Les rendements des produits de fission font partie des données nucléaires sur lesquellesreposent les simulations neutroniques. L’objectif de cette thèse est d’apporter de nouvellesmesures de rendements de fission de deux noyaux fissiles : le 241Pu et l’233U. Ces noyauxappartiennent respectivement au cycle du combustible de l’uranium et à celui du thorium.Ces mesures ont été réalisées auprès du spectromètre de masse Lohengrin de l’InstitutLaue Langevin (ILL) à Grenoble. Le spectromètre est combiné avec une chambre d’ionisationpour mesurer les rendements en masse de l’233U et du 241Pu et avec un dispositif despectrométrie gamma pour déterminer les rendements isotopiques de l’233U.Une nouvelle procédure d’analyse innovante a été mise en place dans le but de maîtriserles systématiques et de réduire les biais expérimentaux. La matrice de variance-covarianceassociée à nos mesures de rendements a ainsi pu être calculée pour la première fois. / Fission product yields are significant nuclear data for neutronic simulations. The purposeof this work is to improve fission yield knowledge for two fissile nuclei : 241Pu and 233U. Thoseare respectively involved in the uranium and thorium nuclear fuel cycle.The measurements are performed at the Lohengrin mass spectrometer of the InstitutLaue-Langevin (ILL) located in Grenoble. The spectrometer is combined with an ionizationchamber to measure mass yields of 241Pu and 233U and with a gamma spectrometry set-upto determine isotopic yields of 233U.A new analysis method of experimental data has been developed in order to controlsystematics and to reduce experimental biases. For the first time, the experimental variancecovariancematrix of our measured fission yields could be deduced.

Os mecanismos de transposição de peixes (MTP) são estruturas ou sistemas que possibilitam a migração da ictiofauna entre as partes de jusante/montante/jusante de uma barragem. As escadas para peixes representam um dos tipos de MTP mais conhecidos no mundo e apresentam diversas configurações geométricas. A escolha do tipo de escada deve atender às características natatórias dos peixes selecionados para transporem o obstáculo. Para algumas espécies, como o salmão, já se conhecem geometrias adequadas, entretanto, para a maioria das espécies isto não ocorre e muitos projetos têm demonstrado desempenho insatisfatório. No Brasil, encontra-se uma imensa diversidade de espécies de peixes, cujas características natatórias diferem em muito das apresentadas pelos salmonídeos. Este fato, associado à crescente exigência da implantação de MTP nos barramentos, através de leis estaduais ambientais, torna necessária a definição de estruturas adequadas à ictiofauna brasileira. A validação dos critérios de projeto passa, obrigatoriamente, por estudos que avaliem as características hidráulicas das estruturas propostas e a interação do fluxo com os padrões natatórios da ictiofauna. O número de pesquisas relacionadas ao funcionamento hidráulico de escadas para peixes vem crescendo, entretanto ainda são insuficientes, não existindo um consenso sobre os critérios, seja para sua caracterização completa, seja para definir sobre quais parâmetros devem ser considerados. Os padrões de turbulência do escoamento em escadas para peixes, cujas características supõem-se relacionarem-se com o grau de aceitação ou rejeição das espécies, são praticamente desconhecidos. Neste trabalho realizou-se a caracterização hidráulica através do estudo experimental, de três tipos de escadas para peixes: (1) com ranhura vertical; (2) com descarregador de superfície e (3) com orifício de fundo. As estruturas foram construídas nos laboratórios do Instituto de Pesquisas Hidráulicas da Universidade Federal do Rio Grande do Sul – IPH/UFRGS e no Laboratório Nacional de Engenharia Civil – LNEC – Portugal. As seguintes medições foram realizadas: velocidades em três direções, em diferentes planos dos tanques, com velocímetros acústicos Doppler (ADV) e níveis de água da superfície livre com pontas linimétricas e réguas graduadas. Para a estrutura com ranhura vertical ainda foram medidas as pressões médias e suas flutuações, junto ao fundo do canal, com transdutores piezoresistivos. Para cada uma das estruturas realizaram-se ensaios com três descargas. Além de definidas as características médias do escoamento, os dados de velocidades, que passaram por um processo de aplicação de filtros sem substituição, possibilitaram a avaliação de parâmetros de turbulência, entre eles a energia cinética da turbulência, a intensidade da turbulência e as tensões de Reynolds. No modelo da escada para peixes com ranhuras verticais verificou-se que os parâmetros hidráulicos estão de acordo com estruturas similares da bibliografia, entre eles, coeficiente de descarga, vazão adimensional e coeficiente de cisalhamento. Os campos de pressão junto ao fundo refletem o comportamento da superfície livre do escoamento. Através do mapeamento das velocidades dentro do tanque da escada do tipo ranhura vertical, foi possível caracterizar as duas zonas de recirculação e a região do jato principal. Os máximos valores médios de velocidade encontram-se na seção da ranhura, não excedendo 1,00 m/s (no modelo). Além disso, foi possível reconhecer as regiões de maior energia cinética da turbulência que apresentaram valores de até 1000 cm2/s2 na região do jato principal, as quais coincidem com as zonas de maiores tensões de Reynolds da ordem de até 30 N/m2. A partir das velocidades médias e em função das velocidades de nado dos peixes obtidos na literatura, foi possível a identificação de locais que atuam como “barreiras” ao deslocamento de determinada espécie. A avaliação qualitativa do comportamento da trajetória dos peixes dentro dos tanques mostrou-se de acordo com a definição destas “barreiras” hidráulicas e com a avaliação do comportamento dos campos de energia cinética da turbulência e das tensões de Reynolds. No modelo da escada do tipo descarregador de superfície observaram-se as máximas velocidades médias sobre o descarregador, com valores de até 1,73 m/s. Verifica-se que na maior parte do tanque as velocidades médias não ultrapassam 40% da velocidade potencial. Foram encontrados valores de energia cinética da turbulência até 2000 cm2/s2, com valores na maior parte do tanque em torno de 200 cm2/s2. Quanto aos campos de tensões de Reynolds, têm-se, na maior parte do tanque, os valores entre −5 e 5 N/m2, sendo que na região do jato mergulhante, os valores chegam até 30 N/m2. Na escada com orifícios de fundo verificaram-se as maiores velocidades médias nos planos sob influência do fluxo principal proveniente do orifício. Não foi possível a medição da velocidade na seção da abertura, sendo que os valores medidos no tanque não ultrapassaram 50% da velocidade potencial. Os máximos valores de energia cinética da turbulência atingem até 2000 cm2/s2 junto ao fundo, enquanto na região central do tanque, o valor médio é um pouco inferior a 200 cm2/s2. Os valores de tensão de Reynolds encontram-se