<p> In the mid-1970s, an enormous open-ocean polynya developed in the Weddell Sea. Since the Weddell Polynya's occurrence, no polynya of similar size or duration has been observed in the region. A polynya of this magnitude could significantly affect global abyssal ocean properties via increased Antarctic Bottom Water (AABW) formation and large Weddell Sea water mass property perturbations. However, the scarcity of 1970s Weddell Sea observations, coupled with the sparseness of abyssal ocean observations, make it difficult to study this phenomenon's oceanic impact without models. This dissertation examines the influence of Weddell Polynyas on abyssal ocean water mass properties and circulation using the GFDL CM2G coupled climate model. </p><p> Abyssal ocean temperature, salinity, and water mass changes resulting from Weddell Polynyas are quantified in CM2G and compared to observations. The model polynyas initially cool the abyssal Southern Ocean and South Atlantic, but 2-3 decades after polynya cessation the same regions warm as they relax toward their mean state. Composites of multiple, spontaneously-occurring polynyas in CM2G reveal that up to 10% of recently observed warming in the abyssal Southern Ocean could be the result of the 1970s Weddell Polynya recovery. </p><p> Weddell Polynya transport mechanisms are also investigated. Polynya signal transport is governed by two processes acting on different timescales and spreading at different rates: 1) topographic and planetary waves that act on interannual-to-decadal timescales, and 2) advection that acts on decadal-to-centennial timescales. Both mechanisms generate property changes on isobaths. Despite different spreading rates, the advective and wave signals act contemporaneously in many Southern Hemisphere abyssal basins. The combined effect and relative magnitude of the two signals dictates the prevailing property changes. </p><p> During Weddell Polynyas, vigorous exchange occurs between the surface and deep waters, resulting in increased abyssal ventilation. In climate models, ideal age tracer is often used to investigate oceanic ventilation. This tracer suffers from several flaws that detract from its suitability as a ventilation diagnostic. We develop a new tracer, ?-age, that rectifies some of ideal age's problematic aspects and examine its utility in an offline tracer model.</p>

Source: Dissertation Abstracts International, Volume: 41-01, Section: B, page: 0113. / Thesis (Ph.D.)--The Florida State University, 1980.

Elemental composition and time series analysis of concentrations of aerosols collected on Hawaii above and below the inversion layer in spring and summer 1979 have provided insight into the sources, transport processes, and chemistry of different aerosol components present in this marine atmosphere. Aerosols were sampled by 3.7-hour time resolution streakers (Nuclepore filter strip and sliding suction orifice) and analyzed by particle-induced X ray emission (PIXE) for elements heavier than Mg. At Mauna Loa Observatory (3.4 km altitude) April 23 to May 6, 1979, during a haze period caused by east Asian desert dust storms, a continental dust component of up to 20 (mu)g m('-3) or more and an aerosol sulfate component, sometimes in excess of 8 (mu)g m('-3), were measured. High coherency at long periodicities between the two components suggests pollution from China and Japan as the sulfur source. Since a 21-hour, not 24-hour, variation was prominent, an initially diurnal periodicity in upward transport foreshortened during transport across the Pacific, e.g. by wind speed convergence, is suggested. Continental dust was also found to be abundant, at 1-2 (mu)g m('-3) median concentration, in spring 1979 at 1.2 km (Hawaiian Volcano Observatory), but was essentially absent in the summer. Over both seasons, the local basaltic aerosol biweekly medians varied 100-fold, 0.2-18 (mu)g m('-3), a variation associated with rainfall. Compositional differences between the continental dust at HVO and MLO may indicate the presence of two distinct desert dust plumes over Hawaii. An additional component of volcanic emissions from Kilauea, containing aerosol P, Cl, and S, was also observed at HVO. The presence of all three elements was sporadic, especially for P, and all were found with concentrations of up to 1 (mu)g m('-3). However, P and Cl were almost never detected simultaneously, suggesting a volatility release of HCl from aerosol particles by reaction with strong acid associated with P. If such strong acid is due to H(,2)SO(,4), it is not well predicted by S, which may include other sulfates or sulfur. / Source: Dissertation Abstracts International, Volume: 43-03, Section: B, page: 0663. / Thesis (Ph.D.)--The Florida State University, 1982.

