Piston cores and single-channel seismic profiles were collected from the western South Orkney Plateau to investigate glacial history, survey the seismic stratigraphy, and test the feasibility of paleoceanographic interpretation as a site survey for O.D.P Leg 113 drilling operations. Data reveal evidence for past grounded ice to 240 meters depth, and expanded floating ice cover over the entire plateau. Paleoceanographic interpretation is difficult because nearly 75% of slope cores are affected by sediment gravity flow. All dip-oriented seismic lines show large-scale slumping. Surface sample textural data indicate that wind-driven currents are redistributing sediments in a predictable pattern to about 450 meters depth. Total grain size analyses are necessary to differentiate ice-rafted-debris from other sands. Seismic data support previous interpretations of a passive margin setting, and show up to three seismic sequences within sediment fill on the margin and plateau.

The Caribbean carbonate crash was a time of increased regional carbonate dissolution at the middle to late Miocene transition. It is marked by five dissolution episodes, occurring from 12-10 Ma, characterized by significant reductions in carbonate mass accumulation rates (CO$\sb3$ MAR). We determined carbonate content and CO$\sb3$ MAR for sites 998-1000 over the middle to late Miocene interval. Stable isotope composition of benthic foraminifers is used to track changes in deep water masses. Carbonate mineralogies are determined for the shallow site 1000 (927 m water depth) to detect dissolution of metastable carbonates at sub-thermocline depths. The geochemical changes during the carbonate period are similar to those that occur during the Caribbean Quaternary interglacial stages, times when Antarctic Intermediate Water (AAIW) filled the Caribbean to abyssal depths. The initiation of North Atlantic Deep Water (NADW) production during this period, possibly caused by the partial closing of the Isthmus of Panama and the opening of Pedro Channel (northern Nicaragua Rise), led to reorganization of global thermohaline circulation. The increase of return flow that passes through the Caribbean may have brought corrosive AAIW into the Caribbean, causing dissolution of carbonate sediment at the sea floor.

Fluorine is abundant in the oceans, and carbonate fluorapatite precipitation in marine sediment is a significant removal mechanism. A paucity of data has led to several misconceptions of fluorine geochemistry and carbonate fluorapatite (CFA) diagenesis, a fluorine and carbonate substituted apatite (Ca10 (PO4)5.74(CO3)0.26F 2.26). This thesis will revise and expand our understanding of fluorine and CFA in deep sediment by proving fluorine's availability, documenting CFA precipitation, and demonstrating alkalinity controls on fluorine and CFA. Because CFA is the main mineral phase of economic phosphorite deposits, its development has warranted much study. However, previous research focused on PO43- and F concentrations in pore waters of shallow sediment (<10 meters) and is incomplete. Here, previous work is improved by directly measuring fluorine concentrations in sediments and pore waters from deep cores (>100 meters) near the upwelling zone of the Peru margin. These results are then related to CFA diagenesis.

Previous reconstructions for anomaly 5 and older anomalies flanking the Southwest Indian Ridge (SWIR) have indicated no statistically significant evidence for motion between the Nubian and Somalian plates. Recently, an analysis of current plate motion across the SWIR indicates significantly different Nubia-Antarctica and Somalia-Antarctica angular velocities. Herein the motion across the SWIR is examined since chron 5 (11 Ma) and chron 6 (20 Ma). I identify 238 Anomaly 5 crossings, 140 Anomaly 6 crossings, and many fracture zone crossings. The new results show that the Nubia-Antarctica and Somali a-Antarctica rotations differ significantly. The results have several implications. (1) Earlier motion differs significantly from that since 3.2 Ma, implying that Nubia-Somalia motion began before 3.2 Ma. (2) Poles of rotation for motion since chron 5 and since chron 6 poles differ significantly from those after 3.2 Ma suggesting a component of right-lateral shearing has occurred along the East African rift since 11 and 20 Ma. (3) Data suggest that the boundary between Nubia and Somalia is narrow where it intersects the SWIR near the Andrew Bain fracture zone complex.

