Thesis: Ph. D., Joint Program in Biological Oceanography (Massachusetts Institute of Technology, Department of Civil and Environmental Engineering; and the Woods Hole Oceanographic Institution), 2014. / Cataloged from PDF version of thesis. / Includes bibliographical references. / Viral predation on bacteria in the ocean liberates carbon from the particulate fraction, where it is accessible to higher trophic levels, and redirects it to the dissolved fraction, where it supports microbial growth. Although viruses are highly abundant in the ocean little is known about how their interactions with bacteria are structured. This challenge arises because the diversity of both bacteria and viruses is exceedingly high and interactions between them are mediated by specific molecular interactions. This thesis uses heterotrophic bacteria of the genus Vibrio as a model to quantify virus-host interactions in light of host population structure and ecology. The methods developed in this thesis include streamlining of standard bacteriophage protocols, such as the agar overlay, and facilitate higher throughput in the isolation and characterization of novel environmental virus-host systems. Here, >1300 newly isolated Vibrio are assayed for infection by viral predators and susceptibility is found to be common, though total concentrations of predators are highly skewed, with most present at low abundance. The largest phylogenetically-resolved host range cross test available to date is conducted, using 260 viruses and 277 bacterial strains, and highly-specific viruses are found to be prevalent, with nearly half infecting only a single host in the panel. Observations of blocks of multiple viruses with nearly identical infection profiles infecting sets of highly-similar hosts suggest that increases in abundance of particular lineages of bacteria may be important in supporting the replication of highly specific viruses. The identification of highly similar virus genomes deriving from different sampling time points also suggests that interactions for some groups of viruses and hosts may be stable and persisting. Genome sequencing reveals that members of the largest broad host-range viral group recovered in the collection have sequence homology to non-tailed viruses, which have been shown to be dominant in the surface oceans but are underrepresented in culture collections. By integrating host population structure with sequencing of over 250 viral genomes it is found that viral groups are genomically cohesive and that closely-related and co-occurring populations of bacteria are subject to distinct regimes of viral predation. / by Anne Kathryn Marie Kauffman. / Ph. D.

Thesis (Ph. D.)--Joint Program in Oceanography/Applied Ocean Science and Engineering (Massachusetts Institute of Technology, Dept. of Biology; and the Woods Hole Oceanographic Institution), 2011. / Cataloged from PDF version of thesis. / Includes bibliographical references. / Larval supply is an important process linking reproductive output to recruitment of benthic marine invertebrates. Few species-specific studies of bivalve larvae have been performed due to the lack of suitable methods for species identification. This thesis focused on applying a method to identify larvae from field samples from Waquoit Bay, MA using shell birefringence patterns. This method was then used to address variability in larval supply for three bivalve species on weekly, tidal, and hourly scales. Sampling weekly for six months during two years showed large variability in larval concentrations on this time scale. Abundances of most species were related to bay temperature, and species distributions among sampling sites were indicative of transport potential and population coherence. Greater growth of larvae in 2009 compared to 2007 was attributed to more wind-induced mixing and better food availability in 2009. Integrative samples over each tidal event for a 14-day period demonstrated that larvae were mostly constrained by water masses. During a period when there were sharp tidal signals in temperature and salinity, larval concentrations were higher in bay water compared to coastal waters on incoming tides. After a storm event, water mass properties were less distinct between tidal events and a semidiurnal signal in larval concentrations was no longer apparent. The timing of periods of high larval concentrations did not always coincide with periods of highest water mass flux reducing net export in some cases. On an hourly scale, the vertical distribution of larvae affected by water column stratification and strength of tidal flow. Strong currents and a fresh upper layer both prevented larvae from concentrating at the surface. There was little evidence of peaks in larval concentrations associated with a given tidal period. Species-specific data can provide new perspectives on larval transport. For the three species studied, Anomia simplex, Guekensia demissa, and Mercenaria mercenaria, different source areas, patterns for growth, and potential for export were observed. Applying species-specific identification methods to future studies of bivalve larval transport has the potential to relate larval abundance to settlement patterns, an important component of larval ecology and shellfish management. / by Christine Mingione Thompson. / Ph.D.