Models incorporating a single baroclinic mode and realistic coastline geometry are used to analyze the linear, dynamic response to estimates of the interannual wind field over the tropical Pacific and the seasonally varying winds of the tropical Atlantic. The interannual variability of the tropical Pacific is studied for the period January, 1961 to December, 1978. Model pycnocline variations at several locations are similar to the observed sea level fluctuations. El Nino events are depicted as periods when the pycnocline is persistently deep along the eastern boundary. Remotely forced equatorial Kelvin waves are responsible for this response. The character of each simulated El Nisno is strongly dependent on the relation between zonal wind stress changes in the western an central equatorial Pacific. A rapid shoaling of the pycnocline in the western tropical Pacific during each El Nino is caused by westward-propagating Rossby waves. Interannual pycnocline displacements in the central equatorial Pacific are determined by the superposition of Kelvin waves excited to the west and first-mode Rossby waves generated to the east. / The forced periodic response to the seasonal wind field of the tropical Atlantic is a spatially dependent combination of a locally forced response, Kelvin waves, Rossby waves, and multiple wave reflections. The seasonal displacements of the model pycnocline are compared with observed dynamic height. Annual and semiannual fluctuations dominate the seasonal signal throughout the basin. In general, the distribution of amplitude and phase are similar for annual changes in dynamic height and pycnocline depth. Major features of the seasonal response are reproduced, e.g. an east-west tilting of the equatorial pycnocline about a pivot point, the seasonal pycnocline movement along the northern and southern coast of the Guinea Gulf, and a significant change of phase in the ocean variability north and south of the ITCZ. The relative importance between local and remote forcing is determined for several parts of the basin. The wind-driven annual signal in the Gulf of Guinea is due to zonal wind stress fluctuations west of the gulf. The seasonal response in the western equatorial and northernmost parts of the basin are primarily local. / Source: Dissertation Abstracts International, Volume: 43-10, Section: B, page: 3173. / Thesis (Ph.D.)--The Florida State University, 1982.

A model is formulated to study the influence of large coastline and shelf topography variations on the dynamics of stratified, low-frequency wind-driven frictional shelf waters. At low frequencies, currents nearly follow the isobaths. Therefore, the coordinate system should be a curvilinear system in which the coordinate lines are parallel and perpendicular to the isobaths. Writing the low frequency problem for stratified, frictional, time-dependent linear wind forced shelf flow in these coordinates, it is found that, even when the topography varies alongshore on scales comparable to the shelf width, the solution can be expressed as a sum of curvilinear coastally trapped waves (CCTW) scattered by alongshore variations in bottom topography and bottom friction. The CCTW amplitudes satisfy coupled, forced-wave equations of the same form as those of the straight coast case. The unforced CCTWs speed up when the shelf forms an embayment and when there is no curvature and the shelf is wider. / Calculations show that changes in topography and coastline affect both velocity and sea level, especially the velocity. For example, for frictionless flow and a mode 1 CTW incident to a narrowing shelf, the alongshore velocity amplitude changes markedly in the bend region. In this case curvature and conservation of relative vorticity explain the changes in amplitude of the alongshore coastal velocity at the bend. When friction is included the same qualitative changes occur but the changes in velocity amplitude are smaller because of damping. Similar dynamics apply when a shelf widens. / Calculations were also done with measured wind forcing and results compared with measured sea level on the West Florida Shelf. Since the sea level observations were made at the coast, a formula was derived to estimate sea level at the coast given sea level at the model coast. The model successfully calculated the wind-driven West Florida Shelf sea level fluctuations even in the Big Bend region where coastline and bottom topography vary rapidly alongshore. / Source: Dissertation Abstracts International, Volume: 53-11, Section: B, page: 5622. / Major Professor: Allan J. Clarke. / Thesis (Ph.D.)--The Florida State University, 1992.