entre −30 e 30 N/m2, com a maioria das regiões entre −5 e 5 N/m2. Os valores máximos e médios de energia cinética da turbulência e tensões de Reynolds para os modelos com descarregador de superfície e com orifício de fundo encontram-se na mesma faixa. Isto indica que, sob o ponto de vista técnico, possivelmente o critério de escolha entre essas duas estruturas recai nas características da ictiofauna. A passagem com ranhura vertical permite a escolha da profundidade preferencial de nado. No entanto, nessa estrutura, verificase que as componentes médias e turbulentas, nas regiões de descanso, comparando-as com os valores máximos do jato principal, são superiores proporcionalmente, às observadas nas zonas de recirculação das outras duas estruturas. As informações biológicas disponíveis na literatura não permitem a definição de condições preferenciais em relação aos parâmetros de turbulência entre as estruturas aqui avaliadas. No entanto, as informações obtidas nesse trabalho indicam que a energia cinética da turbulência e as tensões de Reynolds podem ser indicativos da tolerância ou preferência dos peixes até certos níveis de turbulência. / Fish facilities are structures or systems that enable fish passage through dams or obstructions. Fishways represent one of the most common fish facilities types worldwide, presenting different geometries and designs. The choice and design of these structures must attend the fish swimming performance and biological characteristics. For some species, as salmon, there are defined designs that can be successfully applied, however, that does not occur for the majority of the species and many projects have showed unsatisfactory efficiency. In Brazil, there is an immense diversity of fish species, whose swimming characteristics are strongly different than the other known species like the salmons. This fact, associated with the increasing requirement of fish facilities implantation in dams, mainly through environmental State Laws, requires the definition of structures adapted to the Brazilian fish. The design criteria validation needs, necessarily, studies to evaluate hydraulic characteristics on structure proposals and the interaction of the flow with swimming abilities. The number of researches related to the hydraulic functioning of fishways is increasing, however they are still insufficient and there is not a consensus on the criteria, either for its complete characterization, either to define which parameters should be considered. The flow turbulence patterns in fishways, whose characteristics are assumed to be related with the degree of acceptance or rejection of the species, are practically unknown. In this study a hydraulic characterization was carried out through an experimental study, including three kinds of fishways: (1) with vertical slots; (2) with rectangular notches and (3) with bottom orifices. The facilities were set up in the Hydraulic Research Institute of the Federal University of Rio Grande do Sul (IPH/UFRGS) and in the National Laboratory of Civil Engineering – LNEC - Portugal. The following measurements were carried out: three-direction velocities, in a 3D-mesh in one pool of each structure, with Acoustic Doppler Velocimeters and water level of the free surface with a point gauge and scales. In the vertical slot fishway it was carried out complementary measurements of pressure in the bottom of the channel, with transducers. For each structure three discharges were tested. Besides defining the mean flow characteristics, the velocity time data, filtered (through a digital process) without substitution, allowed to analyze some turbulence parameters, as turbulence kinetic energy, turbulence intensity and Reynolds’ shear stresses. In the vertical slot fishway model it was verified that the hydraulic parameters are in agreement with similar structures of the literature, among them, discharge coefficient, adimensional discharge and friction factor. The bottom pressure field agrees with the behavior of the free-surface flow. Two recirculation zones and the area of the main jet were characterized through the velocities distribution inside the vertical slot fishway pool. The maximum mean velocity values were found in the slot section, not exceeding 1.00 m/s (in the model). Moreover, it was possible to recognize the areas with larger turbulence kinetic energy that presented values of up to 1000 cm²/s² in the main jet area, which correspond to the largest Reynolds’ shear stresses values of up to 30 N/m². Considering mean velocities data and fish swimming capabilities, it was possible to identify regions that are insurmountable by the fish. The qualitative approach of the fish trajectory inside the structure agrees with the insurmountable regions described through mean velocities and with the distribution of turbulence kinetic energy and Reynolds’ stresses. In the fishway model with rectangular notches, the maximum mean velocities were observed on the weir, with values of up to 1.73 m/s. In the major part of the pool, mean velocities do not surpass 40% of the potential velocity. Values up to 2000 cm²/s² for turbulence kinetic energy were found, with values in the major part of the pool close to 200 cm²/s². For Reynolds’ stresses, the major part of the structure works with values in the range of −5 and 5 N/m², and in the jet entrance pool region the values are of up to 30 N/m². In the fishway with bottom orifices the largest mean velocities were verified in the plans under influence of the main flow connecting consecutive orifices. The measurement of the velocities in the orifice section was not possible and the values measured in the pool did not exceed 50% of the potential velocity. The maximum values of turbulence kinetic energy reached up to 2000 cm²/s² close to the bottom channel, while in the central area of the pool, the mean value is lower than 200 cm2/s2. The values of Reynolds’ shear stresses are between −30 and 30 N/m2, with the major part between−5 and 5 N/m2. The maximum and mean values of turbulence kinetic energy and Reynolds’ stresses in the models with rectangular notches and with orifice are in the same range. It indicates that the choice between these two structures relapses in the fish swimming characteristics. The passage with vertical slot allows the choice of the swimming depth preference. However, in the resting areas of this structure, it is verified that the mean and turbulent components when compared with the maximum values of the main jet, are higher proportionally, to the ones observed in the recirculation zones of the other two structures. The biological information available in the literature does not allow the definition of preferential conditions in relation to the turbulence parameters among the structures here appraised. However, the information obtained in this work indicates that the turbulence kinetic energy and Reynolds’ shear stress can be indicatives of the tolerance or preference of the fish to certain turbulence levels.