Examination of Late Pleistocene to recent coastal lithosomes on the east Texas continental shelf can help refine current models aimed at understanding how coastal environments respond to a variety of forcing mechanisms including changes in subsidence rate, sediment supply (climate), and eustasy. High-resolution seismic data, sediment cores, paleontologic data, and radiocarbon ages were examined from bay complexes (fluvial, bay-head delta, middle bay, tidal delta, and barrier shoreline environments), offshore banks, incised valleys, and the Brazos Delta. Within the study area, preservation of these deposits on the shelf has been variable. Coastal lithosomes had a high preservation potential in the eastern portion of the study area (around the Trinity/Sabine incised valley) and a low preservation potential in the western portion of the study area (offshore of Follets Island and the Brazos Delta). Sabine, Heald, Shepard, and Thomas banks, located above and adjacent to the Trinity and Sabine incised fluvial valleys, represent submerged barrier shorelines. The shoreline submergence events have been correlated with flooding surfaces located within the Trinity incised valley. Maps of the paleoenvironments bound by these flooding surfaces indicate that each paleoshoreline submergence event is associated with estuarine environments being shifted tens of kilometers landward. Rising sea level during the Holocene was the forcing mechanism behind these events. Freeport Rocks Bathymetric High, a bank located further to the west, also represents a submerged barrier shoreline. However, this feature was not emplaced during the Holocene transgression; rather, it was deposited during the middle Wisconsin (oxygen isotope stage 3) sea-level highstand. The shoreline associated with the bank was mapped regionally at -15 m +/- 2 m. This suggests sea level was around 15--30 m shallower than what oxygen isotope and coral records indicate for stage 3. A detailed sedimentary and geomorphologic study was undertaken on the Brazos Delta, Texas to better define the facies architecture and controlling processes on wave-dominated delta evolution. The Brazos Delta is primarily composed of fine-grained sediments; prodelta clay composes more than half of the sediment volume. The facies architecture is not representative of the classic strandplain model for wave-dominated deltas due to the strong influence of floods on deltaic evolution.

The Maldive Archipelago in the equatorial Indian Ocean is only the uppermost part of a more than 3-km thick carbonate platform. The Maldive platform contains a 50 Ma-long sedimentation record and has a relatively simple tectonic history. The interpretation of 6000 km of Shell 2-D seismic data and information from two industry and three ODP wells was the basis for the reconstruction of the platform evolution and assessment of controls on platform development. The evolution of the Maldives platform was essentially twofold. During the first stage (Eocene-early Oligocene), tectonic control played the dominant role in the establishment and geographic distribution of shallow water carbonates. A series of shallow water carbonate platforms were formed in the early Eocene on basement highs separated by two deep, narrow, and continuous graben systems. The platforms aggraded and backstepped in the Eocene and early Oligocene in response to relative sea level rise driven mostly by tectonic subsidence. The second stage of the platform evolution (late Oligocene-Quaternary) was predominantly controlled by sea level fluctuations. A significant fall in sea level at the early-late Oligocene transition, with a magnitude possibly up to a 100 m, was recorded in the paleo-bathymetry of the Shell ARI-1 well. In the late Oligocene and early Miocene, the platforms first aggraded, partially drowned, and later backstepped in response to a substantial long-term sea-level rise. At the end of the early Miocene, a series of aggrading flat top carbonate banks, a small remnant of the Eocene-Oligocene neritic carbonate system, were established on the periphery of the central basin, the predecessor of the modern Inner Sea of the Maldives. During the middle Miocene, the bank margins prograded for 10--15 km. The progradation was driven by five complete sea level cycles, with each cycle represented by a relative sea-level fall and a subsequent rise. The reconstructed late Oligocene-middle Miocene relative sea level history of the Maldives corresponds well with the newly-published ice-volume record based on the temperature-corrected benthic foraminifera oxygen isotope data. The late Oligocene-middle Miocene depositional geometries of the Maldive platform appear to have recorded eustatic sea-level fluctuations.

