Variações temporais na tensão de cisalhamento do vento, na Plataforma Continental Sudeste do Brasil (PCSE), perturbam o equilíbrio geostrófico gerando movimentos com frequências próximas à inercial local. Estas perturbações se propagam horizontalmente e verticalmente, interagindo com o fluxo médio e alterando as características hidrográficas e hidrodinâmicas. Foram analisados dados observacionais de corrente, registrados por correntógrafos fundeados às isóbatas de 50 m e 100 m, ao largo de Arraial do Cabo (RJ) e Ubatuba (SP), de vento registrados por bóias meteorológicas e de salinidade e temperatura perfilados em função da profundidade em estações hidrográficas, obtidos no âmbito do Projeto DEPROAS (Dinâmica do Ecossistema de Plataforma da Região Oeste do Atlântico Sul), entre 2001 e 2002. Os resultados das análises indicaram que a média de duração dos eventos de oscilações inerciais é 7,5 dias com desvio padrão de 3,8 dias, sendo 6,8 o número médio de oscilações em cada evento (desvio padrão de 3,3 oscilações). O período inercial efetivo médio foi calculado em 26,5 h com deslocamento médio da frequência inercial em 12,2%, devido ao desvio Doppler causado pelas interações com a vorticidade relativa do fluxo básico. As correntes inerciais horizontais, filtradas a partir das séries correntográficas, apresentaram valores entre 3 e 10 cm/s. Foi observada propagação vertical das oscilações inerciais e calculado o valor da velocidade de grupo vertical em -2,59 .10-2 cm.s-1. A partir desse resultado, foi obtido o valor de 28,3 m de profundidade para o valor máximo de velocidade vertical das partículas de água na onda interna inercial forçada pela oscilação das isotermas, em acordo com resultados observados de máxima corrente inercial logo abaixo da camada de mistura. O valor calculado para a amplitude da oscilação vertical das isotermas foi 17,2 m, próximo aos 19 m registrado com os resultados observacionais em períodos de mudança do campo de ventos. O modelo numérico hidrodinâmico ROMS (Regional Ocean Modelling System) comprovou resultados observacionais em relação à capacidade de mudanças na tensão de cisalhamento do vento, em períodos menores que o inercial local, gerarem oscilações inerciais. Os resultados numéricos com o ROMS permitiram analisar a interação das oscilações inerciais com o fluxo médio em diferentes regiões da PCSE, através de diferentes condições de estratificação da coluna de água e, a partir da trajetória de pseudoderivadores, estimar os raios das trajetórias com ordem 10 km. / Temporal variations in the wind stress, on the South Brazil Bight, disturb the geostrophic equilibrium and generate motion with near inertial frequencies. The disturbances propagate horizontally and vertically, interacting with the basic motions and altering the hydrodynamics and hydrographic characteristics. Observational data of currents, collected by currentmeters deployed in the 50 m and 100 m isobaths, offshore Arraial do Cabo (RJ) and Ubatuba (SP), of winds collected in meteorological buoys and of salinity and temperature vertically profiled in hydrographic stations, during the project DEPROAS (Dinâmica do Ecossistema de Plataforma da Região Oeste do Atlântico Sul), in the years of 2001 and 2002, have been analyzed. Results showed that the mean duration of the inertial oscillations events is 7.5 days, with standard deviation of 3.8 days, being 6.8 the mean number of oscillations in each event (standard deviation of 3.3 oscillations). The calculated mean effective inertial period is 26.5 h, with a mean difference of the local inertial frequency of 12.2% due to the Doppler shift caused by interactions with the relative vorticity of the basic flow. The horizontal inertial currents, filtered from the current time series, presented values between 3 and 10 cm/s. Vertical propagation of the inertial oscillations were observed and the calculated value for the vertical group velocity was -2.59 . 10-2 cm/s. From these results, the calculated value for the depth of maximum water vertical velocity was 28.3 m for the internal inertial wave forced by the isotherms oscillations, in accordance with results observed of maximum vertical velocities occurring below the mixing layer. The calculated value for the amplitude of vertical oscillations of isotherms was 17.2 m, approximately equal to the 19 m observed in the data set during periods of wind shift. The numerical hydrodynamical model ROMS (Regional Ocean Modeling System) confirmed the observational results that changes in the wind field with periods smaller than the local inertial period generate inertial oscillations. The numerical results with ROMS allowed to analyze the interaction between the inertial oscillations and the basic flow in different regions of the South Brazil Bight, due to vertical different stratification characteristics of the sea water, and from the trajectories of pseudo-drifters it was possible to estimate the radius of the trajectories as having order of magnitude of 10 km.