Round Island is a small (208 ha) islet of volcanic origin located 22.5 km north east of mainland Mauritius and has been classified as a nature reserve since 1957. Two sites were chosen for the installation of environmental monitoring equipment. A series of Gerlach troughs were installed to capture surface sediment transported by runoff, which were used to document sediment yields and determine the particle size distribution. Overall, rainfall and erosivity on Round Island is far less, when compared to mainland Mauritius. However, erosivity from Round Island (2,314.76 MJ.mm.ha-1.h-1.yr-1) is slightly above the global average of 2,190 MJ.mm.ha-1.h-1.yr-1. In terms of sediment transport, the annual sediment movement rates for Round Island were established during this study (0.1248 t.ha-1.yr-1) and were found to be considerably lower than Mauritius (10 t.ha-1.yr-1), as well as other tropical island such as Kauai (0.86 t.ha-1.yr-1) and O’ahu (0.6 t.ha-1.yr-1). Thus, although the estimated rates of soil erosion are very low for humid tropical regions, these rates only reflect the contemporary environmental conditions and cognisance of the landscape history should be incorporated into assessments of soil erosion / Geography / M. Sc. (Geography)

A différents endroits du monde, on observe que la méthode d’exploitation minière par chambres et piliers est associée à deux aléas de grande ampleur se manifestant en surface : les affaissements progressifs et les effondrements brutaux. Le bassin ferrifère lorrain est sujet à ces deux types d’aléas notamment depuis la fermeture massive des mines à partir des années 1960. Les effondrements étant plus dangereux que les affaissements, des chercheurs ont développé des méthodes statistiques pour distinguer les zones à risque d’effondrement de celles à risque d’affaissement. L’objectif de cette thèse est de développer des indicateurs mécaniques capables de faire la même distinction, afin d’améliorer la compréhension physique des phénomènes mis en jeux et de valider les indicateurs statistiques. Nous faisons l’hypothèse que les effondrements et affaissements sont tous deux causés par un endommagement des piliers en profondeur. Le Chapitre 1 est consacré à une étude bibliographique qui synthétise les connaissances relatives au comportement mécanique des piliers. Nous constatons que les piliers sont caractérisés par deux critères : le facteur de sécurité (résistance en compression/contrainte verticale moyenne) qui détermine s’ils sont susceptibles de se rompre, et un critère de stabilité des piliers en cas de rupture qui s’exprime comme la raideur relative des piliers par rapport à celle de la roche encaissante. Deux limites des approches actuelles sont identifiées : 1) la question de la contrainte verticale moyenne des piliers (dénominateur du facteur de sécurité) a été relativement peu étudiée pour les piliers aux dimensions irrégulières, caractéristiques des anciennes mines ; 2) l’instabilité de la rupture des piliers est évaluée par un modèle 1D proposé dans les années 1960-1970, se basant sur des considérations énergétiques. Ce modèle très courant n’a toutefois fait l’objet que de peu de développements dans des configurations géomécaniques plus réalistes que celle du modèle de départ. Le Chapitre 2 développe une nouvelle méthode permettant de prendre en compte l’irrégularité géométrique des piliers pour estimer leur contrainte verticale et donc d’avoir une meilleure précision sur l’estimation du facteur de sécurité. Nous introduisons le concept du taux de défruitement relatif et montrons sur la base de résultats de modélisation numérique, qu’une fonction quadratique de ce terme procure une meilleure estimation de la contrainte verticale moyenne que la méthode traditionnelle de l’aire tributaire. Le Chapitre 3 porte sur l’application d’une méthode originale de modélisation explicite de l’instabilité des piliers, basée l’énergie cinétique. Nous montrons que nos résultats de modélisation en 2D sont compatibles avec le modèle 1D traditionnel. La procédure de modélisation employée permet également de localiser et de quantifier les excès d’énergie cinétique lors de l’instabilité. Les quantités d’énergie cinétique modélisées sont du même ordre de grandeur que les magnitudes de Richter d’énergie sismique typiques des ruptures brutales de piliers réels. Nous traitons dans le Chapitre 4 la problématique des effondrements du bassin ferrifère lorrain. Nous montrons sur la base de modélisations 3D que les critères de rupture et d’instabilité des piliers permettent de bien y distinguer les zones affaissées et les zones effondrées. Les secteurs affaissés et effondrés sont caractérisés par un facteur de sécurité moyen des piliers inférieur à 1 du point de vue de leur limite d’élasticité. Cependant, les secteurs effondrés se distinguent des secteurs affaissés par un facteur de sécurité moyen des piliers également inférieur à 1 du point de vue de leur pic de résistance. Ceci est confirmé par des modèles numériques simples en 2D, suggérant par ailleurs que les caractéristiques du recouvrement ont peu d’effet sur l’occurrence des effondrements. Certaines études fournissant des conclusions contraires, nous avançons des arguments pour en discuter / At different locations around the world, the room-and-pillar mining method truns out to be associated with two major types of hazards affecting the surface: the continuous subsidence and the sudden collapse. The Lorraine iron basin (France) have known on many occasions both continuous subsidences and sudden collapses, notably since the decline of the French mining activity in the 1960’s. The collapses are more dangerous than the continuous subsidences. This is why researchers and engineers developed statistical methods capable of distinguishing between the sectors subjected to a collapse hazard and those subjected to a continuous subsidence hazard. The objective of that study if to develop indicators capable of making the same distinction but on the basis of mechanical criteria, for the purpose of improving the understanding of the collapse mechanism and for validating the statistical indicators. The main assumption of this work is that both the continuous subsidences and the collapses are caused by damage of the pillars within the mines. Chapter 1 presents a literature review concerning the actual knowledge about the pillars mechanical behaviour. We notice that pillars are traditionally characterized by two criteria: the factor of safety (pic compressive resistance/average vertical stress), which determines whether or not pillars are susceptible to fail, and a criterion of stability in case of a failure, which consists of comparing the flexural stiffness of the surrounding rock to the pillars axial stiffness in the post-pic domain of their behaviour. We identify two limits regarding the traditional approaches: i) The question of the average vertical stress has only been little studied in the case of pillars with irregular geometry, which is often the case in old mines; ii) The model allowing to estimate whether or not an instability is susceptible to occur comes from a 1D model proposed