Beach profile models predict the changes in bathymetry along a line perpendicular to the shoreline. These models are used to forecast bathymetric changes in response to storms, sea level rise or human activities such as dredging and beach nourishment. Process-based models achieve this by simulating the physical processes that drive the sediment transport as opposed to behavior models which simulate observed profile changes without resolving the underlying processes. Some of these processes are wave shoaling and breaking, boundary layer streaming, and offshore-directed undertow currents. These hydrodynamic processes control the sediment processes such as sediment pick-up from the bottom, diffusion of the sediment across the water column and its advection with waves and currents. For this study, newly developed sediment transport and boundary layer models were coupled with existing models of wave transformation, nearshore circulation and bathymetry update, to predict beach profile changes. The models covered the region from the dry land to a depth of 6-8 meters, spanning up to 500 meters in the cross-shore direction. The modeling system was applied at storm time scales, extending from a couple of hours to several days. Two field experiments were conducted at Myrtle Beach, SC, involving the collection of wave, current and bathymetric data as a part of this study. The results were used to calibrate and test the numerical models along with data from various laboratory studies from the literature. The sediment transport model computes the variation of sediment concentrations over a wave period and over the water column, solving the advection-diffusion equation using the Crank-Nicholson finite-difference numerical scheme. Using a new approach, erosion depth thickness and sediment concentrations within the bed were also predicted. The model could predict sediment transport rates for a range of conditions, within a factor of two. It successfully computed the sediment concentration profile over the water column and within the bed and its variation throughout a wave period. Erosion depth and sheet flow layer thickness were also predicted reasonably well. Wave heights across the profile were predicted within ten percent when the empirical wave breaking parameter was tuned appropriately. Mean cross-shore velocities contain more uncertainty, even after tuning. The importance of capturing the location of the maximum, near-bottom, cross-shore velocity when predicting bar behavior was shown. Bar formation, erosion, accretion, onshore and offshore bar movement were all computed with the model successfully