Since oceanic eddies migrate westward, they eventually reach the western boundaries of the oceans. Therefore, it is of interest to find out in which direction a surface eddy moves after eddies collide with walls. First, we show analytically that, as a result of the image effect, a cyclone moves southward whereas an anticyclone translates northward along a wall. Then, we use an isopycnic, two-layer, primitive equation, numerical model on a $\beta$-plane to study the image effect, the $\beta$-force, and the "rocket" effect altogether. It is numerically found that, out of these three mechanisms, the image effect is the most dominant one in determining the eddy's final migration along the wall. Finally, it is analytically found that, in the presence of an offshore-sloping-down bottom, the topographic effect tends to causes a bottom eddy to migrate southward along the wall. / Our second aim is to study the detailed temporal evolution of an eddy colliding with a wall on an f-plane using a barotropic, as well as a one-and-a-half-layer contour dynamics model. At t = 0, the circular eddy is conceptually cut off by the wall. It is found that, for t $>$ 0, part of the eddy's fluid is gradually advected along the wall and forms a new eddy. Then, the new eddy migrates away from the parent eddy due to the image effect. The results of this contour dynamics study are confirmed by those of the isopycnic, primitive equation model on an f-plane. / Possible applications of these models to various oceanic situations are discussed. / Source: Dissertation Abstracts International, Volume: 53-11, Section: B, page: 5622. / Major Professor: Doron Nof. / Thesis (Ph.D.)--The Florida State University, 1992.

The parameter optimization problem in oceanography is studied, in which an oceanic model with thermodynamics is used to determine the surface thermal forcing field by the adjoint technique. Two datasets are chosen, the climatological monthly-mean sea surface temperature (SST) and winds. The seasonal variability of the surface heat flux distribution over the tropical Pacific ocean is investigated. / The use of a priori information is investigated in the formulation of the cost function. Experimental evidence has verified that adding a priori information of the estimated parameters can increase the probability for the solution to be unique. The a priori information also plays the role of bogus data. It serves not only to increase the number of observations but to improve the conditioning of the Hessian. / The results are very promising. It has been possible, for the first time, to calculate the seasonal surface heat flux patterns which are consistent with both the ocean's physics and the observations. / Source: Dissertation Abstracts International, Volume: 53-11, Section: B, page: 5622. / Major Professor: James J. O'Brien. / Thesis (Ph.D.)--The Florida State University, 1992.

Previous investigations disclosed a strong near-bottom equatorward contour-following flow at the base of the Scotian Rise near the 4,900 m isobath ((TURN)40(DEGREES)N, 62(DEGREES)W) which coexisted with a filament of relatively fresh cold water (termed the Cold Filament); energetic fluctuations with time scales 0(30-90 days) existed in current-meter data. A comparison is made of long (8-12 mos.) records of three near-bottom current-meters with satellite-derived frontal positions of Gulf Stream meanders and rings. The energetic fluctuations coincide with, and most probably result from, the movement of Gulf Stream meanders and rings. However, near the 4,500 m isobath, a 40-day energetic event indicates the apparent presence of topographic Rossby waves, which appear to fit Pedlosky's theory of a bottom-trapped baroclinic topographic Rossby wave resulting from destabilization of a sheared flow by topography. The evidence indicates that the strong equatorward contour-following flow near the 4,900 m isobath is decoupled from, and not part of, the Deep Western Boundary Current, which exists upslope of the 4,000 m isobath on the Scotian Rise. A search of archived hydrographic data of the western North Atlantic shows the Cold Filament to be an ubiquitous feature near the base of the Continental Rise from the Grand Banks Ridge (50(DEGREES)W) to 24(DEGREES)N. The Cold Filament is inferred to be a tracer for an equatorward contour-following flow. A sketch of the partial near-bottom circulation of the western North Atlantic is made based primarily on the Cold Filament; it agrees more closely with a model of the deep circulation by Wunsch and Grant than with a recent model by Hogg. A rough estimate of the dissipation of eddy kinetic energy as a result of the interaction of the Gulf Stream meanders and rings with the bottom indicates that this mechanism may account for roughly 50% of the energy input by the wind into the subtropical gyre of the western North Atlantic. / Source: Dissertation Abstracts International, Volume: 45-04, Section: B, page: 1140. / Thesis (Ph.D.)--The Florida State University, 1984.

