Water quality in coastal waters is of great socio-economic concern. Anthropogenic activities along the coast have led to an increasing number of impaired waterbodies and degraded ecosystems. To manage water quality issues, accurately modeling the coastal water quality is vitally important. One traditional way to model water quality is using numerical models. Despite great advances in hydrodynamic modeling over the past few decades, water quality simulation is still challenging as the performance of water quality model depends on how well the complex biogeochemical processes are parameterized. While numerical models are the dominant tool for water quality modeling, there are increasing efforts in developing data-driven models in marine sciences. Several major challenges associated with data-driven models for coastal water quality are addressed in this dissertation. These challenges include difficulties in high-dimensional simulation, missing records in observational data, and uncertain watershed loadings. A data-driven model for coastal water quality is introduced in this dissertation. The proposed model has three major components including (1) forcing transformation auto-selection, (2) empirical orthogonal functions (EOF), and (3) neural network. It uses EOF to extract principal components of the target variable and applies a neural network to simulate the temporal variations of nontrivial components. Different from previous empirical models, the approach is able to simulate three-dimensional variations of water quality variables and it does not use in situ measured physical conditions but only external forcings as model inputs. The robustness of the model is verified with applications to predict temporal-spatial distributions of key water quality variables, including dissolved oxygen (DO) and Chlorophyll-a (Chl-a) concentration in Chesapeake Bay. Using a major portion of historical shipboard monthly measurements and corresponding external forcings for training, the model shows good performance in terms of predicting both seasonal and interannual variations for the testing period. The model is also tested for high-resolution simulation using Visible Infrared Imaging Radiometer Suite (VIIRS) Chl-a data. The missing records in the satellite data are effectively interpolated by Data Interpolating Empirical Orthogonal Functions (DINEOF). An overall satisfactory model performance demonstrates that by combining DINEOF and machine learning, it is feasible to use data-driven models to predict high-resolution spatiotemporal variations of water quality variables in coastal waters. Finally, to address the uncertainty in watershed loading, a typically important forcing for coastal water quality, an inverse method is introduced to estimate loading by combining observation and numerical model. In this method, an estuary is divided into multiple segments. Water and material fluxes between neighboring segments are computed from a set of linear equations derived from mass balance and the relationship between residence time and water fluxes. With sparse observational data, inversely estimated loadings agree well with loadings from a previously calibrated watershed model, demonstrating the reliability of the method. Overall, this dissertation demonstrates the feasibility of using data-driven approaches to model three-dimensional coastal water quality. With the rapidly accumulated observational data and quick advances in machine learning techniques, data-driven approaches have great potential for water quality modeling and environmental management in the future.

Global climate predictions hinge on our understanding of the global carbon cycle, and in particular, the role of the Southern Ocean (SO). Sea surface measurements across the SO are sparse and subject to temporal, spatial and seasonal biases. These biases arise from the inaccessibility of the SO due to the high-risk weather conditions and ice coverage experienced during the winter. This study looks at the feasibility of autonomous measuring platforms in the SO for the purpose of reducing the uncertainty bias observed in the SO, constraining the global carbon budget and observing the seasonal cycle of carbonate chemistry in seawater. The high resolution Wave Glider (WG) dataset located in the sub-Antarctic zone (SAZ) of the SO, during the spring-summer bloom period of 2013/2014, resolves the seasonal cycle of TA from pCO2 and pH (WGTA) and the empirical expression for TA (Lee et al., 2006) using salinity and temperature and measurements of pH. The discrepancy between WGTA and the calculated TA gives rise to a summer bias in the seasonal cycle of TA attributed to the uptake of nitrate during the bloom period, and the entrainment of nitrate in the pre-bloom period. The effect of this bias on estimating pCO2 indicates that the amplitude of the pCO2 seasonal cycle may be overestimated by as much as 3.6% during the pre-bloom period. An assessment of the Lee et al., (2006) TA expression (LeeTA) in the SO regime against shipboard observations showed a significant regional different in TA between the Atlantic Ocean and Indian Ocean sectors of the SO at the onset of winter, where LeeTA overestimates TA observations in the Atlantic Ocean sector. This further emphasises the seasonal bias of the TA algorithm which provides an averaged TA across the SO as a whole. Hence, it is proposed that a regional formulation be developed for the prediction of TA in each ocean sector of the SO. To further assess the performance of empirical expressions for TA, the Carter et al., (2014) LIAR expression, utilized by the Southern Ocean Carbon and Climate Observations and Modeling (SOCCOM) project were compared to shipboard underway measurements of TA in the autumn-winter season, and WGTA. The LIAR expression showed a strong dependence on salinity that coincided with the summer bias of the Lee et al., 2006 formulation. This reenforces that estimates of TA in the SO cannot resolve biologically driven changes in the seasonal cycle of TA, and measurements of pH alone are not enough to elucidate the accurate pCO2 estimates if TA is not constrained by the seasonal cycle of nitrate.