in the 1960’s and that has never been developed furthermore for being applied to real configurations in various geomechanical conditions. In Chapter 2, we develop a new method for taking the irregularity of the pillars dimensions into account in estimating their average vertical stress, for the purpose of having a better precision in the factor of safety calculation. We introduce the concept of relative extraction ratio and show, on the basis on numerical modelling, that a quadratic function of this ratio is more precise in estimating the average pillar stress than the classical tributary area method.Chapter 3 is devoted to the development and the application of an original explicit method for modelling pillar instability based on the kinetic energy. We show that the numerical results obtained in 2D configurations are compatible with the traditional 1D model of pillar instability. Moreover, we show how the modelling method we are using helps to locate and to quantify the kinetic energy in excess induced by the instability. The magnitudes of modelled kinetic energies are similar to Richter magnitudes of typical seismic energies recorded during real pillar bursts. The problem of the Lorraine iron basin collapses is treated in Chapter 4. Based on 3D modelling results, we show that pillar failure and instability criteria are capable of well distinguishing between the sectors affected by collapses and those affected by continuous subsidences in the iron basin. Both the sectors subjected to collapses and continuous subsidences are characterized by a factor of safety lower than one regarding the limit of elasticity. But the collapsed sectors are distinguished by a factor of safety also lower than one regarding the pic compressive resistance. This observation is confirmed by simplified 2D modelling, which suggests, furthermore, that the properties of the overburden have no effect on the occurrence of the collapses. Some studies provide contrary conclusions. So, we propose some points to be discussed

Os mecanismos de transposição de peixes (MTP) são estruturas ou sistemas que possibilitam a migração da ictiofauna entre as partes de jusante/montante/jusante de uma barragem. As escadas para peixes representam um dos tipos de MTP mais conhecidos no mundo e apresentam diversas configurações geométricas. A escolha do tipo de escada deve atender às características natatórias dos peixes selecionados para transporem o obstáculo. Para algumas espécies, como o salmão, já se conhecem geometrias adequadas, entretanto, para a maioria das espécies isto não ocorre e muitos projetos têm demonstrado desempenho insatisfatório. No Brasil, encontra-se uma imensa diversidade de espécies de peixes, cujas características natatórias diferem em muito das apresentadas pelos salmonídeos. Este fato, associado à crescente exigência da implantação de MTP nos barramentos, através de leis estaduais ambientais, torna necessária a definição de estruturas adequadas à ictiofauna brasileira. A validação dos critérios de projeto passa, obrigatoriamente, por estudos que avaliem as características hidráulicas das estruturas propostas e a interação do fluxo com os padrões natatórios da ictiofauna. O número de pesquisas relacionadas ao funcionamento hidráulico de escadas para peixes vem crescendo, entretanto ainda são insuficientes, não existindo um consenso sobre os critérios, seja para sua caracterização completa, seja para definir sobre quais parâmetros devem ser considerados. Os padrões de turbulência do escoamento em escadas para peixes, cujas características supõem-se relacionarem-se com o grau de aceitação ou rejeição das espécies, são praticamente desconhecidos. Neste trabalho realizou-se a caracterização hidráulica através do estudo experimental, de três tipos de escadas para peixes: (1) com ranhura vertical; (2) com descarregador de superfície e (3) com orifício de fundo. As estruturas foram construídas nos laboratórios do Instituto de Pesquisas Hidráulicas da Universidade Federal do Rio Grande do Sul – IPH/UFRGS e no Laboratório Nacional de Engenharia Civil – LNEC – Portugal. As seguintes medições foram realizadas: velocidades em três direções, em diferentes planos dos tanques, com velocímetros acústicos Doppler (ADV) e níveis de água da superfície livre com pontas linimétricas e réguas graduadas. Para a estrutura com ranhura vertical ainda foram medidas as pressões médias e suas flutuações, junto ao fundo do canal, com transdutores piezoresistivos. Para cada uma das estruturas realizaram-se ensaios com três descargas. Além de definidas as características médias do escoamento, os dados de velocidades, que passaram por um processo de aplicação de filtros sem substituição, possibilitaram a avaliação de parâmetros de turbulência, entre eles a energia cinética da turbulência, a intensidade da turbulência e as tensões de Reynolds. No modelo da escada para peixes com ranhuras verticais verificou-se que os parâmetros hidráulicos estão de acordo com estruturas similares da bibliografia, entre eles, coeficiente de descarga, vazão adimensional e coeficiente de cisalhamento. Os campos de pressão junto ao fundo refletem o comportamento da superfície livre do escoamento. Através do mapeamento das velocidades dentro do tanque da escada do tipo ranhura vertical, foi possível caracterizar as duas zonas de recirculação e a região do jato principal. Os máximos valores médios de velocidade encontram-se na seção da ranhura, não excedendo 1,00 m/s (no modelo). Além disso, foi possível reconhecer as regiões de maior energia cinética da turbulência que apresentaram valores de até 1000 cm2/s2 na região do jato principal, as quais coincidem com as zonas de maiores tensões de Reynolds da ordem de até 30 N/m2. A partir das velocidades médias e em função das velocidades de nado dos peixes obtidos na literatura, foi possível a identificação de locais que atuam como “barreiras” ao deslocamento de determinada espécie. A avaliação qualitativa do comportamento da trajetória dos peixes dentro dos tanques mostrou-se de acordo com a definição destas “barreiras” hidráulicas e com a avaliação do comportamento dos campos de energia cinética da turbulência e das tensões de Reynolds. No modelo da escada do tipo descarregador de superfície observaram-se as máximas velocidades médias sobre o descarregador, com valores de até 1,73 m/s. Verifica-se que na maior parte do tanque as velocidades médias não ultrapassam 40% da velocidade potencial. Foram encontrados valores de energia cinética da turbulência até 2000 cm2/s2, com valores na maior parte do tanque em torno de 200 cm2/s2. Quanto aos campos de tensões de Reynolds, têm-se, na maior parte do tanque, os valores entre −5 e 5 N/m2, sendo que na região do jato mergulhante, os valores chegam até 30 N/m2. Na escada com orifícios de fundo verificaram-se as maiores velocidades médias nos planos sob influência do fluxo principal proveniente do orifício. Não foi possível a medição da velocidade na seção da abertura, sendo que os valores medidos no tanque não ultrapassaram 50% da velocidade potencial. Os máximos valores de energia cinética da turbulência atingem até 2000 cm2/s2 