Interannual and interdecadal variability of the equatorial Pacific are examined using a new 60-year monthly historical pseudo-stress data set. The monthly mean pseudo-stress fields (1930-1989) are assembled from Comprehensive Ocean-Atmosphere Data Set pseudo-stresses using a variation of Cressman's (1959) objective analysis scheme, climate basis functions obtained from the FSU pseudo-stress product (1966-1990) and a technique called Vector Group Renormalization. The new wind fields are used to force an ocean model (Kubota and O'Brien, 1984). Model estimates of tropical Pacific current and model upper layer thickness (ULT) variability are then obtained for the period of interest. Observed sea level and spatially averaged sea surface temperature (SST) anomalies are used to validate the hindcasts. / Interannual fluctuations in modeled and observed sea level fluctuations are compared at Galapagos and Truk, yielding Correlation (r) values of 0.73 and 0.71, respectively. The comparison of the interannual fluctuations in modeled ULT and observed SST anomalies, which were both spatially averaged over a subdomain in the eastern Pacific basin, yielded a correlation (r) value of 0.64. / Interdecadal fluctuations in eastern Pacific model ULT are found to be qualitatively consistent with those in the spatially averaged observed SST anomalies. Interdecadal variations are shown to play a significant role in modulating the amplitude of El Nino events. Comparison of interdecadal fluctuations in global mean land air temperature and eastern Pacific ULT suggests a connection between eastern Pacific and global mean land air temperature warming for interdecadal time scales. / Source: Dissertation Abstracts International, Volume: 54-12, Section: B, page: 6118. / Major Professor: James J. O'Brien. / Thesis (Ph.D.)--The Florida State University, 1993.

Reduced gravity and two-layer numerical models have been used to study the circulation in the westernmost basin of the Mediterranean Sea, the Alboran Sea. Circulation is forced by flow through a 20 km wide port in the western boundary representing the Strait of Gibraltar. / The reduced gravity model domain is a rectangle measuring 600 km x 160 km with 10km x 5 km grid resolution. When forced by an eastward or northeastward inflow, the model solutions evolve to a steady state which exhibits a meandering current. The first meander of the current forms the northern boundary of an anticyclonic gyre. Horizontal dimensions of the gyre are strongly dependent up on the inflow angle, vorticity associated with the incoming current, magnitude of the incoming transport and the north-south extent of the basin. The meandering current is considered a standing Rossby wave with a highly distorted vorticity trajectory due to the interaction of the current with the northern and southern boundaries. When velocity (transport) is increased, the wavelength increases approximately as SQRT.(v). As a result the anticyclonic gyre shifts east as velocity increases and west as velocity decreases. These solutions show that bottom topography, winds and coastline features are not necessary mechanisms for the formation of the gyre. / Two-layer model solutions were obtained using realistic topography, geometry and a westward moving lower layer. The addition of the lower layer flow and topography slightly distorted the circulation in the upper layer, particularly in the southern half of the basin. With topography included, the lower layer flow followed a path similar to that observed for the Mediterranean Deep Water. When topography was removed, the flow followed the path taken by the Levantine Intermediate Water. / Experiments using climatological wind to force the model show that a strong wind driven circulation from the Balearic Sea causes cyclonic circulation to form in the location of the Alboran gyre. This cyclonic circulation, however, is very weak and when combined with the Atlantic water inflow serves only to slightly weaken the anticyclonic gyre. / Source: Dissertation Abstracts International, Volume: 46-02, Section: B, page: 0458. / Thesis (Ph.D.)--The Florida State University, 1985.