This study focuses on the habitats of cetaceans in the Mid-Atlantic Bight, a region characterized by bathymetric diversity and the presence of distinct water masses (i.e. the shelf water, slope water, and Gulf Stream). The combination of these features contributes to the hydrographic complexity of the area, which furthermore influences biological productivity and potential prey available for cetaceans. The collection of cetacean sighting data together with physical oceanographic data can be used to examine cetacean habitat associations. Cetacean habitat modeling is a mechanism for predicting cetacean distribution patterns based on environmental variables such as bathymetric and physical properties, and for exploring the potential ecological implications that contribute to cetacean spatial distributions. We can advance conservation efforts of cetacean populations by expanding our knowledge of their habitats and distribution. Generalized additive models (GAMs) were developed to predict the spatial distribution patterns of sperm whales (Physeter macrocephalus), pilot whales (Globicephala spp.), bottlenose dolphins (Tursiops truncatus), and Atlantic spotted dolphins (Stenella frontalis) based on significant physical parameters along the continental shelf-break region in the Mid-Atlantic Bight. Data implemented in the GAMs were collected in the summer of 2006 aboard the NOAA R/V Gordon Gunter. These included visual cetacean survey data collected along with physical data at depth via expendable bathythermograph (XBT), and conductivity-temperature-depth (CTD) instrumentation. Additionally, continual surface data were collected via the ship’s flow through sensor system. Interpolations of physical data were created from collected point data using the inverse distant weighted method (IDW) to estimate the spatial distribution of physical data within the area of interest. Interpolated physical data, as well as bathymetric (bottom depth and slope) data were extracted to overlaid cetacean sightings, so that each sighting had an associated value for nine potentially significant physical habitat parameters. A grid containing 5x5 km grid cells was created over the study area and cetacean sightings along with the values for each associated habitat parameter were summarized in each grid cell. Redundant parameters were reduced, resulting in a full model containing temperature at 50 m depth, mixed layer depth, bottom depth, slope, surface temperature, and surface salinity. GAMs were fit for each species based on these six potentially significant parameters. The resultant fit models for each species predicted the number of individuals per km2 based on a unique combination of environmental parameters. Spatial prediction grids were created based on the significant habitat parameters for each species to illustrate the GAM outputs and to indicate predicted regions of high density. Predictions were consistent with observed sightings. Sperm whale distribution was predicted by a combination of depth, sea surface temperature, and sea surface salinity. The model for pilot whales included bottom slope, and temperature at 50 m depth. It also indicated that mixed layer depth, bottom depth and surface salinity contributed to group size. Similarly, temperature at 50 m depth was significant for Atlantic spotted dolphins. Predicted bottlenose dolphin distribution was determined by a combination of bottom slope, surface salinity, and temperature at 50 m depth, with mixed layer depth contributing to group size. Distribution is most likely a sign of prey availability and ecological implications can be drawn from the habitat parameters associated with each species. For example, regions of high slope can indicate zones of upwelling, enhanced vertical mixing and prey availability throughout the water column. Furthermore, surface temperature and salinity can be indicative of patchy zones of productivity where potential prey aggregations occur. The benefits of these models is that collected point data can be used to expand our knowledge of potential cetacean “hotspots” based on associations with physical parameters. Data collection for abundance estimates, higher resolution studies, and future habitat surveys can be adjusted based on these model predictions. Furthermore, predictive habitat models can be used to establish Marine Protected Areas with boundaries that adapt to dynamic oceanographic features reflecting potential cetacean mobility. This can be valuable for the advancement of cetacean conservation efforts and to limit potential vessel and fisheries interactions with cetaceans, which may pose a threat to the sustainability of cetacean populations.

