It has been suggested that the transport of the Antarctic Circumpolar Current is set, essentially, by the southward Sverdrup flux at latitudes just north of Drake Passage. Although this idea is consistent with observations, it has been critized in that Sverdrup dynamics fail at Drake Passage latitudes. Here, we think of the total transport as being comprised of two components: one associated with the basin-like dynamics to the north of Drake Passage and the other associated with the dynamics of the Drake Passage latitude band itself. The Drake Passage latitudes are often simulated using channel models with bottom topography. For a two-layer channel, large topography effectively blocks geostrophic contours at depth and allows zonally-reconnecting contours in the upper layer. This concentrates the through-channel transport in the model's upper layer. Furthermore, it is argued that the statistical steady state for wind-driven channel flow (that is not too viscous) should be baroclinically unstable. Assuming marginal instability then leads to an estimate of the through-channel transport. / A two-layer primitive-equation channel model with bottom topography and wind forcing is used to test this relationship. Model integrations are made to obtain statistical steady states for a range of parameters. The Rossby radius and the wind strength are varied as the theory predicts that transport should go like the square of the former and be relatively insensitive to the latter. Integrations to test the robustness of these results to model resolution are also conducted.

A numerical model study of the seasonal cycle of sea-ice cover and ocean circulation in the Arctic Ocean is presented. The investigation is carried out in four parts using the coupled sea ice-ocean model of Oberhuber (1993a). The Oberhuber model is the first global ocean general circulation model to use Lagrangian isopycnal coordinates. First, a sensitivity study is given of the sea-ice model, which is both dynamic and thermodynamic. The robustness of the sea-ice component in an uncoupled mode is demonstrated. Secondly, the addition of a snow model to the coupled sea ice-ocean model of Oberhuber (1993a) is described. The inclusion of snow is shown to be important for obtaining a good simulation of ice thickness in both the Arctic and Antarctic. Thirdly, the coupled ice-ocean model is used to investigate the general circulation of the Artic Ocean and its connection with the North Atlantic. The cyclonic motion of the Atlantic layer within the Arctic is correctly simulated. Fourthly, a sensitivity study of the Arctic mixed-layer circulation is presented.

Since the beginning of the sixties, coastal trapped waves have been one of the most popular subjects in physical oceanography. Initially, we discuss their characteristics in the Southeast Hudson Bay, using the Brink and Chapman model. Then we compare the normal modes obtained with the observations made in Kuujjuarapik (Hudson Bay) in 1986 using the technique of Freeland et al (1986). This area is difficult to model because of rapid changes of stratification in the water column during the ice break-up period. The best fit was found for two topographic waves both having a period of three days with phase speeds of 70 cm/s and 13 cm/s. In this work it has been shown that the response function of the system due to atmospheric pressure gradients has a maximum at a period corresponding to the period of the topographic waves.

Observations of tidally forced flow in a constricted region of a highly stratified sound are examined as a problem of two-layer hydraulic exchange. It is shown that the narrowest section and the region downstream of the narrowest section were subject to internal hydraulic control. Bores moved upstream and evolved into packets of internal solitary waves with 3-6 minute period when the tide turned to ebb. Using results from different models of the solitary wave based on the KdV equation, it is shown that the second-order nonlinear term must be included in the two-layer model. The results from a first-order continuously stratified model gave similar good results. This implies that two-layer models may ignore some properties of the real fluid and that the internal solitons are also sensitive to the stratification characteristics of the water column. It is found that the mixing was related to both the vertical velocity shear, the hydraulic characteristics of the flow and the presence of solitary waves.