Estuaries are highly productive and characterized by complex shoreline geometry and topography. Multiple materials produced within or transported into estuaries include non-living dissolved/particulate materials and living organisms. Estuarine circulation determines material transport and distributions, which further impact estuarine ecosystem to support abundant fauna/flora and their vulnerability to escalated anthropogenic inputs. Short-term material transport process in Chesapeake Bay (CB) has been studied for years, yet its long-term characteristics and the transport of materials with settling and biological behaviors that interact with physical transport have not been fully studied. This study aims to understand the transport of non-living and non-motile dissolved/particulate materials, and motile algae in the bay under different timescales, via hydrodynamic model, transport timescale, remote sensing, and an improved particle tracking model with algal bloom dynamics and vertical migrations.Freshwater discharge and northwesterly wind together contribute more than 90% of the spatiotemporal variations in water exchange (inflow and outflow) in CB at seasonal to annual scales. The outflow responds positively to river discharge, while the inflow increases with river discharge due to enhanced gravitational circulation, then levels, and gradually declines due to overwhelming seaward barotropic currents. A time lag should be considered when computing the mean residence time (RT) of an embayment via its volume and its outflow since RT depends on the forward hydrodynamics whereas outflow was determined by previous hydrodynamics. The geometry and topography of CB impact greatly on material transport. The shallow region near bay mouth, i.e., Rappahannock Shoal, increases water recirculation of surface outflow back into the middle to upper bay, thus, increases material (e.g., dissolved nutrients and organic matter) retention inside the bay. Besides topographic effects, gravitational circulation and river outflow dominate the transport of dissolved materials, whereas tidal contributions are localized near the mouth. For particulate materials, specifically the surface-produced particulates, settling, resuspension, and interactions with bottom sediments dominate their downward transport and vertical distributions. The contribution of freshwater discharge on vertical transport time is weak. The overall “shallow-deep-shallow” topography also prolongs the resting of particulates in the deep-channel sediments. For particulates with biological behaviors, such as the motile harmful algal species Margalefidinum polykrikoide, its vertical migration behaviors alter the transport pathways and bloom concentrations as the latter varies with surrounding environmental conditions. Like estuary is a diverse environment, material transport in CB is subject to different drivers, whose contributions to material transport are not easy to quantify and compare. Transport timescale can help compare the importance of different drivers under a common scale and understand complex aquatic ecosystems, e.g., quantifying retention efficiency associated with topography; quantifying material retention time in water column and sediments to compare transport time with biochemical processes; revealing that the near-mouth shallow region increasing the retention of dissolved and particulate materials inside the bay and regulating the ecosystem of CB. For harmful algal bloom that is highly regulated by shore-term transport and vertical migrations of algae, particle tracking technique with embedded algal bloom dynamics and behaviors could simulate multiple biophysical scales and form the basis for a bloom forecast system in CB.