This thesis reports the development, testing, and application of computer programs for simulating Lagrangian-Stochastic particle dispersion in coastal seas, with particular application to tide and storm induced dispersion in South Australian seas. The three-dimensional tidal equations are briefly discussed for the two types of surge models used, and finite-difference methods for numerically solving these equations are considered. Different methods of simulating flows at open sea boundaries are investigated. The method of particle tracking and the development of the particle tracking model is also described. Various tests are conducted to investigate both the advective and diffusive aspects of dispersion, and a number of scenarios for the simulation of open (ocean) and closed (coastal) boundaries are considered. Various aspects of the particle tracking routine are given specific characteristics according to the nature of the particle being considered. Application of the tide and storm surge model to the Great Australian Bight is described. This uses spherical polar co-ordinates to account for the curvature of the earth, and an oblique boundary element to increase accuracy of the coastline representation. The effect of a low pressure system moving from west to east across the Bight and the resulting significant observed surge at Thevenard during the storm of April 1996 is simulated. This storm resulted in a significant number of deaths in aquaculture farms containing southern bluefin tuna (Thunnas maccoyii) within the Boston Bay region to the extreme east of the Bight due to the agitation of almost neutrally buoyant organic sediments at the sea floor. The effects of this storm are further considered using a Cartesian co-ordinate fine-grid local model of Boston Bay, in Spencer Gulf, South Australia, where both tidal and storm (wind and outside surge) induced flows are simulated. The dispersion of suspended neutrally buoyant sediment throughout the region is considered, and compared with the mortalities of tuna at various farms within the region. Tidal and storm induced currents in the Gulf St. Vincent region, South Australia, have also been modelled using Cartesian co-ordinates. Detailed consideration has been given to the modelling of tides, winds, atmospheric pressures and outside surges from the two open boundaries in Investigator Strait and Backstairs Passage. The information obtained has enabled the modelling of a number of storm surge scenarios. Further to this, various simulations of the dispersion of the larvae of the western king prawn (Penaeus latisulcatus) have been driven using the storm surge model developed. These incorporate currents near the surface and the sea floor, as well as the consideration of changes in behaviour during the life history of the larvae. / Thesis (Ph.D.)--Engineering, Computer and Mathematical Sciences, 2000.

Continental margin sediments play an important role in marine biogeochemical cycles, partly due to high primary production rates and efficient export of organic matter to sediments in margin environments. This thesis presents studies of solute exchange in fine-grained sediments representative of slope and rise environments, and carbon and nitrogen cycling in sandy sediments dominant in continental shelves worldwide. Results of these studies advance understanding of the role of benthic processes on marine ecosystems. In fine-grained sediments, solute exchange by diffusion, biological mixing and bioirrigation can be quantified using in situ flux chambers with inert tracer additions. Mechanistic models of chamber tracer transport across the seabed indicate that in organic-rich sediments, bioirrigation and mixing dominate over a wide range of bottom water oxygen levels, reflecting the patchiness and versatility of benthic macrofaunal communities. Positive correlations between benthic oxygen and tracer fluxes appear site-specific. Reliable chamber volume estimates derived from mechanistic models reveal that empirical fits to chamber tracer datasets may overestimate chamber volume and benthic solute fluxes. The biogeochemistry of sandy, highly permeable sediments that dominate continental shelves is largely unknown because of the difficulty in sampling and studying them under natural conditions. Novel sediment reactors were developed and used to mimic in situ porewater advection and natural sedimentary conditions. Compositional changes of natural seawater, with and without the addition of