junto ao fundo, enquanto na região central do tanque, o valor médio é um pouco inferior a 200 cm2/s2. Os valores de tensão de Reynolds encontram-se entre −30 e 30 N/m2, com a maioria das regiões entre −5 e 5 N/m2. Os valores máximos e médios de energia cinética da turbulência e tensões de Reynolds para os modelos com descarregador de superfície e com orifício de fundo encontram-se na mesma faixa. Isto indica que, sob o ponto de vista técnico, possivelmente o critério de escolha entre essas duas estruturas recai nas características da ictiofauna. A passagem com ranhura vertical permite a escolha da profundidade preferencial de nado. No entanto, nessa estrutura, verificase que as componentes médias e turbulentas, nas regiões de descanso, comparando-as com os valores máximos do jato principal, são superiores proporcionalmente, às observadas nas zonas de recirculação das outras duas estruturas. As informações biológicas disponíveis na literatura não permitem a definição de condições preferenciais em relação aos parâmetros de turbulência entre as estruturas aqui avaliadas. No entanto, as informações obtidas nesse trabalho indicam que a energia cinética da turbulência e as tensões de Reynolds podem ser indicativos da tolerância ou preferência dos peixes até certos níveis de turbulência. / Fish facilities are structures or systems that enable fish passage through dams or obstructions. Fishways represent one of the most common fish facilities types worldwide, presenting different geometries and designs. The choice and design of these structures must attend the fish swimming performance and biological characteristics. For some species, as salmon, there are defined designs that can be successfully applied, however, that does not occur for the majority of the species and many projects have showed unsatisfactory efficiency. In Brazil, there is an immense diversity of fish species, whose swimming characteristics are strongly different than the other known species like the salmons. This fact, associated with the increasing requirement of fish facilities implantation in dams, mainly through environmental State Laws, requires the definition of structures adapted to the Brazilian fish. The design criteria validation needs, necessarily, studies to evaluate hydraulic characteristics on structure proposals and the interaction of the flow with swimming abilities. The number of researches related to the hydraulic functioning of fishways is increasing, however they are still insufficient and there is not a consensus on the criteria, either for its complete characterization, either to define which parameters should be considered. The flow turbulence patterns in fishways, whose characteristics are assumed to be related with the degree of acceptance or rejection of the species, are practically unknown. In this study a hydraulic characterization was carried out through an experimental study, including three kinds of fishways: (1) with vertical slots; (2) with rectangular notches and (3) with bottom orifices. The facilities were set up in the Hydraulic Research Institute of the Federal University of Rio Grande do Sul (IPH/UFRGS) and in the National Laboratory of Civil Engineering – LNEC - Portugal. The following measurements were carried out: three-direction velocities, in a 3D-mesh in one pool of each structure, with Acoustic Doppler Velocimeters and water level of the free surface with a point gauge and scales. In the vertical slot fishway it was carried out complementary measurements of pressure in the bottom of the channel, with transducers. For each structure three discharges were tested. Besides defining the mean flow characteristics, the velocity time data, filtered (through a digital process) without substitution, allowed to analyze some turbulence parameters, as turbulence kinetic energy, turbulence intensity and Reynolds’ shear stresses. In the vertical slot fishway model it was verified that the hydraulic parameters are in agreement with similar structures of the literature, among them, discharge coefficient, adimensional discharge and friction factor. The bottom pressure field agrees with the behavior of the free-surface flow. Two recirculation zones and the area of the main jet were characterized through the velocities distribution inside the vertical slot fishway pool. The maximum mean velocity values were found in the slot section, not exceeding 1.00 m/s (in the model). Moreover, it was possible to recognize the areas with larger turbulence kinetic energy that presented values of up to 1000 cm²/s² in the main jet area, which correspond to the largest Reynolds’ shear stresses values of up to 30 N/m². Considering mean velocities data and fish swimming capabilities, it was possible to identify regions that are insurmountable by the fish. The qualitative approach of the fish trajectory inside the structure agrees with the insurmountable regions described through mean velocities and with the distribution of turbulence kinetic energy and Reynolds’ stresses. In the fishway model with rectangular notches, the maximum mean velocities were observed on the weir, with values of up to 1.73 m/s. In the major part of the pool, mean velocities do not surpass 40% of the potential velocity. Values up to 2000 cm²/s² for turbulence kinetic energy were found, with values in the major part of the pool close to 200 cm²/s². For Reynolds’ stresses, the major part of the structure works with values in the range of −5 and 5 N/m², and in the jet entrance pool region the values are of up to 30 N/m². In the fishway with bottom orifices the largest mean velocities were verified in the plans under influence of the main flow connecting consecutive orifices. The measurement of the velocities in the orifice section was not possible and the values measured in the pool did not exceed 50% of the potential velocity. The maximum values of turbulence kinetic energy reached up to 2000 cm²/s² close to the bottom channel, while in the central area of the pool, the mean value is lower than 200 cm2/s2. The values of Reynolds’ shear stresses are between −30 and 30 N/m2, with the major part between−5 and 5 N/m2. The maximum and mean values of turbulence kinetic energy and Reynolds’ stresses in the models with rectangular notches and with orifice are in the same range. It indicates that the choice between these two structures relapses in the fish swimming characteristics. The passage with vertical slot allows the choice of the swimming depth preference. However, in the resting areas of this structure, it is verified that the mean and turbulent components when compared with the maximum values of the main jet, are higher proportionally, to the ones observed in the recirculation zones of the other two structures. The biological information available in the literature does not allow the definition of preferential conditions in relation to the turbulence parameters among the structures here appraised. However, the information obtained in this work indicates that the turbulence kinetic energy and Reynolds’ shear stress can be indicatives of the tolerance or preference of the fish to certain turbulence levels.