Comprehension of global oceanic currents and, ultimately, of climate variability requires the use of computer modelling. Although much effort has been spent on the accuracy of traditional finite difference (FD) models used in ocean modelling, there are still concerns, especially since these models have a crude representation of the geometry of oceanic basins. Such a crude representation may influence the accuracy of modelling boundary currents, or unrealisticly represent the impinging of eddies or the propagation of Kelvin waves along the coastline. This motivated the use of alternative modelling techniques applied on completely irregular geometries such as finite element (FE) and spectral element (SE) methods. In this thesis, we want to investigate the accuracy and cost-effectiveness of these three numerical methods in irregular domains and to understand to which extent the unstructured grid FE and SE methods constitute an improvement over the more traditional FD methods. To accomplish this, we limit ourselves to modelling the shallow water equations in presence of irregular coastlines with no bottom topography. / In the first part of the thesis, we compare the performances of FD methods on Cartesian grids with FE and SE methods in various geometries for linear and non-linear applications. We argue that the SE method is to a certain extent superior to FD methods. In a second part, we study the influence of step-like walls on vorticity budgets for wind-driven shallow water FD models. We show that vorticity budgets can be very sensitive to the FD formulation. This has certain implications for using vorticity budgets as a diagnostic tool in FD models. In the final part, we use a SE shallow water model for investigating the "inertial runaway problem" in irregular domains for the single-gyre Munk problem. Ideally, one would like the statistical equilibrium observed at large Reynolds number to be insensitive to model choices that are not well founded, e.g., the precise value of the viscous coefficient, and choice of dynamic boundary condition. Simple models of geophysical flows are indeed very sensitive to these choices. For example, flows typically converge to unrealisticly strong circulations, particularly under free-slip boundary conditions, even at rather modest Reynolds numbers. This is referred to as the "inertial runaway problem". We show that the addition of irregular coastlines to the canonical problem helps to slow considerably the circulation, but does not prevent runway.

The tidal flow characteristics around L'Anse du Petit Mitis were described using velocity data from six moored current metres. L'Anse du Petit Mitis is a shallow bay in the lee of the downstream slanting Pointe Mitis. The predominantly M$ sb2$ current showed the following signs of tide induced transient recirculation within the bay: the M$ sb2$ phase within the bay is 65$ sp circ$ ahead of that outside of the bay; the amplitude of the M$ sb4$ and Eulerian mean constituents are relatively large within the bay; and the M$ sb2$ tidal ellipse and the Eulerian mean are oriented towards the tip of Pointe Mitis. The effects of tide induced transient recirculation were not found to extend to the station furthest within the bay. Flow in the bay region was found to be an order of magnitude weaker than the free stream tidal flow.

The interactions between the atmosphere and the oceans play a critical role in determining our climate. These generally consist of various exchanges of heat, mass and momentum between the two media through the air-sea interface. Therefore, the physical state of the upper few meters of the oceans influences the rate at which these exchanges take place. Furthermore, these surface waters are of importance in affecting the primary biological production of the seas. In this context, knowledge of the upper-layer monthly averaged thermohaline state, i.e. temperature (T) and salinity (S) as a function of latitude-longitude and depth, is necessary for further climatological/oceanographic studies in the Gulf of St. Lawrence (GSL). The primary goal of this research is to produce, using basic statistical analyses, monthly mean fields of T and S related quantities at various depths throughout the GSL. The historical hydrographic dataset covers the last 75 years. / Objective fields of sea surface temperature (SST), salinity (SSS) and chlorophyll-a (Chl-a) were also computed and compared with other similar climatologies (when available). (Abstract shortened by UMI.)

