Mesoscale turbulence is ubiquitous in the surface ocean and has significant impact on the large-scale ocean circulation and its interaction with the climate. Ocean currents are most energetic in the mesoscale range on the scales of 20-200 km and recent studies have shown that the surface kinetic energy associated with the mesoscale undergo a large seasonal modulation. At scales below the mesoscale where geostrophic approximation breaks down lies the submesoscale (1-20 km). It is at this scale that baroclinic instabilities feed off the available potential energy stored in the deep wintertime mixed layers, known as mixed-layer instability, and in return energize the mesoscale via inverse energy cascade under the constraint of stratification and rotation. Mixed-layer instability (MLI) is inherently submesoscale due to the depth scale associated with it. We show the robustness of MLI on global scale in modulating seasonality in surface mesoscale turbulence by analyzing outputs from a Community Earth System Model fully ocean-atmosphere coupled run with eddying resolution. Due to the rigorous vertical velocities associated with mesoscale turbulence, in the context of climate, they have been shown to make major contributions to the transport of heat and tracers including carbon. More recently, it has been argued that submesoscale heat transport may dominate over the mesoscale. We ask the same question for tracers: What is the relative contribution of submesoscale transport (local effect) over the energized mesoscale via inverse energy cascade (remote effect)? In order to investigate their impact on the dynamics and tracer transport, we run our own seasonally resolving submesoscale permitting channel model configured to represent the zonal-mean view of the Southern Ocean coupled to a full biogeochemical model. The Southern Ocean is unique in that, apart from it being the only zonally re-entrant basin on Earth, it is one of the high-nutrient low-Chlorophyll oceans and iron is predominantly the limiting nutrient for primary production within the open-ocean region. As the basin responsible for generating the densest water mass properties, i.e. Antarctic Bottom Water, and outcropping isopycnals, primary production and the associated biological carbon pump have been of long interest to the biogeochemical and climate community. We provide an independent estimate from satellite observations of the seasonal cycle in phytoplankton biomass by taking advantage of the biogeochemical Argo floats, in which we show that the biomass reaches its maximum around December in the open-ocean region. Our modelled ecosystem reaches its maximum in November, roughly a month earlier, likely due to the lack of aeolian dust input at the surface, and glacial and bathymetric sources from the south in our model. Utilizing spectral analysis and the generalized Omega equation, we decompose the eddy transport of heat and iron to its submesoscale (local) and mesoscale (remote) contributions. With the exception near the surface where mixed-layer instability is active, our results indicate that mesoscale vertical transport is of first-order significance in calculating the budgets and supplying iron across the mixed-layer base to the surface where phytoplankton can effectively photosynthesize.

The contribution of phytoplankton productivity to higher trophic levels in salt marshes is not well understood. My study furthers our understanding of possible mechanisms controlling phytoplankton productivity, abundance, and community composition in salt marshes. Across three consecutive springs (2001 to 2003), I sampled the upper Nueces Delta in south Texas, a shallow, turbid, salt marsh system stressed by low freshwater inflow and wide ranging salinity (<15 to >300 ppt). Water column productivity and respiration were estimated using a light-dark bottle technique, and phytoplankton biovolume and community composition were determined using inverted light microscopy. To determine their effect on the phytoplankton community, zooplankton and bacterioplankton abundance and several physical parameters were also assessed. Meaningful relationships among the numerous variables evaluated in this study were identified using principal component analysis (PCA). Despite high turbidity, phytoplankton productivity and biovolume were substantial. Resuspension appeared to play a major role in phytoplankton dynamics, as indicated by a positive relationship between ash weight and biovolume that explained up to 46% of the variation in the PCA. Negative relationships between zooplankton grazers and pennate diatoms of optimal sizes for these grazers suggested a functional grazing food chain in this system. Salinity also may have been important in phytoplankton dynamics, whereas nutrients appeared to play a minor role. Salinity increases may have been responsible for a decoupling observed between phytoplankton and grazers during late spring. Findings suggest hypotheses for future studies focused on the role of phytoplankton in salt marshes, particularly those stressed by reduced freshwater inflow and high salinities.