abstract: This thesis focuses on an improved understanding of the dynamics at different length scales of wind farms in an atmospheric boundary layer (ABL) using a series of visualization studies and Fourier, wavelet based spectral analysis using high fidelity large eddy simulation (LES). For this purpose, a robust LES based neutral ABL model at very high Reynolds number has been developed using a high order spectral element method which has been validated against the previous literature. This ABL methodology has been used as a building block to drive large wind turbine arrays or wind farms residing inside the boundary layer as documented in the subsequent work. Studies conducted in the thesis involving massive periodic wind farms with neutral ABL have indicated towards the presence of large scale coherent structures that contribute to the power generated by the wind turbines via downdraft mechanisms which are also responsible for the modulation of near wall dynamics. This key idea about the modulation of large scales have seen a lot of promise in the application of flow past vertically staggered wind farms with turbines at different scales. Eventually, studies involving wind farms have been progressively evolved in a framework of inflow-outflow where the turbulent inflow is being fed from the precursor ABL using a spectral interpolation technique. This methodology has been used to enhance the understanding related to the multiscale physics of wind farm ABL interaction, where phenomenon like the growth of the inner layer, and wake impingement effects in the subsequent rows of wind turbines are important owing to the streamwise heterogeneity of the flow. Finally, the presence of realistic geophysical effects in the turbulent inflow have been investigated that influence the flow past the wind turbine arrays. Some of the geophysical effects that have been considered include the presence of the Coriolis forces as well as the temporal variation of mean wind magnitude and direction that might occur due to mesoscale dynamics. This study has been compared against field experimental results which provides an important step towards understanding the capability of the mean data driven LES methodology in predicting realistic flow structures. / Dissertation/Thesis / Turbulent flow past a large periodic wind farm / Doctoral Dissertation Mechanical Engineering 2018

Os mecanismos de transposição de peixes (MTP) são estruturas ou sistemas que possibilitam a migração da ictiofauna entre as partes de jusante/montante/jusante de uma barragem. As escadas para peixes representam um dos tipos de MTP mais conhecidos no mundo e apresentam diversas configurações geométricas. A escolha do tipo de escada deve atender às características natatórias dos peixes selecionados para transporem o obstáculo. Para algumas espécies, como o salmão, já se conhecem geometrias adequadas, entretanto, para a maioria das espécies isto não ocorre e muitos projetos têm demonstrado desempenho insatisfatório. No Brasil, encontra-se uma imensa diversidade de espécies de peixes, cujas características natatórias diferem em muito das apresentadas pelos salmonídeos. Este fato, associado à crescente exigência da implantação de MTP nos barramentos, através de leis estaduais ambientais, torna necessária a definição de estruturas adequadas à ictiofauna brasileira. A validação dos critérios de projeto passa, obrigatoriamente, por estudos que avaliem as características hidráulicas das estruturas propostas e a interação do fluxo com os padrões natatórios da ictiofauna. O número de pesquisas relacionadas ao funcionamento hidráulico de escadas para peixes vem crescendo, entretanto ainda são insuficientes, não existindo um consenso sobre os critérios, seja para sua caracterização completa, seja para definir sobre quais parâmetros devem ser considerados. Os padrões de turbulência do escoamento em escadas para peixes, cujas características supõem-se relacionarem-se com o grau de aceitação ou rejeição das espécies, são praticamente desconhecidos. Neste trabalho realizou-se a caracterização hidráulica através do estudo experimental, de três tipos de escadas para peixes: (1) com ranhura vertical; (2) com descarregador de superfície e (3) com orifício de fundo. As estruturas foram construídas nos laboratórios do Instituto de Pesquisas Hidráulicas da Universidade Federal do Rio Grande do Sul – IPH/UFRGS e no Laboratório Nacional de Engenharia Civil – LNEC – Portugal. As seguintes medições foram realizadas: velocidades em três direções, em diferentes planos dos tanques, com velocímetros acústicos Doppler (ADV) e níveis de água da superfície livre com pontas linimétricas e réguas graduadas. Para a estrutura com ranhura vertical ainda foram medidas as pressões médias e suas flutuações, junto ao fundo do canal, com transdutores piezoresistivos. Para cada uma das estruturas realizaram-se ensaios com três descargas. Além de definidas as características médias do escoamento, os dados de velocidades, que passaram por um processo de aplicação de filtros sem substituição, possibilitaram a avaliação de parâmetros de turbulência, entre eles a energia cinética da turbulência, a intensidade da turbulência e as tensões de Reynolds. No modelo da escada para peixes com ranhuras verticais verificou-se que os parâmetros hidráulicos estão de acordo com estruturas similares da bibliografia, entre eles, coeficiente de descarga, vazão adimensional e coeficiente de cisalhamento. Os campos de pressão junto ao fundo refletem o comportamento da superfície livre do escoamento. Através do mapeamento das velocidades dentro do tanque da escada do tipo ranhura vertical, foi possível caracterizar as duas zonas de recirculação e a região do jato principal. Os máximos valores médios de velocidade encontram-se na seção da ranhura, não excedendo 1,00 m/s (no modelo). Além disso, foi possível reconhecer as regiões de maior energia cinética da turbulência que apresentaram valores de até 1000 cm2/s2 na região do jato principal, as quais coincidem com as zonas de maiores tensões de Reynolds da ordem de até 30 N/m2. A partir das velocidades médias e em função das velocidades de nado dos peixes obtidos na literatura, foi possível a identificação de locais que atuam como “barreiras” ao deslocamento de determinada espécie. A avaliação qualitativa do comportamento da trajetória dos peixes dentro dos tanques mostrou-se de acordo com a definição destas “barreiras” hidráulicas e com a avaliação do comportamento dos campos de energia cinética da turbulência e das tensões de Reynolds. No modelo da escada do tipo descarregador de superfície observaram-se as máximas velocidades médias sobre o descarregador, com valores de até 1,73 m/s. Verifica-se que na maior parte do tanque as velocidades médias não ultrapassam 40% da velocidade potencial. Foram encontrados valores de energia cinética da turbulência até 2000 cm2/s2, com valores na maior parte do tanque em torno de 200 cm2/s2. Quanto aos campos de tensões de Reynolds, têm-se, na maior parte do tanque, os valores entre −5 e 5 N/m2, sendo que na região do jato mergulhante, os valores chegam até 30 N/m2. Na escada com orifícios de fundo verificaram-se as maiores velocidades médias nos planos sob influência do fluxo principal proveniente do orifício. Não foi possível a medição da velocidade na