In part 1, the spatial and temporal relationships between subarctic Canadian sea-ice cover and atmospheric forcing are investigated by analyzing sea-ice concentration, sea-level pressure, and surface air temperature, as well as ice thickness, runoff, and sea surface temperature, from 1953-1988. The sea-ice anomalies in Hudson Bay, Baffin Bay and the Labrador Sea are found to be related to the North Atlantic Oscillation (NAO) and the Southern Oscillation (SO). From the application of a spatial Student's t-test and a Monte Carlo simulation, we found that sea ice responds significantly to the SO event only in summer, and to the NAO event only in winter. / A spectral analysis shows that sea-ice cover in Baffin Bay and the Labrador Sea responds to the SO and SAT fluctuations at about 1.7 year, 3-5 year and about 8-10 year periods. In addition, a sea-ice signature associated with the so-called "climate jump" during the early 1960s was found. The ice thickness and ice-melt date data are also used to verify the above findings. / In part 2, the three-dimensional ocean general circulation model of Blumberg and Mellor (1983, 87) was used to simulate the winter and summer ocean circulation in Hudson Bay under specified atmospheric forcing and runoff. This model uses vertical sigma coordinates and horizontal orthogonal curvilinear coordinates. The vertical viscosity and diffusivity are computed using the Mellor-Yamada (1982) second-order (2.5) turbulence closure, while the values for similar horizontal parameters are calculated using the Smagorinsky (1963) parameterization. The new version of this model uses the semi-implicit scheme for the shallow water equations (Casulli, 1990). A consistent, modified radiation boundary condition has been developed for the surface elevation and the normal barotropic velocity for the universal multi-level, baroclinic model with strong vertical stratification. The surface cyclonic circulation in summer and winter, due in part to the boundary inflow from Roes Welcome Sound to the northwest of the domain, has been well simulated. / In part 3, a dynamic thermodynamic model of sea ice with viscous-plastic rheology (Hibler, 1979; 1980) is used to simulate the seasonal cycle of sea-ice motion, thickness, compactness, and growth rate in Hudson Bay under monthly climatological atmospheric forcing. The simulated results for ice cover in other seasons also compare favourably with the observed climatology and with measurements from satellites. In particular, the model gives complete sea-ice cover in winter and ice-free conditions in late summer. (Abstract shortened by UMI.)

A series of numerical experiments are conducted using the Bryan-Cox Ocean General Circulation Model to investigate the potential existence of low-frequency variability of the thermohaline circulation under seasonal forcing. Experiments are performed with different combinations of a seasonal cycle being present or not on the restoring temperature, the surface freshwater flux fields (mixed boundary conditions) and the surface wind forcing. / Despite the presence of the forcing on the dominant seasonal timescale, it is found that the system may oscillate at the decadal period or longer. The decadal variability is excited by changes in the net surface density flux which are due to the advection of temperature and salinity anomalies in the model domain. The magnitude of the seasonal cycle also plays an important role in determining the timescale of variability. Violent overturning events may occur on the century timescale under seasonal forcing. The magnitudes of the flushes are reduced compared to those found in similar experiments without the presence of a seasonal cycle.

A numerical modelling approach is chosen to study equilibrium and time-dependent aspects of the ocean's thermohaline circulation. In the first part, the roles of basin geometry and surface buoyancy forcing in determining the asymmetry of the present-day thermohaline circulation are considered. An idealized flat-bottomed two-basin model of the Atlantic and Pacific is found to favour equilibria with sinking in the southern hemisphere only (Southern Sinking) or also in the North Atlantic (Conveyor), even under a freshwater flux forcing field with more precipitation over the North Atlantic than over the North Pacific. / Another new result is the range of Conveyor equilibria found under mixed boundary conditions. Rare cases with North Pacific sinking are characterized by a very fresh halocline in the Southern Ocean and a reversed pole-to-pole surface density contrast. A more quantitative investigation leads to an approximately linear relationship between the Atlantic overturning and the meridional gradient of zonally-averaged depth-integrated steric height from the northern boundary of the ocean to the southern tip of Africa; on the other hand, the local linear relationships postulated in most two-dimensional plane models of the overturning circulation could not be validated. / In the second part, the climatology of a global ocean model is presented, and the importance in the model of the warm water route of the Conveyor through the Indian Ocean relative to the cold water route through Drake Passage is noted. The implied ocean heat and freshwater transports from the Canadian Climate Centre second generation atmospheric general circulation model are then presented, and are shown to be incompatible with the present-day thermohaline circulation. / Finally, in the third part, a simple new parameterization of the sea surface temperature-evaporation feedback is developed as an extension of the traditional mixed boundary conditions. The positive sign of the feedback for the thermohaline circulation is demonstrated, and three examples featuring decadal, century and millennial timescale variability in one-hemisphere idealized basins are discussed. No fundamental alterations of the mechanisms under mixed boundary conditions are found, although the timescale is altered or the variability interrupted sooner in some cases.