Thesis (Ph. D.)--Joint Program in Oceanography/Applied Ocean Science and Engineering (Massachusetts Institute of Technology, Dept. of Earth, Atmospheric, and Planetary Sciences, and the Woods Hole Oceanographic Institution), 2003. / Includes bibliographical references (p. 132-139). / Iron (Fe) is an essential micronutrient for marine phytoplankton often limiting phytoplankton growth due to its low concentration in the ocean and thus playing a role in modulating the ocean's biological pump. In order to understand controls on global Fe distribution, the decoupling between Fe and P04 and the sensitivity of surface nutrient concentrations to changes in aeolian iron supply, I use a hierarchy of ocean circulation and biogeochemistry models. I formulate a mechanistic model of iron cycling which includes scavenging onto sinking particles and complexation with an organic ligand. The iron cycle is coupled to a model of the phosphorus cycle. The aeolian source of iron is prescribed. This system is examined in the context of a highly idealized box model. With appropriate choice of parameter values, the model can be brought into consistency with the relatively sparse ocean observations of iron in the oceans. I implement this biogeochemical scheme in a coarse resolution ocean general circulation model, guided by the box model sensitivity studies. This model is also able to reproduce the broad regional patterns of iron and phosphorus. In particular, the high macro-nutrient concentrations of the Southern Oceans result from iron limitation in the model. I define a tracer, Fe* that quantifies the degree to which a water mass is iron limited. Surface waters in high nutrient, low chlorophyll regions have negative Fe* values, indicating Fe limitation, because aeolian surface dust flux is not sufficient to / (cont.) compensate for the lack of iron in upwelled waters. The oceanic residence time of Fe is [approximately] 285 years in the model, confirming that transport plays an important role in controlling deep water [Fe[Tau]]. Globally, upwelling accounts for 40% of 'new' iron reaching the euphotic zone. Due to the potential ability of iron to change the efficiency of the carbon pump in the remote Southern Ocean, I study Southern Ocean surface P04 response to increased aeolian dust flux. My box model results suggest that a global ten fold increase in dust flux can support a P04 drawdown of [approximately]0.25[mu]M, while the GCM results suggest a P04 drawdown of 0.5 [mu]M. / by Payal Parekh. / Ph.D.