This thesis reports the development, testing, and application of computer programs for simulating Lagrangian-Stochastic particle dispersion in coastal seas, with particular application to tide and storm induced dispersion in South Australian seas. The three-dimensional tidal equations are briefly discussed for the two types of surge models used, and finite-difference methods for numerically solving these equations are considered. Different methods of simulating flows at open sea boundaries are investigated. The method of particle tracking and the development of the particle tracking model is also described. Various tests are conducted to investigate both the advective and diffusive aspects of dispersion, and a number of scenarios for the simulation of open (ocean) and closed (coastal) boundaries are considered. Various aspects of the particle tracking routine are given specific characteristics according to the nature of the particle being considered. Application of the tide and storm surge model to the Great Australian Bight is described. This uses spherical polar co-ordinates to account for the curvature of the earth, and an oblique boundary element to increase accuracy of the coastline representation. The effect of a low pressure system moving from west to east across the Bight and the resulting significant observed surge at Thevenard during the storm of April 1996 is simulated. This storm resulted in a significant number of deaths in aquaculture farms containing southern bluefin tuna (Thunnas maccoyii) within the Boston Bay region to the extreme east of the Bight due to the agitation of almost neutrally buoyant organic sediments at the sea floor. The effects of this storm are further considered using a Cartesian co-ordinate fine-grid local model of Boston Bay, in Spencer Gulf, South Australia, where both tidal and storm (wind and outside surge) induced flows are simulated. The dispersion of suspended neutrally buoyant sediment throughout the region is considered, and compared with the mortalities of tuna at various farms within the region. Tidal and storm induced currents in the Gulf St. Vincent region, South Australia, have also been modelled using Cartesian co-ordinates. Detailed consideration has been given to the modelling of tides, winds, atmospheric pressures and outside surges from the two open boundaries in Investigator Strait and Backstairs Passage. The information obtained has enabled the modelling of a number of storm surge scenarios. Further to this, various simulations of the dispersion of the larvae of the western king prawn (Penaeus latisulcatus) have been driven using the storm surge model developed. These incorporate currents near the surface and the sea floor, as well as the consideration of changes in behaviour during the life history of the larvae. / Thesis (Ph.D.)--Engineering, Computer and Mathematical Sciences, 2000.

Deutlich ansteigende interkontinentale Handelsbeziehungen bilden die wesentliche Grundlage das Wachstum der Containerschifffahrt. Die Schnittstelle der Containerschifffahrt zu den landgebundenen Verkehrsträgern bilden Containerterminals. Deren Entwicklung muss im Zusammenhang mit der Containerschifffahrt gesehen werden. Die Entwicklung des Containerschifffahrtsmarktes wird sich auch in Zukunft weiter dynamisch fortsetzen. Prognosen zur Umschlagsentwicklung sind für die Containerterminals von besonderem Interesse, da die Containerterminals bestrebt sind, genügend Kapazitäten bereitzustellen. Im Umfeld eines Containerterminals sowie in ihnen selbst können diverse Kapazitätsengpässe entstehen. In vielen Fällen ist dringender Handlungsbedarf hinsichtlich der Beseitigung von potenziellen Engpässen gegeben. Eine Option stellt dabei der Neubau von Containerterminals dar. Der Hafenbau wird in dichtbesiedelten Räumen zunehmend schwieriger und teurer. Die aus heutiger Sicht zur Verfügung stehenden Flächen werden langfristig nicht ausreichen, um das Umschlagsaufkommen der Zukunft zu bewältigen. Deswegen kann die Errichtung von herkömmlichen Containerterminals langfristig nicht als ausschließliche Option zur Beseitigung von Umschlagsengpässen angesehen werden. Das Konzept eines Offshore-Containerterminals (OCT) stellt einen alternativen Ansatz dar, der zukünftige Engpässe vermeidet oder ihre Auswirkungen mindert. Ziel des Konzepts ist die Verlagerung von Transshipments von den herkömmlichen Containerterminals auf einen OCT. Dieses Konzept wird auf seine wirtschaftliche Tragfähigkeit untersucht. Letzlich stellt das Konzept des OCTs als Transshipment-Hub eine Möglichkeit dar, die unter heutigen Verhältnissen zwar unrentabel erscheint, unter zukünftigem Expansionsdruck und Kostensteigerungen bei der Erschließung neuer Terminalflächen jedoch als Konzept ökonomisch konkurrenzfähig sein kann und eventuell ohne Alternative ist.

