Cette thèse étudie la formation et l'évolution des jets et des vortexdans les atmosphères planétaires turbulentes, à l'aide d'une doubleapproche de simulations numériques et d'expériences delaboratoire. Pour l'approche numérique, un modèle en fluidesshallow-water quasi-géostrophique à deux couches dans le plan betaavec des conditions canal a été utilisé. Comme dans Panetta (1988), onimplémente un cisaillement vertical pour représenter le gradientlatitudinal de température moyenné spatialement, qui est partiellementmaintenu par un forçage thermique. Les instabilités baroclinesaffaiblissent le gradient de température, alors que le forçagethermique le restaure, ce qui crée une dynamique non-linéaire trèsriche.Tout d'abord, nous avons considéré l'écoulement sur un fond plat, etavons modélisé les mouvements convectifs par des paires decyclones/anticyclones ou `hetons' comme dans Thomson (2016). Nousobtenons ainsi des jets principalement baroclines, oscillants entredes phases calmes et des phases turbulentes, où l'écoulement perd sazonalité. Des vortex se forment à partir des jets méandreux etl'énergie zonale diminue alors que l'énergie tourbillonnaireaugmente. Ces phases turbulentes durent typiquement pendant unepériode de relaxation du forçage thermique. On étudie les effets ducisaillement vertical, du forçage thermique et des hetons, enregardant les transferts d'énergie entre les énergies cinétiques etpotentielles, leurs composantes barotropes et baroclines ainsi queleurs composantes zonales et tourbillonnaires. Ceci nous amène àrepenser le paradigme classique des transferts d'énergie présenté dansSalmon (1982). De plus, nous étudions comment une analyse de stabilitélinéaire de l'écoulement zonal instantané est reliée aux phases calmeset turbulentes.Ensuite, nous considérons l'effet d'une topographie de grande échelle,comme une première approche pour comprendre le rôle de la topographiedans la formation des jets et des vortex. Nous utilisons le mêmemodèle que dans la première étude mais nous ajoutons un fondtopographique linéaire méridionalement, qui a l'avantage de dépendred'un seul paramètre, la pente. Une pente négative approfondit lacouche inférieure par rapport à un fond plat, ce qui augmente lepotentiel des instabilités baroclines, alors qu'une pente positive aun effet stabilisateur. Nous supprimons le forçage par les hetons etperturbons l'écoulement grâce à une zone de Rossby de faibleamplitude dans la couche inférieure à l'instant initial. L'effetprincipal du forçage par les hetons est d'agir comme une sorted'amortissement : les fluctuations de l'énergie sont constamment plusextrêmes que sans forçage. Une analyse de stabilité linéaire esteffectuée afin de déterminer les zones de stabilité etd'instabilité.Pour l'étude expérimentale, nous utilisons une cuve tournanteremplie par deux couches de fluides avec une stratification au sel etun couvercle rigide en rotation différentielle. Nous étudions unfront barocliniquement instable dans le régime des vacillationsd'amplitude, qui est caractérisé par l'émergence et ladisparition de vortex de grande échelle. L'analyse de deuxexpériences à la limite de la géostrophie, avec des nombres deRossby de Ro=0.4 et Ro=0.6, montre des comportement trèsdifférents. Pour un faible nombre de Rossby, nous observons desdipôles baroclines alors que pour un large nombre de Rossby nousobtenons des vortex barotropes. Nous examinons l'activité des ondesde petite échelle par différentes méthodes qui révèlent laprésence d'ondes d'inertie gravité comme précurseures del'émergence des vortex.Afin de poursuivre nos recherches sur les fronts à l'interface entredeux couches de fluides immiscibles, nous avons développé unenouvelle méthode de détection de la hauteur et de la pente baséesur les lois optiques de la réfraction. Les équations théoriquesassociées sont résolues numériquement et validées à l'aidede plusieurs situations idéalisées. / This thesis investigates the formation and evolution of jets andvortices in turbulent planetary atmospheres using a dual approach ofhigh-resolution numerical simulations and novel laboratoryexperiments. A two-layer quasi-geostrophic beta-channel shallow watermodel is used for the numerical study. As in Panetta (1988), avertical shear is implemented to represent a spatially-meanlatitudinal temperature gradient, which is partially maintained bythermal damping. Baroclinic instabilities work to erode thetemperature gradient, while thermal damping acts to restore it. Asthe basic state vertical shear is unstable, the thermal damping cannotlead to a full recovery, thus modifying subsequent instabilities andleading to rich nonlinear dynamical behaviour.First, we consider flow over a flat bottom, and model convectivemotions like those thought to occur on Jupiter by pairs ofcyclones/anti-cyclones or `hetons' as in Thomson (2016). We therebyobtain predominantly baroclinic jets, oscillating between quiescentphases, when jets are zonal and the energy is nearly stationary, andturbulent phases, when the flow loses its zonality, vortices pinch offfrom the meandering jets, and zonal energy components drop while eddyenergy components increase. These turbulent phases typically last fora thermal damping relaxation period. The impacts of vertical shear(baroclinicity), thermal damping and heton forcing are comprehensivelyinvestigated by considering the energy transfers occurring betweenkinetic and potential energy, their barotropic and baroclinic parts aswell as their zonal and eddy parts. This leads to a rethinking of theclassic paradigm of energy transfer presented by Salmon (1982), asthis paradigm is too simplistic to explain the results found.Then, we consider the effect of large-scale bottom topography, as afirst approach to understanding the role of topography in jet andvortex formation. We use the same model as in the first study butinclude a linearly sloping topography which has the advantage of beingcharacterised by a single parameter, the slope. We omit the hetonforcing and instead perturb the flow with a small amplitude Rossbywave initially. The main effect of heton forcing is actually to act asa kind of damping: energy fluctuations are consistently less extremethan when no forcing is used. A linear stability analysis is carriedout to motivate a series of nonlinear simulations investigating theeffect of topography, in particular, differences from the flat bottomcase previously examined. We find that destabilising topography makesthe jets more dynamic.In the experimental part, a two-layer salt-stratified fluid is used ina rotating tank with a differentially rotating lid to generate theshear across the interface. We consider a baroclinically unstablefront in the regime of amplitude vacillation, which is found to becharacterised by the sequential emergence and disappearance of alarge-scale vortex. Analysing two similar experiments at the limit ofgeostrophy, with different Rossby numbers Ro=0.4 and Ro=0.6, showssurprisingly different behaviours, with a baroclinic dipole for small,and a barotropic vortex for the large Rossby number. The small-scalewave activity is explored using different methods, and the resultssuggest small, spontaneously-arising inertia-gravity waves precedingthe emergence of the vortex which stirs the interface, thus having animpact on the mixing between the two layers. The recovery period ofthe amplitude vacillation, as well as the intensity of the vortex,increases with the Rossby number.For further research on fronts at two-layer immiscible interfaces, avery accurate novel optical method has been developed to detect theheight and slope, based on the refractive laws of optics. Theassociated theoretical equations are solved numerically and validatedin various idealised situations.