Thesis (Ph. D.)--Joint Program in Biological Oceanography (Massachusetts Institute of Technology, Dept. of Biology; and the Woods Hole Oceanographic Institution), 2005. / Includes bibliographical references (p. 157-171). / Bacteriophage have been proposed to be the most abundant organisms on the planet, at an estimated 10³¹ particles globally (Hendrix et al., 1999). The majority of bacteriophage isolates (96%) are double-stranded DNA tailed phages (Caudovirales). These phages possess a distinctive icosehedral head, with a protein tail structure protruding from a single vertex. This organelle determines host specificity and provides the mechanism of passage of the phage genome into the host cell. Phages infecting differing microbial hosts may have access to a global pool of genes, albeit at different levels. Marine cyanobacteria of the genera Prochlorococcus and Synechococcus are numerically dominant photosynthetic cells in the large oligotrophic gyres of the open oceans, and contribute an estimated 30% to the oceanic photosynthetic budget. Cyanophages have been isolated which propagate on many strains of Synechococcus and Prochlorococcus. Cyanophages can effect community structure and succession through lytic infection of their hosts, and have implications in lateral gene transfer, mediated through lysogeny, mixed infections, pseudolysogeny, and transduction. / (cont.) The broad host ranges (between genera) observed in some phages indicates that lateral gene transfer is not confined to cells of the same strain. These phage/host interactions begin by host recognition by the tail of the infecting phage. Few studies have examined the structural proteins of cyanophage, partially due to the lack of a robust protocol for the growth and purification of phage particles. Cyanophage Syn5 is a short-tailed phage isolated from the Sargasso Sea by Waterbury and Valois (1993) which infects Synechococcus strain WH8109. Methods of growing the host cells and the phage, and concentrating the phage by PEG precipitation were developed. These methods led to highly concentrated purified phage stocks, to titers of 1012 particles/ml. Preliminary characterization of the growth of Syn5 gave a burst size of approximately 30 phage/cell and a lytic period of approximately 10 hours when inoculated into exponentially growing host cells acclimated to a temperature of 26⁰C and a light intensity of 50[mu]E m⁻² s⁻¹. Isolation of the phage nucleic acid yielded dsDNA molecules of approximately 40kb. The Syn5 particles were comprised of twelve structural proteins, as determined by SDS-PAGE. / (cont.) The most intense band on the gel was assigned to the capsid protein of Syn5 ([approx.] 35kDa). However, it was not possible to distinguish putative tail proteins via this method. Purified Syn5 particles were sent to the Pittsburgh Bacteriophage Institute for genome sequencing. The completed Syn5 genome was 46,214 bp long with a 237bp terminal repeat. Annotation of the completed Syn5 genome identified 61 putative ORFs, and revealed that Syn5 appeared closely related to the enteric phage T7 and cyanophages P-SSP7 and P60, as determined by gene similarity and synteny, although the genome was [approx.] 10kb longer than T7. Syn5 appeared to possess a more extensive DNA replisome that T7, containing copies of genes that encoded proteins of known T7 host co-factors, such as thioredoxin, utilized by the T7 DNAP. Several large ORFs were identified between the gene encoding the putative tail fiber and the gene encoding the putative terminase. These ORFs encoded proteins similar to some fibrous sequences within the NCBI non- redundant (nr) gene sequence database as of March, 2005; but had unknown functions within the phage. Unlike other recently sequenced cyanophages, SynS did not contain any photosynthetic genes. / (cont.) The structural proteins of SynS, as visualized by SDS-PAGE, were characterized by mass-spectroscopy and N-terminal sequencing. This allowed the assignment of sequences to putative ORFs within the Syn5 genome. The Syn5 particle was comprised of eleven discreet protein chains of molecular weight 152kDa, 139kDa, 99kDa, 90kDa, 66kDa, 60kDa, 47kDa, 35kDa, 22kDa, 21kDa, and 16kDa. The identified proteins included the portal, capsid, two tail tube proteins, and three internal virion proteins. Each of the genes encoding these proteins were found in the same gene order in the Syn5 genome as the corresponding genes were ordered in the T7 genome. There were three unidentifiable proteins within the particle (66kDa, 47kDa, and 16kDa). These mapped to the area of the SynS genome between the gene encoding the putative tail fiber and the gene encoding the putative terminase. No minor capsid or decorative capsid proteins were detected. The copy numbers of the corresponding protein chains were similar to those known for T7, with the exception of the tail fiber, which was present at a number of three chains per particle in comparison to T7's eighteen per particle. / (cont.) Polyclonal antibodies were raised against Syn5 particles. A Western blot with these antibodies showed that the tail fiber and the two unknown fibrous sequences were highly antigenic. This evidence implies that the unknown structures may act as host recognition proteins in addition to the tail fiber. Characterization of these novel proteins may provide insight to the host recognition abilities of cyanophages. An additional study was also carried out, investigating the high temperature limit of the growth of phage P22. The results revealed that the production of infectious particles was limited by the temperature sensitivity of the folding and assembly of the P22 tailspike protein. This work has been published and is included in the Appendix. / by Welkin Hazel Pope. / Ph.D.