La captación de carbono por la vegetación es a escala global el flujo más grande de CO2 e influencia en gran medida el funcionamiento de los ecosistemas. Sin embargo, su variabilidad temporal y espacial sigue siendo poco conocida y difícil de estimar. Las técnicas de teledetección pueden ayudar a calcular mejor la producción primaria bruta (GPP) terrestre, que es la expresión a nivel de ecosistemas del proceso de la fotosíntesis. El objetivo principal de esta tesis fue encontrar una manera de estimar la variabilidad espacial y temporal de la eficiencia en el uso de la radiación (RUE) a escala de ecosistema y por lo tanto mejorar la estimación de la GPP de la vegetación terrestre por medio de datos de teledetección. Se abordaron cuatro objetivos específicos. El primero fue analizar y sintetizar la literatura científica sobre la relación entre el Índice de Reflectancia Fotoquímica (PRI), un índice espectral vinculado a la eficiencia fotosintética, y diversas variables ecofisiológicas a través de un amplio rango de tipos funcionales de plantas y ecosistemas. El segundo objetivo fue analizar y sintetizar los datos de la variabilidad espacial de la GPP y la variabilidad espacial y temporal de la RUE y sus controles climáticos para un amplio rango de tipos de vegetación, desde la tundra a la selva tropical. El tercer objetivo fue comprobar si diferentes índices espectrales, es decir, el PRI, el NDVI (Normalized Difference Vegetation Index) y EVI (Enhanced Vegetation Index), derivados del Moderate Resolution Imaging Spectroradiometer (MODIS) son buenos estimadores de la captación de carbono a diferentes escalas temporales en un bosque mediterráneo. El cuarto objetivo fue evaluar el uso de MODIS PRI como estimador de la RUE en un amplio rango de tipos de vegetación mediante el uso de datos sobre la captación de carbono de la vegetación derivados de las torres de covarianza turbulenta.Las principales conclusiones de esta tesis son que hay una coherencia emergente de la relación RUE-PRI que sugiere un sorprendente grado de convergencia funcional de los componentes bioquímicos, fisiológicos y estructurales que afectan la eficiencia de captación de carbono a escala de hoja, de cobertura y de ecosistemas. Al complementar las estimaciones de la fracción de radiación fotosintéticamente activa interceptada por la vegetación (FPAR), el PRI permite mejorar la evaluación de los flujos de carbono a diferentes escalas, a través de la estimación de la RUE. Una segunda conclusión apoya la idea de que el funcionamiento anual de la vegetación es más limitado por la disponibilidad de agua que por la temperatura. La variabilidad espacial de la RUE anual y máxima puede explicarse en gran medida por la precipitación anual, más que por el tipo de vegetación. Una tercera conclusión es que, si bien EVI puede estimar el incremento diametral anual de los troncos, y el PRI puede estimar la fotosíntesis neta diaria nivel de hoja y la eficiencia en el uso de radiación, el papel del NDVI es más limitado como un estimador de cualquier parte del ciclo del carbono en bosques mediterráneos. Por lo tanto, el EVI y el PRI son excelentes herramientas para el seguimiento del ciclo del carbono en los bosques mediterráneos. Por último, el PRI derivado de información satelital disponible libremente, presenta una relación positiva significativa con la RUE para un amplio rango de diferentes tipos de bosques, incluso en años determinados, en bosques caducifolios. En general, esta tesis proporciona un mejor entendimiento de los controles espacial y temporal de la RUE y abre la posibilidad de estimar RUE en tiempo real y, por tanto, la captación de carbono de los bosques a nivel de ecosistemas a partir del PRI. / Carbon uptake by vegetation is the largest global CO2 flux and greatly influences the ecosystem functions. However, its temporal and spatial variability is still not well known and difficult to estimate. Remote sensing techniques can help to better estimate the terrestrial gross primary production (GPP), that is the ecosystem level expression of the photosynthesis process or the rate at which the ecosystem's producers capture CO2. The main objective of this thesis was to find a way to estimate the spatial and temporal variability of the Radiation Use Efficiency (RUE) at the ecosystem scale and therefore to arrive to more accurate ways to estimate GPP of terrestrial vegetation by means of remotely sensed data. Four specific objectives were addressed in this thesis. The first objective was to examine and synthesize the scientific literature on the relationships between the Photochemical Reflectance Index (PRI), a narrow-band spectral index linked to photosynthetic efficiency, and several ecophysiological variables across a wide range of plant functional types and ecosystems. The second objective was to analyze and synthesize data for the spatial variability of GPP and the spatial and temporal variability of the RUE and its climatic controls for a wide range of vegetation types, from tundra to rain forest. The third objective was to test whether different spectral indices, i.e. PRI, NDVI (Normalized Difference Vegetation Index) and EVI (Enhanced Vegetation Index), derived from the MODerate resolution Imaging Spectroradiometer (MODIS) can be indicators of carbon uptake at different temporal scales by analyzing the relationships between detailed ecophysiological variables at the stand level in a Mediterranean forest. The fourth objective was to assess the use of MODIS PRI as surrogate of RUE in a wide range of vegetation types by using data on carbon uptake of the vegetation derived from eddy covariance towers. The main conclusions of this thesis are that there is an emerging consistency of the RUE-PRI relationship that suggests a surprising degree of functional convergence of biochemical, physiological and structural components affecting leaf, canopy and ecosystem carbon uptake efficiencies. By complementing the estimations of the fraction of photosynthetically active radiation intercepted by the vegetation (fPAR) PRI enables improved assessment of carbon fluxes at different scales, through the estimation of RUE. A second conclusion supports the idea that the annual functioning of vegetation is more constrained by water availability than by temperature. The spatial variability of annual and maximum RUE can be largely explained by annual precipitation, more than by vegetation type. A third conclusion is that while EVI can estimate annual diametric wood increment, and PRI can estimate daily leaf level net photosynthesis and radiation use efficiency, the role NDVI is more limited as a surrogate of any part of the carbon cycle in this type of forest. Therefore, EVI and PRI are excellent tools for vegetation monitoring of carbon cycle in the Mediterranean forests, the first ones we tested in this thesis. Finally, the PRI derived from freely available satellite information was also found to present significant positive relationship with the RUE for a very wide range of different forest types, even in determined years, the deciduous forests. Overall, this thesis provides a better understanding of the spatial and temporal controls of the RUE and opens the possibility to estimate RUE in real time and, therefore, actual carbon uptake of forests at the ecosystem level using the PRI.Keywords carbon cycle, Normalized Difference Vegetation Index, Enhanced Vegetation Index, Photochemical Reflectance Index, primary productivity, photosynthesis, remote sensing, climatic controls, eddy covariance, radiation use efficiency, terrestrial vegetation.