Deutlich ansteigende interkontinentale Handelsbeziehungen bilden die wesentliche Grundlage das Wachstum der Containerschifffahrt. Die Schnittstelle der Containerschifffahrt zu den landgebundenen Verkehrsträgern bilden Containerterminals. Deren Entwicklung muss im Zusammenhang mit der Containerschifffahrt gesehen werden. Die Entwicklung des Containerschifffahrtsmarktes wird sich auch in Zukunft weiter dynamisch fortsetzen. Prognosen zur Umschlagsentwicklung sind für die Containerterminals von besonderem Interesse, da die Containerterminals bestrebt sind, genügend Kapazitäten bereitzustellen. Im Umfeld eines Containerterminals sowie in ihnen selbst können diverse Kapazitätsengpässe entstehen. In vielen Fällen ist dringender Handlungsbedarf hinsichtlich der Beseitigung von potenziellen Engpässen gegeben. Eine Option stellt dabei der Neubau von Containerterminals dar. Der Hafenbau wird in dichtbesiedelten Räumen zunehmend schwieriger und teurer. Die aus heutiger Sicht zur Verfügung stehenden Flächen werden langfristig nicht ausreichen, um das Umschlagsaufkommen der Zukunft zu bewältigen. Deswegen kann die Errichtung von herkömmlichen Containerterminals langfristig nicht als ausschließliche Option zur Beseitigung von Umschlagsengpässen angesehen werden. Das Konzept eines Offshore-Containerterminals (OCT) stellt einen alternativen Ansatz dar, der zukünftige Engpässe vermeidet oder ihre Auswirkungen mindert. Ziel des Konzepts ist die Verlagerung von Transshipments von den herkömmlichen Containerterminals auf einen OCT. Dieses Konzept wird auf seine wirtschaftliche Tragfähigkeit untersucht. Letzlich stellt das Konzept des OCTs als Transshipment-Hub eine Möglichkeit dar, die unter heutigen Verhältnissen zwar unrentabel erscheint, unter zukünftigem Expansionsdruck und Kostensteigerungen bei der Erschließung neuer Terminalflächen jedoch als Konzept ökonomisch konkurrenzfähig sein kann und eventuell ohne Alternative ist.

Atmospheric features, such as tropical cyclones, act as a driving mechanism for many of the major hazards affecting coastal areas around the world. Accurate and efficient quantification of tropical cyclone surge hazard is essential to the development of resilient coastal communities, particularly given continued sea level trend concerns. Recent major tropical cyclones that have impacted the northeastern portion of the United States have resulted in devastating flooding in New York City, the most densely populated city in the US. As a part of national effort to re-evaluate coastal inundation hazards, the Federal Emergency Management Agency used the Joint Probability Method to re-evaluate surge hazard probabilities for Flood Insurance Rate Maps in the New York – New Jersey coastal areas, also termed the New York Bight. As originally developed, this method required many combinations of storm parameters to statistically characterize the local climatology for numerical model simulation. Even though high-performance computing efficiency has vastly improved in recent years, researchers have utilized different “Optimal Sampling” techniques to reduce the number of storm simulations needed in the traditional Joint Probability Method. This manuscript presents results from the simulation of over 350 synthetic tropical cyclones designed to produce significant surge in the New York Bight using the hydrodynamic Advanced Circulation numerical model, bypassing the need for Optimal Sampling schemes. This data set allowed for a careful assessment of joint probability distributions utilized for this area and the impacts of current assumptions used in deriving new flood-risk maps for the New York City area.