Para muitos componentes e estruturas de engenharia a determinação precisa do estado de tensões residuais presente é de fundamental importância para a avaliação de sua integridade estrutural. O método do furo cego (MFC) é um dos métodos mais difundidos para a medição de tensões residuais, no entanto, por se tratar de um ensaio relativamente complexo, a determinação da sua incerteza de medição apresenta uma série de dificuldades, as quais se refletem na inexistência de estudos na literatura que abordem esse assunto de forma completa. Este trabalho tem como objetivo a determinação da incerteza de medição do MFC englobando todas as fontes de incerteza em potencial do método. Para tanto, foi elaborado um procedimento envolvendo a determinação e caracterização das fontes de incerteza tanto na parte experimental como no tratamento matemático dos dados. A partir da caracterização das fontes de incerteza detectou-se como fator principal na determinação da incerteza o erro do operador, que foi possível através do desvio de repetitividade. A fonte de incerteza devido ao erro do operador foi a de determinação mais complexa e envolveu o projeto e a construção de uma máquina especialmente desenvolvida para a aplicação de um estado de tensões homogêneo e que permitisse medir com precisão as tensões residuais com o método do furo cego. Técnicas de medição de tensões residuais por difração de raios-X e monitoramento de tensão com extensometria de resistência elétrica foram utilizadas para a verificação da homogeneidade das tensões nas amostras. A metodologia se mostrou adequada, conduzindo a resultados que permitiram determinar com sucesso a incerteza de medição através de planilhas eletrônicas. / The characterization of the residual stress state of engineering components and structures is of fundamental importance for assessments of their structural integrity. The hole drilling method is one of the most commonly used methods; however, due to its relative complexity, quantifying its measurement uncertainties is not straightforward. It is a semi-destructive method, which means that multiple measurements cannot be repeated in a same location, and stress distributions mean that it is difficult to guarantee that a number of points in the same state will be available in a given volume. This study suggests a global uncertainty measurement which includes a study of repeatability related to the operator. This is thought to be novel in the existing literature. For this, a standard sample and a machine which guarantees a homogeneous stress distribution in the sample were developed. To ensure the homogeneity of the stresses, x-ray diffraction measurements were performed. The results obtained show good repeatability for different operators and good results for the uncertainty of measurements for the method.

This study compares four different methods to measure hydraulic conductivity (K) at two sites on the East Tennessee State University Valleybrook Campus. It compares the K values to each other, to the different K values between the two sites, and to United States Department of Agriculture (USDA) K values. Two field methods, Well Bail Test and Auger Hole Test, and two lab methods, Constant Head Permeameter Test and Grain Size Distribution Test (GSD), were performed on the clay rich Ultisol soils on an East Tennessee hillslope in the Valley and Ridge Physiographic Province. One site was located close to a monitoring well and the other on the floodplain of an existing stream. The Hazen, Alyamani & Sen, and Slichter methods were used to compute K from the GSD Test. The Alyamani & Sen, Slichter, and permeameter methods produced similar K values ranging from 9.52 x 10-6 to 1.25 x 10-3 cm/sec. These are similar to the USDA K values ranging from 9.17 x 10-4 to 2.82 x 10-4 cm/sec. The Hazen method overestimated K and ranged from 8.10 x 10-3 to 1.09 x 10-1 cm/sec. The Well Bail Test yielded a lower K value (ranging from 8.16 x 10-9 to 1.19 x 10-8 cm/sec) than the USDA values as expected for water flow in deeper soil horizons at a depth of 8.50 meters. Comparing these values helped to better understand the difference between various methods to compute the hydraulic conductivity.

Thesis (Ph. D.)--Joint Program in Oceanography (Massachusetts Institute of Technology, Dept. of Earth, Atmospheric and Planetary Sciences; and the Woods Hole Oceanographic Institution), 1999. / Includes bibliographical references. / This study is a geochemical investigation of the evolution of the Kerguelen plume, on the basis of upper mantle and lower crustal xenoliths. Ultramafic xenoliths include harzburgites predominant, a lherzolite, dunites and pyroxenites, whereas lower crustal xenoliths are cumulate gabbros recrystallized under granulite facies conditions. On the basis of the whole rock major element characteristics and trace element abundance patterns in clinopyroxenes, the harzburgites were found to be residues of extensive melting at high pressures within the Kerguelen plume. These were then recrystallized at low pressures and metasomatized by plume generated melts. Details of the metasomatic process were determined from trace element variations in clinopyroxene in connection to texture. This demonstrated that meltrock reaction and the precipitation of new clinopyroxenes occurred by metasomatic carbonatitic melts. It was also found that some of the harzburgites had distinctly unradiogenic Os isotopic compositions and were identified as originating from the sub-Gondwanaland lithosphere. On the basis of major and trace element compositions, the granulite xenoliths were found to be originally gabbroic cumulates formed from plume-derived basaltic melts emplaced at the base of the crust by underplating and subsequently recrystallized isobarically under granulite conditions. The Sr, Nd and Os isotopic compositions of the peridotite and granulite xenoliths demonstrate that the Kerguelen plume is isotopically heterogeneous and displays a temporal progression toward more enriched Sr and Nd isotopic compositions from the Ninetyeast Ridge to granulite xenoliths to Kerguelen basalts and Heard Island basalts. / by Deborah Renee Hassler. / Ph.D.