Estuaries depend on the transport of nutrients and sediments from the open sea to help maintain a prosperous environment. One of the major transport mechanisms is the propagation of large two dimensional vortical structures. At the mouth of an inlet, tidal flow forces the formation of two dimensional vortical structures whose lateral extent is much greater than the water depth. After the starting jet vortex dipole detaches from the inlet, secondary vortices shed due to separation from the inlet boundary and eventually reach the starting-jet dipole. An idealized inlet con figuration was utilized for laboratory experiments detailing the formation and propagation of the vortex structures with water depths of 3, 5, and 9 centimeters and flow Froude scaled to inlets along the Texas coast. Using surface particle image velocimetry, the entrainment of the secondary structures into the vortex system are shown as well as variations in characteristics such as trajectory, size, vorticity, and circulation for the vortices as they move downstream.

Optical spatial solitons in nematic liquid crystals, termed nematicons, have become an excellent test bed for nonlinear optics, ranging from fundamental effects to potential uses, such as designing and demonstrating all-optical switching and routing circuits in reconfigurable settings and guided-wave formats. Following their demonstration in planar voltage-assisted nematic liquid crystal cells, the spatial routing of nematicons and associated waveguides have been successfully pursued by exploiting birefringent walkoff, interactions between solitons, electro-optic controlling, lensing effects, boundary effects, solitons in twisted arrangements, refraction and total internal reflection and dark solitons. Refraction and total internal reflection, relying on an interface between two dielectric regions in nematic liquid crystals, provides the most striking results in terms of angular steering. In this thesis, the refraction and total internal reflection of self-trapped optical beams in nematic liquid crystals in the case of a planar cell with two separate regions defined by independently applied bias voltages have been investigated with the aim of achieving a broader understanding of the nematicons and their control. The study of the refraction of nematicons is then extended to the equivalent refraction of optical vortices. The equations governing nonlinear optical beam propagation in nematic liquid crystals are a system consisting of a nonlinear Schr¨odinger-type equation for the optical beam and an elliptic Poisson equation for the medium response. This system of equations has no exact solitary wave solution or any other exact solutions. Although numerical solutions of the governing equations can be found, it has been found that modulation theories give insight into the mechanisms behind nonlinear optical beam evolution, while giving approximate solutions in good to excellent agreement with full numerical solutions and experimental results. The modulation theory reduces the infinite-dimensional partial differential equation problem to a finite dynamical system of comparatively simple ordinary differential equations which are, then easily solved numerically. The modulation theory results on the refraction and total internal reflection of nematicons are in excellent agreement with experimental data and numerical simulations, even when accounting for the birefringent walkoff. The modulation theory also gives excellent results for the refraction of optical vortices of +1 topological charge. The modulation theory predicts that the vortices can become unstable on interaction with the nematic interface, which is verified in quantitative detail by full numerical solutions. This prediction of their azimuthal instability and their break-up into bright beams still awaits an experimental demonstration, but the previously obtained agreement of modulation theory models with the behaviour of actual nematicons leads us to expect the forthcoming observation of the predicted effects with vortices as well.

