Thesis (Ph. D.)--University of Texas at Austin, 2007. / Vita. Includes bibliographical references.

As plantações florestais, historicamente, estão associada a intensos debates sobre o seu impacto na dinâmica da água do solo ou subterrânea. Nesse contexto, este trabalho buscou avaliar os possíveis impactos da mudança de cobertura do solo, de pastagem para eucalipto, sobre o aquífero freático em zona de afloramento do Sistema Aquífero Guarani. A área de pesquisa está localizada na bacia do Ribeirão da Onça, situada no município de Brotas, no centro-leste do Estado de São Paulo. Essa é monitorada desde 2004. Foram utilizados dois métodos para avaliar o impacto da mudança de uso do solo, o método iWTF e o método de Darcy. A partir desses métodos, estimou-se as taxas de recarga subterrânea, fluxo lateral, variação de armazenamento e percolação profunda. Além disso foram obtidas a variação do nível freático, espessura saturada e gradiente hidráulico horizontal para o aquífero na área de estudo. Durante os anos hidrológicos de 2005 a 2011, a recarga direta variou de 255 e 876 mm (iWTF), equivalente a 21% e 51% da precipitação anual correspondente. Após 2011, com a mudança de cobertura do solo de pastagem para eucalipto, não obstante os anos hidrológicos de 2012 e 2013 apresentarem precipitação próxima da média de longo período, a recarga direta foi estimada em cerca de 210 mm (2012) e 147 mm (2013), equivalentes a 13% e 10% da precipitação anual. A diminuição na recarga direta levou à redução da espessura saturada, variação do nível freático, gradiente hidráulico horizontal, fluxo lateral e armazenamento anual. Além disso, pode-se perceber que a plantação florestal está exercendo influência sobre a área de pastagem adjacente à plantação. Tem-se observado essa influência através da existência de fluxo transversal em direção à plantação florestal, o qual tem elevado o armazenamento no período de estiagem. Os resultados indicam que a plantação de eucalipto afetou significativamente o comportamento sazonal e anual do balanço hídrico na área de estudo. / Historically forest plantations are associated with intense debate about its impact on soil water dynamics or groundwater. Thus, the objective of this study was to evaluate the potential impacts of land cover change of pasture to eucalyptus on the groundwater in an outcrop zone of the Guarani Aquifer System. The study area is located in the Ribeirão da Onça watershed, located in the municipality of Brotas, in the center-east of the state of São Paulo. To assess the impact of land cover change, I used two methods: i. iWTF and ii. Darcy. From these methods, we estimated groundwater recharge rates, lateral flow, water storage variation and deep percolation. Furthermore, I obtained the water table variation, saturated thickness and horizontal hydraulic gradient into the aquifer in the study area. During the hydrological years 2005-2011, the direct recharge through the results of iWTF method varied between 256 and 876 mm, equivalent to 21% and 51% of the corresponding annual precipitation. After 2011, with the land cover change of pasture to eucalyptus, the direct recharge was computed at about 210 mm (2012) and 147 mm (2013), equivalent to 13% and 10% of the annual precipitation. I did not find significant difference in the annual precipitation in these years compared to the historical mean. Therefore, my findings have shown that the decrease of direct recharge are associated to the land cover change of pasture to eucalypts. The decrease in direct recharge leads to the reduction of saturated thickness, the water level variation, horizontal hydraulic gradient, lateral flow and annual storage. Moreover, it can be seen that the forest plantation has influenced the pasture area, adjacent to the plantation. It has been observed that influence through the existence of cross flow toward the forest plantation, which has high storage, especially in the dry season. The results indicate that the eucalyptus plantation significantly affected the seasonal and annual behavior of the water balance in the study area.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Civil Engineering, 1980. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING. / Bibliography: leaves 60-61. / by Gregory Paul Matthews. / M.S.

京都大学 / 0048 / 新制・課程博士 / 博士(農学) / 甲第21151号 / 農博第2277号 / 新制||農||1059(附属図書館) / 学位論文||H30||N5125(農学部図書室) / 京都大学大学院農学研究科地域環境科学専攻 / (主査)教授 村上 章, 教授 川島 茂人, 教授 藤原 正幸 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DFAM