The Natal Valley is a sediment-filled marine basin situated between the east coast of southern Africa (Natal) and the Mozambique Ridge. Geophysical and sedimentological techniques are used in a broad geological study of the mid Natal Valley. Major emphasis is directed to: (a) basin history and tectonic evolution; (b) seismic stratigraphy of the basin fill; (c) recent sedimentary processes and responses. General basin morphology is defined by five major physiographic provinces: continental shelf and slope, Tugela Cone, Central Terrace, Mozambique Ridge and deep basin plain. Thinned (20-25 km) continental crust, attenuated and subsided in response to Gondwana rifting and drifting, underlies the Central Terrace, Tugela Cone and Mozambique Ridge. Southern margins of the Central Terrace and Tugela Cone are cored by a series of subsea floor ridge and pinnacle complexes (Naude, East Tugela and South Tugela Ridges). Geochemical analyses of East Tugela Ridge basalts suggest a transitional origin but with continental affinities. These volcanic marginal ridges may approximately delineate the continental-oceanic crust boundary (COB) in the Natal Valley. To the south, the deep basin plain is underlain by oceanic crust. / The Natal Valley is a sediment-filled marine basin situated between the east coast of southern Africa (Natal) and the Mozambique Ridge. Geophysical and sedimentological techniques are used in a broad geological study of the mid Natal Valley. Major emphasis is directed to: (a) basin history and tectonic evolution; (b) seismic stratigraphy of the basin fill; (c) recent sedimentary processes and responses. General basin morphology is defined by five major physiographic provinces: continental shelf and slope, Tugela Cone, Central Terrace, Mozambique Ridge and deep basin plain. Thinned (20-25 km) continental crust, attenuated and subsided in response to Gondwana rifting and drifting, underlies the Central Terrace, Tugela Cone and Mozambique Ridge. Southern margins of the Central Terrace and Tugela Cone are cored by a series of subsea floor ridge and pinnacle complexes (Naude, East Tugela and South Tugela Ridges). Geochemical analyses of East Tugela Ridge basalts suggest a transitional origin but with continental affinities. These volcanic marginal ridges may approximately delineate the continental-oceanic crust boundary (COB) in the Natal Valley. To the south, the deep basin plain is underlain by oceanic crust.

Thesis (Ph. D.)--Rutgers University, 2009. / "Graduate Program in Oceanography." Includes bibliographical references (p. 197-210).

Thesis (Ph. D.)--University of Washington, 2003. / Vita. Includes bibliographical references (p. 105-110).

Thesis (Ph. D.)--Naval Postgraduate School, December 2003. / Dissertation supervisors: Curtis A. Collins, Franklin B. Schwing. Includes bibliographical references (p. 95-108). Also available online.

<p> Ocean general circulation models of the IPCC class have biases even when simulating present-day conditions, which may bring into question their predictions of future conditions. This dissertation is about tools for, and results from assessing biases in the Community Earth System Model (CESM) ocean component, by itself and when combined with the Biological Ecosystem Cycling (BEC) model. Newly developed tools and their applications are listed. 1. An offline matrix tracer transport model for the ocean component of CESM. 2. A fast Newton-Krylov implicit tracer equilibrium solver for both the annually-averaged and the seasonally-varying circulation. 3. An effective preconditioner for the solver simulating radiocarbon. Application results: For a natural radiocarbon simulation, an equilibrium solution was obtained in 23 model-years, a dramatic decrease from the 4000 model-years reported for time-stepping. The modeled circulation in the deep Pacific Ocean produced radiocarbon ages twice those of observations. 4. A capability for computing the surface origin of water mass fractions as well as the age of the various water masses. </p><p> Application results: The North Atlantic was the major supplier of ventilated water to not only the Atlantic, but also the Pacific and Indian Oceans. A lack of formation of bottom water in the Southern Ocean was discovered. 5. A capability for restricting the tracer simulation domain to a limited region of the ocean while retaining the effectiveness of advection and diffusion fields on the boundary. This reduces computational costs and allows separating local versus remote impacts of tracer sources on the biogeochemical tracer concentrations. This capability has the potential to provide a platform for further biogeochemical studies. </p><p> Application results: The Indian Ocean region was isolated. Global versus regional circulation effects were determined using radiocarbon. Most of the bias within the region was eliminated by using observational, rather than globally calculated values, on the boundaries. Oxygen production and consumption from a CMIP5 BEC simulation were used to drive a regional oxygen model. Boundary values of oxygen from the CMIP5 BEC simulation were replaced with observations, resulting in less bias within the region. However, significant bias in the location of the Arabian Sea oxygen minimum zone remained.</p>