No description available.

This dissertation work is a study of positronium formation for positron-hydrogen and positron-helium collisions in the Ore gap (the energy region between the threshold for ground-state positronium formation and the first excitation level of the target atom) using variational K-matrices. We have fitted the K-matrices using multichannel effective range theories and using polynomials. Using the variational K-matrices and their fits, we have located zeros in the positronium-formation scattering amplitude and corresponding deep minima in the positronium-formation differential cross section. The zeros are related to the vortices in the extended velocity field associated with the positronium-formation scattering amplitude. For positron-hydrogen collisions, we have found two zeros in the positronium-formation scattering amplitude, and corresponding deep minima in the positronium-formation differential cross section, while we have obtained a zero in the positronium-formation scattering amplitude for positron-helium collisions. We have connected the zeros in the positronium-formation scattering amplitude to vortices in the extended velocity fields. Our work shows that vortices can occur for charge exchange in atomic collisions.

Thesis (M.S.)--Missouri University of Science and Technology, 2009. / Vita. The entire thesis text is included in file. Title from title screen of thesis/dissertation PDF file (viewed February 18, 2009) Includes bibliographical references (p. 54-56).

Thesis (Ph. D.)--Worcester Polytechnic Institute. / Keywords: vortex; coherence. Includes bibliographical references (p. 127-151).

Thesis (Ph. D.)--Worcester Polytechnic Institute. / Keywords: singularity; vortex; phase; diffraction; interference; nulling; singularities; coherence; dislocation; optical vortex. Includes bibliographical references (p. 123-146).

O estudo do escoamento ao redor de corpos submersos tem sido foco de muitas pesquisas. Escoamentos ao redor de cilindros constituem sistemas de grande interesse prático em vários campos da engenharia e outras ciências. Neste trabalho procurou-se estudar a dinâmica do escoamento ao redor de um cilindro circular submetido a um movimento forçado analisando as alterações ocorridas na formação da esteira de vórtices. Para a realização das simulações foi utilizada a técnica de Simulação Numérica Direta (DNS), com um esquema de diferenças finitas compacto de sexta ordem, em associação com um método de representação virtual das fronteiras do cilindro. Quatro tipos de trajetória foram estudadas: uma trajetória com oscilação transversal, onde o cilindro é forçado a mover-se na direção transversal em relação ao escoamento, uma trajetória elíptica, uma trajetória em arco e uma trajetória em oito. Buscou-se identificar as principais características destes escoamentos, como os modos e freqüências de desprendimento de vórtices na esteira formada, coeficientes hidrodinâmicos e parâmetros estatísticos pertinentes ao fenômeno estudado. As simulações foram realizadas com diferentes amplitudes longitudinais de movimento sendo que a freqüência de movimento do cilindro foi mantida constante para as trajetórias elíptica, em arco e em oito. Os resultados para o caso do cilindro com oscilação transversal representaram bem os fenômenos físicos ocorridos neste tipo de escoamento. Foi possível identificar as fronteiras da região de lock-in primário através das variações bruscas nos coeficientes hidrodinâmicos e do ângulo de fase entre o deslocamento do cilindro e a força de sustentação. No caso das simulações com trajetória elíptica e número de Reynolds Re = 140, alterações significativas foram identificadas na esteira de vórtices à medida que a amplitude longitudinal do movimento era modificada. Nas simulações com Re = 400 e trajetória transversal, notou-se a formação de uma esteira com dois vórtices e duas camadas cisalhantes. Comparando esta simulação com os resultados das simulações em trajetória elíptica, foram observadas alterações no padrão de desprendimento, sendo formado um par de vórtices na região superior da esteira e um vórtice simples na região inferior (padrão assimétrico P+S). Comparando as simulações com trajetória elíptica e sentidos de movimento opostos, notou-se que o escoamento se inverte, evidenciando um efeito de memória deste escoamento que é função do sentido adotado para a trajetória. Tanto a amplitude de deslocamento longitudinal quanto o tipo de trajetória imposta ao cilindro influenciam significativamente os padrões da esteira de vórtices formada. / The study of flow around submerged bodies has been the focus of many researches. Flow around cylinders are systems of great practical interest in many fields of engineering and other sciences. This work aims to study the flow dynamics around a circular cylinder under a forced movement by analysing changes in the patterns of the wake. The simulations were performed through the Direct Numerical Simulation technique (DNS), with a sixth-order compact finite differences scheme associated to a virtual boundary method to represent the cylinder surface. Four trajectories were studied: a trajectory with transversal oscillation; a trajectory on elliptical curve; a trajectory in arc curve and a trajectory in eight curve. The main features of these flows were identified, e.g. modes and vortex shedding frequencies, hydrodynamic coefficients and statistical parameters. The simulations were performed with different streamwise movement amplitude, where the movement frequency was held constant for the elliptical, arc and eight trajectories. The results for the cylinder in transversal oscillation successfully represented the physical phenomena. It was possible to identify the boundaries of the primary lock-in region through jumps in the hydrodynamical coefficients and the phase angle between the cylinder displacement and the lift force. For the simulations with elliptical trajectory and Reynolds number Re = 140, significant changes were identified in the wake as the streamwise displacement of the cylinder was modified. In simulations with Re = 400 and transversal trajectory, two vortex and two shear layers could be observed. Comparing this simulation with simulations in elliptical trajectory, changes in the vortex shedding pattern were observed. A vortex pair was formed in the top of the wake and a single vortex in the lower wake ( asymmetrical mode P+S). Comparing the simulations with elliptical trajectory and opposite directions of movement, it was noted that the flow is reversed showing some effect of memory due to the direction of the trajectory. Both the amplitude of streamwise displacement and the trajectory type may result in different patterns in the wake.

