waters of the Arctic Ocean can be reproduced using a simple model based on transfer of heat directly from these waters to sea ice. From this we deduce that there is direct sensible heat flux from waters of Atlantic origin to the ice cover, such interaction could play a major role in determining the physical properties of Arctic water masse

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00565

年代:1988

数据来源: WILEY

摘要:

The HEXOS Experiments In the Simulation Tunnel (HEXIST) are part of the continuing Humidity Exchange Over the Sea program (HEXOS). It is actually an international cooperative subprogram focused on the local interactions between spray droplets and the turbulent fields of velocity, temperature, and humidity. It incorporates numerical modelling and wind tunnel experiments. In order to study separately the generation of droplets, their transport and diffusion by turbulence, their evaporation, and their deposition, a new experimental approach has been designed, based on the generation of spray droplets by the bursting of bubbles produced by aeration devices immersed in the water tank of the large air‐sea interaction simulation tunnel at the Institut de Mécanique Statistique de la Turbulence, Marseille, France. This paper presents this approach, the experimental setup, the governing equations, and the results of the feasibility experimen

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00572

年代:1988

数据来源: WILEY

摘要:

Combining recently reported measurements with those reviewed earlier, a general description of concentrations and size distributions of bubbles at various depths in the near‐surface ocean under a wide range of wind velocities is propose

ISSN:0148-0227    

DOI:10.1029/JC093iC01p00587

年代:1988

数据来源: WILEY