Packed beds find extensive application in a wide variety of industries. The objective of the present work is to analyze and evaluate the effects of the wall on structural characteristics, hydrodynamics and heat transfer in packed beds of spheres. As a first attempt, spheres of uniform size are considered. The cylindrical wall of the bed confines the location of the particles thus leading to significant radial variations in void fraction and specific lateral surface area. The two characteristics at any given radial position r* are estimated by defining a concentric cylindrical channel (CCC) of an arbitrary thickness such that its boundaries are equidistant from the cylindrical surface passing through r* and accounting for the solid volumes or lateral surface areas of the segments of spheres (cap, slice, rod and annular ring) contained in the CCC and with centers lying within a distance of a particle radius from r*.The curved boundaries of the sphere segments are rigorously accounted for. The low aspect ratio beds (aspect ratio less than or equal to 2) show three distinct types of behavior. In beds of aspect ratio 2, the void fraction starts from a value of unity at the wall and decreases to a minimum and then increases to unity at the center of the bed. In beds with aspect ratio between l\/¯3/2, there is a continuous decrease in void fraction from unity at the wall to a fairly low value towards the axis and then a slight increase followed by another decrease. The profiles for aspect ratio less than l\/¯3/2 show a continuous decrease from a value of unity at the wall to zero towards the axis. In contrast, beds of high aspect ratio show heavily damped oscillations in the void fraction up to about five particle diameters from the wall and then a constant value. The lateral surface area variations in low aspect ratio beds show a steep fall from a very high value near the wall, and in high aspect ratio beds an oscillatory nature, though not as strong as in the corresponding void fraction profiles. The distribution of flow in packed beds for steady flow of an incompressible Newtonian fluid under isothermal conditions is modeled by using Ergun equation with Brinkman-type correction to account for the viscous effects in the region close to the wall. The confining effect of the wall is incorporated through the radial variations in void fraction and specific lateral surface area. The hydraulic radius in the region next to the wall is modified to take into account the resistance of the wall surface to flow. The resulting model equations with appropriate boundary conditions are solved numerically by collocation technique. The influence of aspect ratio in the range 1.25 to 20.3 and Reynolds number from 0.1 to 1000, the two most important factors affecting the flow behavior, is evaluated. The velocity profiles show a peak in the region close to the wall thus indicating severe channeling effect in this region. The magnitude and location of the peak depend on aspect ratio and Reynolds number. The model predictions agree remarkably with reported experimental data on velocity profiles in a bed of aspect ratio 10.7, and on the effect of Reynolds number on friction factors in beds of low aspect ratio. The radial variations in void fraction, velocity and effective thermal conductivity are incorporated in the two-dimensional pseudo-homogeneous steady-state model to analyze the wall effects on heat transfer in packed beds. Both constant wall temperature and constant wall flux boundary conditions are adopted. The equations are solved numerically using finite difference technique. The radial temperature profiles are seen to be fairly uniform in beds of low aspect ratio thus showing that the often made assumption of complete radial thermal mixing in low aspect ratio beds is valid. Beds of high aspect ratio show strong radial gradients. For constant heat flux condition the slope of the temperature profile remains constant after a small distance from the Inlet thus leading to thermally fully-developed flow. For this condition the heat transfer equations are solved analytically to obtain expressions for Nusselt number and the radial temperature profiles. There is a significant difference in the temperature profiles evaluated in the presence and absence of wall effects. Good agreement is found between the Nusselt numbers obtained from the model and reported experimental data.

Not available / text

Not available / text

This thesis discusses and develops one approach to solve parabolic partial differential equations with random input data. The stochastic problem is firstly transformed into a parametrized one by using finite dimensional noise assumption and the truncated Karhunen-Loeve expansion. The approach, Monte Carlo discontinuous Galerkin (MCDG) method, randomly generates $M$ realizations of uncertain coefficients and approximates the expected value of the solution by averaging M numerical solutions. This approach is applied to two numerical examples. The first example is a two-dimensional parabolic partial differential equation with random convection term and the second example is a benchmark problem coupling flow and transport equations. I first apply polynomial kernel principal component analysis of second order to generate M realizations of random permeability fields. They are used to obtain M realizations of random convection term computed from solving the flow equation. Using this approach, I solve the transport equation M times corresponding to M velocity realizations. The MCDG solution spreads toward the whole domain from the initial location and the contaminant does not leave the initial location completely as time elapses. The results show that MCDG solution is realistic, because it takes the uncertainty in velocity fields into consideration. Besides, in order to correct overshoot and undershoot solutions caused by the high level of oscillation in random velocity realizations, I solve the transport equation on meshes of finer resolution than of the permeability, and use a slope limiter as well as lower and upper bound constraints to address this difficulty. Finally, future work is proposed.