O estudo do escoamento ao redor de corpos submersos tem sido foco atual de muitas pesquisas. Em especial, escoamentos ao redor de obstáculos rombudos constituem sistemas de grande interesse prático em vários campos da engenharia e outras ciências como, por exemplo, na engenharia automobilística, na concepção de edificações submetidas à ação do vento, no estudo de correntes fluviais e marítimas sobre pilares de pontes, em risers, cabos de ancoragem e plataformas de captação de petróleo. O presente trabalho teve como objetivo principal o estudo por Simulação Numérica Direta do escoamento ao redor de um cilindro com dois graus de liberdade, submetido a Vibração Induzida por Vórtices (VIV), buscando o esclarecimento a cerca dos fenômenos tridimensionais de formação e evolução do escoamento em função da interação fluido-cilindro. As equações de Navier-Stokes e da continuidade são resolvidas numericamente usando esquemas de diferenças finitas compacto de sexta ordem na discretização espacial, em associação com um método de representação das fronteiras do cilindro, denominado Método de Fronteiras Imersas. Inicialmente foram realizadas simulações para os casos de cilindro fixo e em movimento forçado, onde os resultados mostraram que o código computacional consegue modelar satisfatoriamente os fenômenos físicos de acordo com os resultados encontrados na literatura. Para as simulações com cilindro fixo, em movimento forçado e em VIV, nota-se que as estruturas do escoamento tridimensional apresentam modos com comprimentos de onda alterados em função do movimento imposto. Os resultados das simulações em VIV com dois graus de liberdade, para Re = 1250 mostraram que as estruturas tridimensionais do escoamento sofrem significativas alterações espaciais e temporais em função dos parâmetros do modelo dinâmico. Foram identificados diferentes fenômenos de formação e emparelhamento dos vórtices que são modificados em função das condições impostas ao modelo massa-mola amortecimento. / The study of flow around submerged bodies has been the focus of many researches. In special, the flow around bluff bodies are systems of great practical interest in many fields of engineering and other sciences, e.g., the flow around vehicles, buildings submitted to the action of the wind, river and sea currents over pillars of the bridges, risers and cables in the oil platforms. The present work aims to study through Direct Numerical Simulation (DNS) the flow dynamics around a circular cylinder under Vortex-Induced Vibration with two degrees of freedom, searching for identify the changes in the three-dimensional structures of the street wake imposed by the interactions between the flow and the immersed cylinder. The simulations were performed using a sixth-order compact finite differences scheme associated to a Immersed Boundary Method (IBM) to represent the cylinder surface. Firstly, simulations of the flow around fixed and forced movement were accomplished. The computational code Incompact3d presented results in agreement with the literature. The validation of the code shown that is possible to obtain satisfactory results relate to the street wake behind cylinder and the force coefficients are in agreement with others experimental and numerical works. In case of free oscillating cylinder, several changes in the three-dimensional vortex shedding modes are observed as the parameters of the oscillating dynamic model was changed. It was observed that different three-dimensional structures are formed in the wake depending on the initial conditions of the mass-stiffness damping model. The results of the simulations at Reynolds numbers equal to 1250 presenting some hydrodynamic phenomena as paring and turbulent structures in the flow that are modifying during the time and in the computational domain.