<p> Coral reefs are greatly impacted by the physical characteristics of the water surrounding them. Incidence and severity of mass coral bleaching and mortality events are increasing worldwide due primarily to increased water temperature, but also in response to other stressors. This decline in reef health demands clearer understanding of the compounding effects of multiple stressors, as well as widespread assessment of coral reef health in near-real time.</p><p> Satellites offer a means by which some of the physical stressors on coral reefs can be measured. The synoptic spatial coverage and high repeat sampling frequency of such instruments allow for a quantity of data unattainable by <i> in situ</i> measurements. Unfortunately, errors in cloudmasking algorithms contaminate satellite derived sea surface temperature (SST) measurements, especially during anomalously cold events. Similarly, benthic interference of satellite-derived reflectance signals has resulted in large errors in derivations of water quality or clarity in coral reef environments.</p><p> This work provides solutions to these issues for the coral reef environments of the Florida Keys. Specifically, improved SST cloudmasking algorithms were developed for both Advanced Very High Resolution Radiometer (AVHRR; Appendix A) and Moderate Resolution Imaging Spectroradiometer (MODIS) data (Appendix B). Both of these improved algorithms were used to reveal the extent and severity of a January 2010 cold event that resulted in widespread mortality of Florida Keys corals. Applied to SST data from 2010, the improved MODIS cloudmasking algorithm also showed improved quantity of SST retrievals with minimal sacrifice in data quality.</p><p> Two separate algorithms to derive water clarity from MODIS measurements of optically shallow waters were developed and validated, one focusing on the diffuse downwelling attenuation coefficient (K_d, m^-1) in visible bands (Appendix C), the other on K_d in the ultraviolet (Appendix D). The former utilized a semi-analytical approach to remove bottom influence, modified from an existing algorithm. The latter relied on empirical relationships between an extensive <i>in situ</i> training dataset and variations in MODIS-derived spectral shape, determined using a stepwise principal components regression. Both of these algorithms showed satisfactory validation statistics, and were used to elucidate spatiotemporal patterns of water clarity in the Florida Keys. Finally, an approach was developed to use Landsat data to detect concurrent MODIS-derived reflectance anomalies with over 90% accuracy (Appendix E). Application of this approach to historical Landsat data allowed for long-term, synoptic assessment of the water environment of the Florida Keys ecosystem. Using this approach, shifts in seagrass density, turbidity increases, black water events, and phytoplankton blooms were detected using Landsat data and corroborated with known environmental events.</p><p> Many of these satellite data products were combined with <i>in situ </i> reports of coral bleaching to determine the specific environmental parameters individually and synergistically contributing to coral bleaching. As such, SST and visible light penetration were found to be parsimoniously explaining variance in bleaching intensity, as were the interactions between SST, wind and UV penetration. These relationships were subsequently used to create a predictive model for coral bleaching via canonical analysis of principal coordinates. Leave-one-out-cross-validation indicated that this model predicted `severe bleaching' and `no bleaching' conditions with 64% and 60% classification success, respectively, nearly 3 times greater than that predicted by chance. This model also showed improvement over similar models created using only temperature data, further indicating that satellite assessment of coral bleaching based only on SST data can be improved with other environmental data. Future work should further supplement the environmental parameters considered in this research with databases of other coral stressors, as well as improved quantification of the temperature at the depth of corals, in order to gain a more complete understanding of coral bleaching in response to environmental stress.</p><p> Overall, this dissertation presents five new algorithms to the field of satellite oceanography research. Although validated primarily in the Florida Keys region, most of these algorithms should be directly applicable for use in other coastal environments. Identification of the specific environmental factors contributing to coral bleaching enhances understanding of the interplay between multiple causes of reef decline, while the predictive model for coral bleaching may provide researchers and managers with widespread, near real-time assessments of coral reef health.</p>

Thesis (Ph. D.)--Johns Hopkins University, 1966. / Bibliography: leaves 66-70.