L'accumulation du soluté à la surface d'une membrane entraîne le phénomène de polarisation de concentration. Ceci est un problème qui affecte tous les systèmes de filtration membranaire car il a pour effet une augmentation de la pression osmotique et par conséquence une réduction substantielle du flux de perméat. Afin de comprendre ce phénomène, nous avons d'abord mené une étude analytique de la filtration tangentielle en solvant pur prenant en compte de l'influence de la pression motrice locale sur le taux de perméation. Lors de cette étude, des solutions analytiques qui augmentent en précision avec l'ordre développé ont ete dérivées. Ensuite nous avons développé une approche analytique qui couple l'hydrodynamique aux transferts de matière pour le cas d'un système de filtration qui opère sous haute pression avec un taux de récupération faible. Dans le but d'intégrer à la fois la dépendance de la pression transmembranaire locale sur le flux de perméat et l'influence de la polarisation de concentration à travers leurs effets osmotiques sur la pression effective, nous avons développé un modèle numérique qui résout l'équation de conservation du soluté couplée aux équations de Navier-Stokes en régime stationnaire dans l'approximation de Prandtl. Nous avons validé cette approche grâce aux solutions analytiques précédemment dérivées. Ensuite, nous avons testé l'influence des principaux paramètres de fonctionnement sur la performance du système et comparé nos résultats avec ceux d'autres modèles numériques. Finalement, la pertinence du modèle a été quantitativement vérifiée grâce à des données tirées des expériences bien documentées en osmose inverse. / Concentration polarization of solute at the membrane surface, because of osmotic pressure effects, is an important phenomenon that can cause substantial reductions in permeation. To understand these phenomena: we first analyze the filtration process for a pure solvent, imposing the influence of the driving pressure on permeation at the membrane. We obtain accurate analytical solutions for the flow fields. We then derive an analytical solution that coupled hydrodynamics to mass transfer for filtration systems working in a situation of High Pressure and Low Recovery. Second, we develop a numerical model that incorporates both physical aspects: the dependency of pressure on permeation and the influence of concentration polarization and their related osmotic effects in the effective pressure at the membrane. For that, the numerical approach solves the solute conservation equation coupled with the Navier-Stokes equations under the steady Prandtl approximation. The solution of the system is performed using a finite difference method of order 2. The validity of this approach is successfully demonstrated with the previous analytical solutions for hydrodynamics, as well as for the coupling with mass transfer. We then test the influence of the main operating parameters (inlet concentration, axial flow rate, operating pressure and membrane permeability) on the performance of the filtration system and compare the results with other numerical models that takes into account concentration polarization phenomenon. Finally, the validity of this model is quantitatively well-proved when using the reported data resulting from reverse osmosis experiments.

Dans cette thèse, nous étudions l'équation de la chaleur couplée avec la loi de Darcy à travers de la viscosité non-linéaire qui dépend de la température pour les dimensions d=2,3 (Hooman et Gurgenci ou Rashad). Nous analysons ce problème en introduisant la formulation variationnelle équivalente et en la réduisant à une simple équation de diffusion-convection pour la température où la vitesse dépend implicitement de la température.Nous démontrons l'existence de la solution sans la restriction sur les données par la méthode de Galerkin et du point fixe de Brouwer. L'unicité globale est établie une fois la solution est légèrement régulière et les données se restreignent convenablement. Nous introduisons aussi une formulation variationnelle alternative équivalente. Toutes les deux formulations variationnelles sont discrétisées par quatre schémas d'éléments finis pour un domaine polygonal ou polyédrique. Nous dérivons l'existence, l'unicité conditionnée, la convergence et l'estimation d'erreur a priori optimale pour les solutions des trois schémas. Par la suite, ces schémas sont linéarisés par des algorithmes d'approximation successifs et convergentes. Nous présentons quelques expériences numériques pour un problème modèle qui confirme les résultats théoriques de convergence développées dans ce travail. L'estimation d'erreur a posteriori est établie avec deux types d'indicateurs d'erreur de linéarisation et de discrétisation. Enfin, nous montrons des résultats numériques de validation. / In this thesis, we study the heat equation coupled with Darcy's law by a nonlinear viscosity depending on the temperature in dimension d=2,3 (Hooman and Gurgenci or Rashad). We analyse this problem by setting it in an equivalent variational formulation and reducing it to an diffusion-convection equation for the temperature where the velocity depends implicitly on the temperature.Existence of a solution is derived without restriction on the data by Galerkin's method and Brouwer's Fixed Point. Global uniqueness is established when the solution is slightly smoother and the dataare suitably restricted. We also introduce an alternative equivalent variational formulation. Both variational formulations are discretized by four finite element schemes in a polygonal or polyhedral domain. We derive existence, conditional uniqueness, convergence, and optimal a priori error estimates for the solutions of the three schemes. Next, these schemes are linearized by suitable convergent successive approximation algorithms. We present some numerical experiments for a model problem that confirm the theoretical rates of convergence developed in this work. A posteriori error estimates are established with two types of errors indicators related to the linearisation and discretization. Finally, we show numerical results of validation.