seção da abertura, sendo que os valores medidos no tanque não ultrapassaram 50% da velocidade potencial. Os máximos valores de energia cinética da turbulência atingem até 2000 cm2/s2 junto ao fundo, enquanto na região central do tanque, o valor médio é um pouco inferior a 200 cm2/s2. Os valores de tensão de Reynolds encontram-se entre −30 e 30 N/m2, com a maioria das regiões entre −5 e 5 N/m2. Os valores máximos e médios de energia cinética da turbulência e tensões de Reynolds para os modelos com descarregador de superfície e com orifício de fundo encontram-se na mesma faixa. Isto indica que, sob o ponto de vista técnico, possivelmente o critério de escolha entre essas duas estruturas recai nas características da ictiofauna. A passagem com ranhura vertical permite a escolha da profundidade preferencial de nado. No entanto, nessa estrutura, verificase que as componentes médias e turbulentas, nas regiões de descanso, comparando-as com os valores máximos do jato principal, são superiores proporcionalmente, às observadas nas zonas de recirculação das outras duas estruturas. As informações biológicas disponíveis na literatura não permitem a definição de condições preferenciais em relação aos parâmetros de turbulência entre as estruturas aqui avaliadas. No entanto, as informações obtidas nesse trabalho indicam que a energia cinética da turbulência e as tensões de Reynolds podem ser indicativos da tolerância ou preferência dos peixes até certos níveis de turbulência. / Fish facilities are structures or systems that enable fish passage through dams or obstructions. Fishways represent one of the most common fish facilities types worldwide, presenting different geometries and designs. The choice and design of these structures must attend the fish swimming performance and biological characteristics. For some species, as salmon, there are defined designs that can be successfully applied, however, that does not occur for the majority of the species and many projects have showed unsatisfactory efficiency. In Brazil, there is an immense diversity of fish species, whose swimming characteristics are strongly different than the other known species like the salmons. This fact, associated with the increasing requirement of fish facilities implantation in dams, mainly through environmental State Laws, requires the definition of structures adapted to the Brazilian fish. The design criteria validation needs, necessarily, studies to evaluate hydraulic characteristics on structure proposals and the interaction of the flow with swimming abilities. The number of researches related to the hydraulic functioning of fishways is increasing, however they are still insufficient and there is not a consensus on the criteria, either for its complete characterization, either to define which parameters should be considered. The flow turbulence patterns in fishways, whose characteristics are assumed to be related with the degree of acceptance or rejection of the species, are practically unknown. In this study a hydraulic characterization was carried out through an experimental study, including three kinds of fishways: (1) with vertical slots; (2) with rectangular notches and (3) with bottom orifices. The facilities were set up in the Hydraulic Research Institute of the Federal University of Rio Grande do Sul (IPH/UFRGS) and in the National Laboratory of Civil Engineering – LNEC - Portugal. The following measurements were carried out: three-direction velocities, in a 3D-mesh in one pool of each structure, with Acoustic Doppler Velocimeters and water level of the free surface with a point gauge and scales. In the vertical slot fishway it was carried out complementary measurements of pressure in the bottom of the channel, with transducers. For each structure three discharges were tested. Besides defining the mean flow characteristics, the velocity time data, filtered (through a digital process) without substitution, allowed to analyze some turbulence parameters, as turbulence kinetic energy, turbulence intensity and Reynolds’ shear stresses. In the vertical slot fishway model it was verified that the hydraulic parameters are in agreement with similar structures of the literature, among them, discharge coefficient, adimensional discharge and friction factor. The bottom pressure field agrees with the behavior of the free-surface flow. Two recirculation zones and the area of the main jet were characterized through the velocities distribution inside the vertical slot fishway pool. The maximum mean velocity values were found in the slot section, not exceeding 1.00 m/s (in the model). Moreover, it was possible to recognize the areas with larger turbulence kinetic energy that presented values of up to 1000 cm²/s² in the main jet area, which correspond to the largest Reynolds’ shear stresses values of up to 30 N/m². Considering mean velocities data and fish swimming capabilities, it was possible to identify regions that are insurmountable by the fish. The qualitative approach of the fish trajectory inside the structure agrees with the insurmountable regions described through mean velocities and with the distribution of turbulence kinetic energy and Reynolds’ stresses. In the fishway model with rectangular notches, the maximum mean velocities were observed on the weir, with values of up to 1.73 m/s. In the major part of the pool, mean velocities do not surpass 40% of the potential velocity. Values up to 2000 cm²/s² for turbulence kinetic energy were found, with values in the major part of the pool close to 200 cm²/s². For Reynolds’ stresses, the major part of the structure works with values in the range of −5 and 5 N/m², and in the jet entrance pool region the values are of up to 30 N/m². In the fishway with bottom orifices the largest mean velocities were verified in the plans under influence of the main flow connecting consecutive orifices. The measurement of the velocities in the orifice section was not possible and the values measured in the pool did not exceed 50% of the potential velocity. The maximum values of turbulence kinetic energy reached up to 2000 cm²/s² close to the bottom channel, while in the central area of the pool, the mean value is lower than 200 cm2/s2. The values of Reynolds’ shear stresses are between −30 and 30 N/m2, with the major part between−5 and 5 N/m2. The maximum and mean values of turbulence kinetic energy and Reynolds’ stresses in the models with rectangular notches and with orifice are in the same range. It indicates that the choice between these two structures relapses in the fish swimming characteristics. The passage with vertical slot allows the choice of the swimming depth preference. However, in the resting areas of this structure, it is verified that the mean and turbulent components when compared with the maximum values of the main jet, are higher proportionally, to the ones observed in the recirculation zones of the other two structures. The biological information available in the literature does not allow the definition of preferential conditions in relation to the turbulence parameters among the structures here appraised. However, the information obtained in this work indicates that the turbulence kinetic energy and Reynolds’ shear stress can be indicatives of the tolerance or preference of the fish to certain turbulence levels.