The primary objectives of this study were to understand the origin of the Belize Barrier Reef and the sequence stratigraphy of the mixed siliciclastic-carbonate system. The data used consisted of 1,400 km of conventional multichannel seismic, 1,300 km of single-channel high resolution seismic, and information from nine wells. Conventional seismic shows elongated NNE-SSW highs and lows (Camels Basin, Camels Hump-Turneffe atoll, Gladden Basin, Glovers atoll-Lighthouse Island). The thrusted block of Camels Hump formed from buttressing of the Maya Mountains against the moving Caribbean plate during the Paleocene. Its load generated Camels Basin, whereas Gladden and Turneffe Basins formed as pull-aparts. The transtension produced the Turneffe and the Glovers-Lighthouse alignments whose tops as well as Camels Hump's were covered by carbonate platforms during the late Eocene/Oligocene. These carbonates were partially drowned in the early/middle Miocene. During the middle/late Miocene, tectonic enhancement of Camels Hump and Glovers highs triggered the collapse of their margins, shedding mass flow deposits. The late Pliocene/early Pleistocene sea-level fall moved the shoreline along the eastern flank of Camels Hump and in the northern Camels Basin where longshore currents redistributed the sediments. The Belize Barrier Reef became established on top of these lowstand siliciclastic coastal deposits in the middle of the Brunhes Epoch (about 0.45 Ma). Once the barrier reef was formed, it established the modern mixed siliciclastic-carbonate system. When sea-level dropped, about 120 m during the Last Glacial Maximum (18,000 years ago), the reef built during the previous interglacial highstand was karstified. The shelf lagoon became a fluvial plain drained by two incised valley systems. The northern fluvial system deposited a lowstand delta at the mouth of the English Cay Channel. This delta was partially reworked by longshore currents. The rise of sea-level slowed at about -65 to -70 m, 11,000 to 10,000 years ago (Younger Dryas) and is documented by a landward-stepped delta. As sea-level rise resumed, the incised valleys became filled with fluvial to estuarine sediments and, then, buried under marine marls. Flooding at about 7,000 years ago reactivated the reef.

Ice algae are an important component of the carbon cycle in the Arctic, and can therefore have an impact on climate. I investigate the dynamics of an ice algae bloom by coupling an algae-nutrient model [Lavoie et al., 2005] with a thermodynamic sea ice model [Huwald et al., 2005]. The sea ice component is a more sophisticated model than what has been used in past Arctic ice algae model studies. To validate the model, I simulate an algal bloom at the base of the ice over a season and compare with data from the Resolute area in the Canadian Arctic Archipelago. Results suggest that bloom dynamics are strongly related to the ice growth/melt rate, with ice melt being the trigger for bloom decline. Being able to accurately model physical conditions is essential before ice algae can be accurately modeled, and some recommendations for improvement are discussed. / Les algues de glace sont une composante importante du cycle du carbone dans l'Arctique et peuvent avoir une influence sur le climat. J'explore la dynamique d'une efflorescence algale par le couplage d'un modèle d'algues /nutriments [Lavoie et al., 2005] avec un modèle de la thermodynamique de la glace de mer [Huwald et al., 2005]. Le modèle de glace est plus sophistiqué que les modèles utilisés dans les études précédentes sur les algues de glace Arctiques. Pour valider le modèle, je simule une efflorescence algale à la base de la glace pendant une saison et on compare avec des données de la région de Resolute dans l'Archipel Canadien Arctique. Les résultats indiquent que la dynamique de l'efflorescence est fortement liée au taux de croissance et de fonte, la fonte étant le déclencheur du déclin de l'efflorescence. La capacité de modéliser avec précision les conditions physiques est essentiel pour simuler correctement les algues de glaces, et quelques améliorations sont suggérées.