In mining engineering, blast-induced ground vibration has become one of the major concerns when production blasts are conducted, especially when the mining areas and the blast sites are near inhabited areas or infrastructure of interest. To comply with regulations, a vibration monitoring program should be developed for each mining operation. The vibration level, which is usually indicated by the peak particle velocity (PPV) of the vibration waveform, should fall below the maximum allowable values. Ideally, when blasting is near structures of interest (power towers, dams, houses, etc.), the vibration level (PPV) should be predicted prior to the actual production blasts. There are different techniques to predict the PPV, one in particular is the signature hole technique. This technique is based on signals and systems theory and uses a mathematical operation called convolution to assess the waveform of the production blast. This technique uses both the vibration waveform of an isolated hole and the timing function given by the timing used in the blast. The signature hole technique requires an isolated single-hole waveform to create a prediction. Sometimes this information is difficult to acquire, as it requires the synthesis of a single-hole vibration waveform from a production blast vibration signal. The topic of ground vibrations from mining blasts, and more specifically the synthesis of a single-hole vibration waveform, has been studied by researchers in past decades, but without any concrete success. This lack of success may be partially due to the complexity and difficulty of modelling and calculation. However, this inverse methodology can be very meaningful if successfully applied in blasting engineering. It provides a convenient and economical way to obtain the single-hole vibration waveform and make the prediction of a production blast waveform easier. This dissertation research involves the theories of deconvolution, linear superposition, and Fourier phases to recover single-hole vibration waveforms from a production waveform. Preliminary studies of deconvolution included spectral division deconvolution and Wiener filtering deconvolution. In addition to the adaptation of such methodologies to the blast vibrations problems, the effectiveness of the two deconvolution methods by the influence of delay interval and number of holes is also discussed. Additionally, a new statistical waveform synthesis method based on the theories of linear superposition, properties of Fourier phase, and group delays was developed. The validation of the proposed methodology was also conducted through several field blasting tests. Instead of synthesizing one normalized single-hole vibration waveform by deconvolution, the proposed statistical waveform synthesis methodology generates a different single-hole vibration waveform for each blast hole. This method is more effective and adaptable when synthesizing single-hole vibration waveforms. Recommendations for future work is also provided to improve the methodology and to study other inverse problems of blast vibrations.

An instrumented unmanned aerial vehicle (UAV) was developed and employed to observe the full range of turbulent motions that exist within the inertial subrange of atmospheric surface layer turbulence. The UAV was host to a suite of pressure, temperature, humidity, and wind sensors which provide the necessary data to calculate the variety of turbulent statistics that characterize the flow. Flight experiments were performed with this aircraft, consisting of a large square pattern at an altitude of 100 m above ground level. In order to capture the largest turbulent scales it was necessary to maximize the size of the square pattern. The smallest turbulent scales, on the other hand, were measured through the use of a fast response constant temperature hot wire anemometer. The results demonstrates that the UAV system is capable of directly measuring the full inertial subrange of the atmospheric surface layer with high resolution and allowing for the turbulence dissipation rate to be calculated directly.

Carbon Fiber Reinforced Polymers (CFRP) are a class of advanced materials widely used in versatile applications including aerospace and automotive industries due to their exceptional physical and mechanical properties. Owing to the heterogenous nature of the composites, it is often a challenging task to machine them unlike metals. Drilling in particular, the most commonly used process for component assembly is critical especially in the aerospace sector which demands parts of highest quality and surface integrity. Conventionally, all composites are machined under dry conditions. While there are drawbacks related to dry drilling, for example, poor surface roughness, there is a need to develop processes which yield good quality parts. This thesis investigates the machining performance when drilling CFRP under cryogenic, MQL and hybrid (CryoMQL) modes and comparing with dry drilling in terms of the machining forces, delamination, diameter error and surface integrity assessment including surface roughness, hardness and sub-surface damage analysis. Additionally, the effect of varying the feed rate on the machining performance is examined. From the study, it is concluded that drilling using coolant/ lubricant outperforms dry drilling by producing better quality parts. Also, varying the feed rate proved to be advantageous over drilling at constant feed.