The contribution of phytoplankton productivity to higher trophic levels in salt marshes is not well understood. My study furthers our understanding of possible mechanisms controlling phytoplankton productivity, abundance, and community composition in salt marshes. Across three consecutive springs (2001 to 2003), I sampled the upper Nueces Delta in south Texas, a shallow, turbid, salt marsh system stressed by low freshwater inflow and wide ranging salinity (<15 to >300 ppt). Water column productivity and respiration were estimated using a light-dark bottle technique, and phytoplankton biovolume and community composition were determined using inverted light microscopy. To determine their effect on the phytoplankton community, zooplankton and bacterioplankton abundance and several physical parameters were also assessed. Meaningful relationships among the numerous variables evaluated in this study were identified using principal component analysis (PCA). Despite high turbidity, phytoplankton productivity and biovolume were substantial. Resuspension appeared to play a major role in phytoplankton dynamics, as indicated by a positive relationship between ash weight and biovolume that explained up to 46% of the variation in the PCA. Negative relationships between zooplankton grazers and pennate diatoms of optimal sizes for these grazers suggested a functional grazing food chain in this system. Salinity also may have been important in phytoplankton dynamics, whereas nutrients appeared to play a minor role. Salinity increases may have been responsible for a decoupling observed between phytoplankton and grazers during late spring. Findings suggest hypotheses for future studies focused on the role of phytoplankton in salt marshes, particularly those stressed by reduced freshwater inflow and high salinities.

Thesis (M.S.)--University of North Carolina at Wilmington, 2003. / Includes bibliographical references (leaves : 41-46).

Graduation date: 1983

Thesis (Ph. D.)--University of Washington, 2007. / Vita. Includes bibliographical references (p. 127-137).

Thesis (Ph. D.)--University of Hong Kong, 1994. / Includes bibliographical references (leaves 275-295).

Thesis (M.S.)--Oregon State University, 1993. / Typescript (photocopy). Includes bibliographical references (leaves 85-87). Also available on the World Wide Web.

Thesis (Ph. D.)--University of Washington, 1996. / Vita. Includes bibliographical references.