In this thesis an analysis of the variability of the liquid freshwater (FW) export from the Arctic Ocean on annual and seasonal timescales is presented. Due to missing long-term observations, the variability of the liquid FW export is not well known or understood. Model simulations are therefore currently the only way to study the variability of the FW export from the Arctic. / To investigate the role of the atmospheric forcing for the variability of the liquid FW export, a model simulation for 1950-2007 from the University of Victoria Earth System Climate Model (UVic ESCM) is analyzed. It is shown that large-scale atmospheric circulation changes generally control the variability of the FW export through changes in the FW storage in the Beaufort Gyre. These changes have a large influence on the variability of the FW export through the Canadian Arctic Archipelago (CAA), whereas the Fram Strait FW export is also influenced by changes in the FW storage in the Eurasian basin. / In order to better understand the differences between the mechanisms driving the export variability through Fram Strait and the CAA, passive dye tracers are added to the ocean module of a state-of-the-art global general circulation model, the Community Climate System Model Version 3 (CCSM3). These tracers allow the identification of FW from different sources, and therefore the individual investigation of the export variability of FW from individual sources. It is shown that the Fram Strait FW export is made up mainly of Eurasian runoff and Pacific FW, whereas the FW exported through the CAA comes primarily from Pacific FW and North American runoff. The variability of the FW exports from individual sources is largely in phase in the CAA, as the CAA FW export is mainly driven by velocity anomalies, not FW concentration anomalies. In Fram Strait on the other hand, FW concentration anomalies contribute as much to the FW export variability as velocity anomalies. The variability of the Fram Strait FW concentrations from the two main FW sources is not in phase, as Pacific FW and Eurasian runoff have different pathways to Fram Strait and their variability is governed by different mechanisms. Whereas the Eurasian runoff export depends strongly on the release of FW from the Eurasian shelf during years with an anticyclonic circulation anomaly (negative Vorticity index), the variability of the Pacific export is mainly controlled by changes in the Pacific FW stored in the Beaufort Gyre, with increased export during years with a cyclonic circulation anomaly (positive Vorticity index). A high vertical resolution of the ocean model is found to be important to resolve the role of FW concentration changes for the Fram Strait FW export variability. / The model simulation also shows that in contrast to the interannual variability, the seasonal variability of the Fram Strait FW export is driven almost entirely by the seasonal cycle of sea-ice melt, with a smaller influence of velocity changes or advected FW concentration changes. The disappearance of the summer sea-ice cover in the Arctic during the 21st century might therefore affect the seasonal cycle of the Fram Strait FW export. / Cette thèse de doctorat présente une étude de la variabilité du flux d'eau douce de l'océan Arctique vers l'Atlantique Nord. Parce qu'il existe peu d'observations sur ce flux d'eau douce, sa variabilité n'est pas bien connue. Par conséqeuent, des simulations numériques sont nécessaires pour l'étude du flux d'eau douce. / Premièrement, nous avons utilisé le "Earth Sytem ClimateModel" de l'université de Victoria (UVic ESCM) pour analyser la variabilité du flux d'eau douce pour la période 1950-2007. Nos résultats indiquent que la circulation atmosphérique détermine la variabilité du flux d'eau douce de l'Arctique par son influence sur le tourbillon de Beaufort. Les changements de la circulation dans le tourbillon de Beaufort ont une grande influence sur le flux d'eau douce par l'archipel canadien. Le flux d'eau douce par le détroit de Fram est aussi influencé par les changements de la circulation dans le tourbillon de Beaufort, mais également par les changements de la circulation océanique dans le bassin européen / Par la suite, nous avons implanté des traceurs, représentant les différents types d'eau douce dans l'Arctique, dans un autre modèle climatique: le "Community Climate System Model Version 3" (CCSM3). Ces traceurs nous permettent d'analyser la variabilité du flux d'eau douce en détails. Nous remarquons que la plupart de l'eau douce exportée par le détroit de Fram provient des fleuves eurasiens et de l'océan Pacifique (passant du Pacifique à l'Arctique par le détroit de Béring). Par contraste, le flux d'eau douce par l'archipel canadien est principalement composé d'eau douce provenant de l'océan Pacifique et des fleuves d'Amérique du Nord. Les variablités associées au flux d'eau douce provenant de différentes sources par l'archipel canadien sont en phase. Ceci est dû au fait que la variabilité du flux d'eau douce est controllée par la vitesse de l'eau dans l'archipel canadien. Par contre, la variabilité du flux d'eau douce par le détroit de Fram est controllée par la vitesse de l'eau et aussi par la concentration en eau douce. En outre, le flux d'eau douce provenant de sources différentes ne sont pas en phase dans le détroit de Fram parce que le trajet de l'eau douce provenant des fleuves eurasiens et celui de l'océan Pacifique sont différents et leur variabilités sont controllées par des mécanismes différents. Nous remarquons que le flux d'eau douce provenant des fleuves eurasiens par le détroit de Fram dépend du transport d'eau douce du plateau eurasien. L'eau douce quitte le plateau eurasien quand la circulation atmosphérique est anti-cyclonique (l'index de vorticité est positif). Par contraste, le flux d'eau douce provenant de l'océan Pacifique est plus fort quand le tourbillon de Beaufort est réduit, et cette situation se produit lorsque la circulation atmosphérique est cyclonique (l'index de vorticité est dans ce cas negatif). De plus, nous remarquons qu'une haute résolution spatiale est nécessaire pour représenter la / Par ailleurs, les simulations numériques avec le CCSM3 révèlent que la variabilité saisonnière du flux d'eau douce est déterminée par la fonte de la glace de mer dans le détroit de Fram. La variabilité saisonnière de la vitesse a peu d'influence sur la variabilité saisonnière du flux d'eau douce, et ne fait que retarder d'un mois le maximum et minimum du flux d'eau douce. Parce que la fonte de la glace de mer dans le détroit de Fram détermine la variabilité saisonnière du flux d'eau douce par le détroit de Fram, la disparition de la glace de mer dans l'Arctique en été au cours du 21e siècle (selon les prédictions des modèles climatiques) pourrait changer la variabilité saisonnière du flux d'eau douce.