Submitted by Mery Piedad Zamudio Igami (mery@ipen.br) on 2017-03-10T15:20:01Z No. of bitstreams: 1 22060.pdf: 3959215 bytes, checksum: 0a58b276270c30e02d2f6935bafba229 (MD5) / Made available in DSpace on 2017-03-10T15:20:01Z (GMT). No. of bitstreams: 1 22060.pdf: 3959215 bytes, checksum: 0a58b276270c30e02d2f6935bafba229 (MD5) / Neste trabalho avaliou-se a ação positiva da substituição de lantânio por praseodímio e de lantânio por magnésio na performance eletroquímica de eletrodos de ligas de armazenamento de hidrogênio em estado bruto de fusão e com tratamento térmico. O La foi substituído por Mg nas ligas La0,7-xMgxPr0,3Al0,3Mn0,4Co0,5Ni3,8 (x=0,0-0,7) e por Pr nas ligas La0,7-yPryMg0,3Al0,3Mn0,4Co0,5Ni3,8 (y=0,0-0,7). Os parâmetros eletroquímicos analisados foram ativação, capacidade de descarga, retenção da capacidade de descarga, autodescarga e alta taxa de descarga. As ligas apresentaram comportamento passivo em relação à corrosão. As análises por MEV/EDS e por DRX com refinamento por Rietveld revelaram a presença majoritária de fases similares às fases LaNi5, PrNi5, LaMg2Ni9 e PrMg2Ni9 em função das composições das ligas estudadas. Os parâmetros de rede e os volumes da célula unitária das fases diminuíram com a substituição crescente de La por Mg e de La por Pr. As capacidades de descarga máxima decresceram com a substituição crescente de La por Mg e de La por Pr, acompanhando o decréscimo da abundância da fase similar à fase LaNi5 e o aumento da abundância da fase similar à fase LaMg2Ni9. Comparativamente, menores taxas de autodescarga e maior estabilidade cíclica foram observadas para o eletrodo da liga na condição x=0,1, ao passo que o eletrodo da liga na condição y=0,0 apresentou maiores valores de alta taxa de descarga, indicando melhor performance cinética. / Tese (Doutorado em Tecnologia Nuclear) / IPEN/D / Instituto de Pesquisas Energéticas e Nucleares - IPEN-CNEN/SP

Summer monsoon current (SMC) in the north Indian Ocean (NIO) is an open ocean current that flows eastward and enter into Bay of Bengal (BoB) during southwest monsoon (June–September). South of Sri Lanka, the SMC turns north-eastward instead of following eastward course and feeds into the BoB. Understanding the dynamics of SMC is crucial to understanding the interaction between Arabian sea (AS) and BoB. In the year 2009, the current moved north-eastward, meandered and then finally terminated into the southeast BoB. The northward bend of the current southeast of Sri Lanka has been attributed to the interaction of the eastward SMC with Rossby waves radiated from eastern boundary. An anticyclonic vortex formed right of the north-eastward meandering current which was associated with significantly high speeds. Reasons behind the unusually high speeds of SMC in this region remain unknown. Processes involved in the interaction of eddies with eastward SMC, the meandering of the current and its ultimate termination in southeast BoB are also not understood. Our study investigates the evolution, intensfication and meandering of SMC around Sri Lanka using an Indian Ocean general circulation model (MOM4p1) simulation for the year 2009. The model simulation, when compared with observational data OSCAR, showed good agreement. The study also explores the role of local and remote forcing in modulating the dynamics of SMC in the region. An eddy kinetic energy budget analysis for the region was performed which indicates the region to be a zone of significant eddy activity. Both barotropic, baroclinic instabilities were found to be the dominant mechanisms behind the generation of eddies. Based on eddy energetic analysis, the evolution of SMC was classified into stages i ii of onset, intensification, anticyclonic bend, anticyclonic vortices formation and meandering. Effect of eddies on mean flow were studied with the help of a transformed Eulerian mean (TEM) approach under quasi-geostrophic approximation. Eddy potential vorticity fluxes appearing in the TEM momentum equation and eddy enstrophy decay, divergence of eddy enstrophy advection from eddy enstrophy equation, helped to un-derstand when, where and how the eddies tended to drive the mean flow. Rossby waves and other westward propagating eddies arriving from the east, energise the SMC in June and induce an acceleration tendency on the mean flow through regions associated with upgradient eddy potential vorticity flux. In addition to the eddies, local winds also play a crucial role in driving the mean flow. Wind power, surface mean ocean kinetic energy and available potential energy (APE) were computed, integrated over the region of interest and compared to each other. The effect of local winds appear to be predominant in driving the mean flows as it not only increases the surface mean kinetic energy of the SMC but also raises the isopycnals and builds up large amount of APE in the ocean. Baroclinic instability takes place in late July and early August associated with the release of APE which flattens the isopycnals and thus weakens the SMC. Consequently the SMC meanders in course of time and flows into southeast BoB.

Les interactions non-covalentes ont une importance fondamental pour la chimique. Les interactions entre un catalyseur et le substrat, interactions entre matériaux, synthèse des enantiomers parmi autres réactions chimiques sont décrivent par interactions non-covalentes. Elles sont fondamental pour designer nouveaux matériaux. Les interactions non-covalents sont très suivant visualisées à partir de mesures de contactes atomiques qui utilisent des donnes de rayon de van de Waals. Cette type d'approximations ne sont pas très flexible pour comprendre l'interaction avec l'environnement. Aujourd'hui les approximations qui utilisent des fonctions dans l'espace réel (i.e. la densité électronique) sont très utilisent pour visualiser les interactions non-covalentes. Dans cet thèse, on analyse la méthode NCI pour visualiser interactions chimiques. On analyse les gradient réduit de la densité, ingrédient fondamental dans la méthode NCI. On montré que cette fonction est liée au la densité d'énergie cinétique et au comportement bosonique du système. On montre que la méthode NCI peut être utilisée pour analyser tous les types d'interactions; dès interactions covalentes aux non-covalentes. Finalement la méthode est appliqué à la réactivité chimique. / Non-covalent interactions are of paramount importance in chemistry. Interactions between a catalyst and its substrate, self-assembly of nanomaterials, enantiomer production and many other chemical reactions, are most of the time non-covalent in nature. They are also fundamental for crystallographic analysis, since they set up the scenario for molecular crystallization, whose guiding rules are still a fruitful filed of research. Non-covalent interactions are frequently visualized using distance dependent contacts, generally without consideration of hydrogen atoms. Most of these interactions are usually identified by the use of tabulated van der Waals radii, which are not flexible enough to reveal the interplay with the environment. New approaches, based on 3D functions that can be derived either form experiment or computation (e.g. the electro density) are now widely used to identify and visualize non-covalent interactions. In this thesis we analyse the NCI method, and namely, its main ingredient, the reduced density gradient. Its capabilities for visualizing chemical interactions are examined. This 3D function is then, connected with the kinetic energy density and a interpretation of the reduced density gradient in terms of the bosonic behaviour of the electronic system is presented. Then, the NCI method is applied to visualise and analyse chemical interactions: from covalent to non-covalent interactions. The chemical reactivity is also addressed. The NCI method is applied to rationalised the outcome of several reactions.