Western boundary intensification is a major feature of the ocean circulation. It is understood that the change in the Coriolis parameter with latitude is the principal component responsible for this phenomenon. However, traditional arguments for western boundary intensification also call for an interplay between forcing and dissipation. In this thesis, a new geometry, namely the periodic meridional channel, is compared to the closed basin setting in decaying, two-dimensional, beta-plane turbulence. By isolating the effects of the western and eastern boundaries, persistent western boundary intensification is shown to be possible as a balance between nonlinear effects and Rossby wave reflections. The conditions under which this occurs are explained, and particular attention is given to the importance of resolution in resolving the Rossby wave dynamics at the boundaries. A discussion of the regimes that do not exhibit western boundary intensification is also included. / L'intensification des frontières ouest est une caractéristique majeure de la circulation océanique. Il est reconnu que la variation du paramètre de Coriolis avec la latitude est l'élément principalement responsable de ce phénomène. Néanmoins, les arguments traditionnels pour l'intensification des frontières ouest demandent aussi la présence d'une force externe et de dissipation. Dans cette thèse, une nouvelle géométrie, c'est-à-dire le canal méridional périodique, est comparée à l'environnement du bassin fermé dans la turbulence libre sur le plan beta en deux dimensions. L'isolation des effets aux frontières ouest permet l'émergence d'une intensification des frontières ouest persistante en tant qu'équilibre entre les effets non linéaires et la réflexion des ondes de Rossby. Les conditions sous lesquelles ce processus est réalisé sont expliquées, et une attention particulière est portée à l'importance de la résolution pour résoudre la dynamique des ondes de Rossby aux frontières. Une analyse des régimes qui ne manifestent pas d'intensification des frontières ouest est aussi incluse.