L’assimilation variationnelle de données 4D-Var est une technique très utilisée en géophysique, notamment en météorologie et océanographie. Elle consiste à estimer des paramètres d’un modèle numérique direct, en minimisant une fonction de coût mesurant l’écart entre les sorties du modèle et les mesures observées. La minimisation, qui est basée sur une méthode de gradient, nécessite le calcul du modèle adjoint (produit de la transposée de la matrice jacobienne avec le vecteur dérivé de la fonction de coût aux points d’observation). Lors de la mise en œuvre de l’AD 4D-Var, il faut faire face à des problèmes d’implémentation informatique complexes, notamment concernant le modèle adjoint, la parallélisation du code et la gestion efficace de la mémoire. Afin d’aider au développement d’applications d’AD 4D-Var, le logiciel YAO qui a été développé au LOCEAN, propose de modéliser le modèle direct sous la forme d’un graphe de flot de calcul appelé graphe modulaire. Les modules représentent des unités de calcul et les arcs décrivent les transferts des données entre ces modules. YAO est doté de directives de description qui permettent à un utilisateur de décrire son modèle direct, ce qui lui permet de générer ensuite le graphe modulaire associé à ce modèle. Deux algorithmes, le premier de type propagation sur le graphe et le second de type rétropropagation sur le graphe permettent, respectivement, de calculer les sorties du modèle direct ainsi que celles de son modèle adjoint. YAO génère alors le code du modèle direct et de son adjoint. En plus, il permet d’implémenter divers scénarios pour la mise en œuvre de sessions d’assimilation.Au cours de cette thèse, un travail de recherche en informatique a été entrepris dans le cadre du logiciel YAO. Nous avons d’abord formalisé d’une manière plus générale les spécifications deYAO. Par la suite, des algorithmes permettant l’automatisation de certaines tâches importantes ont été proposés tels que la génération automatique d’un parcours “optimal” de l’ordre des calculs et la parallélisation automatique en mémoire partagée du code généré en utilisant des directives OpenMP. L’objectif à moyen terme, des résultats de cette thèse, est d’établir les bases permettant de faire évoluer YAO vers une plateforme générale et opérationnelle pour l’assimilation de données 4D-Var, capable de traiter des applications réelles et de grandes tailles. / Variational data assimilation 4D-Var is a well-known technique used in geophysics, and in particular in meteorology and oceanography. This technique consists in estimating the control parameters of a direct numerical model, by minimizing a cost function which measures the misfit between the forecast values and some actual observations. The minimization, which is based on a gradient method, requires the computation of the adjoint model (product of the transpose Jacobian matrix and the derivative vector of the cost function at the observation points). In order to perform the 4DVar technique, we have to cope with complex program implementations, in particular concerning the adjoint model, the parallelization of the code and an efficient memory management. To address these difficulties and to facilitate the implementation of 4D-Var applications, LOCEAN is developing the YAO framework. YAO proposes to represent a direct model with a computation flow graph called modular graph. Modules depict computation units and edges between modules represent data transfer. Description directives proper to YAO allow a user to describe its direct model and to generate the modular graph associated to this model. YAO contains two core algorithms. The first one is a forward propagation algorithm on the graph that computes the output of the numerical model; the second one is a back propagation algorithm on the graph that computes the adjoint model. The main advantage of the YAO framework, is that the direct and adjoint model programming codes are automatically generated once the modular graph has been conceived by the user. Moreover, YAO allows to cope with many scenarios for running different data assimilation sessions.This thesis introduces a computer science research on the YAO framework. In a first step, we have formalized in a more general way the existing YAO specifications. Then algorithms allowing the automatization of some tasks have been proposed such as the automatic generation of an “optimal” computational ordering and the automatic parallelization of the generated code on shared memory architectures using OpenMP directives. This thesis permits to lay the foundations which, at medium term, will make of YAO a general and operational platform for data assimilation 4D-Var, allowing to process applications of high dimensions.

L’aérosolthérapie a pour objectif de délivrer un médicament dans les voies respiratoires. Le nébuliseur pneumatique est un dispositif permettant de générer des gouttelettes de liquide de diamètre micrométrique. Son processus d’atomisation a cependant été peu analysé. Ainsi, les performances du nébuliseur, caractérisées par le diamètre des gouttes et la masse de médicament inhalable par le patient, et atteignent un palier. Notre travail consiste à utiliser un modèle numérique diphasique en 3D basé sur une géométrie donnée et paramétré sous ANSYS Fluent. Plusieurs méthodes sont utilisées pour caractériser expérimentalement la génération de l’aérosol : l’ombroscopie, la diffractométrie laser et l’anémométrie phase Doppler. Notre modèle est validé par rapport aux données expérimentales et peut donc être exploité pour analyser les processus de génération. L’influence de plusieurs paramètres physiques sur les caractéristiques de l’aérosol produit est étudiée. Ainsi, l’étape de génération de gouttelettes est optimisée pour le développement d’un nouveau nébuliseur. Le transport des gouttes aux poumons du patient est optimisé empiriquement. / The purpose of aerosol therapy is to deliver drugs into respiratory airways. The jet nebulizer is a device used to generate liquid droplets with a diameter lower than 5 μm. However its atomization process was not much analyzed. Nebulizer performances, which are characterized with droplet size and drug mass inhaled by the patient, are empirically optimized and have reached a plateau. Our work consists in setting a 3D diphasic numerical model on ANSYS Fluent, based on a given geometry. Several methods are used to experimentally characterize aerosol generation: shadowgraphy, laser diffractometry and phase Doppler anemometry. Our model is validated by experimental data and helps predicting generation processes. The influence of several geometric and physical parameters on the output is studied. From these data, droplet generation is optimized for the development of a new nebulizer. Droplet transport to the patient lungs is empirically optimized.

Les machines frigorifiques à absorption ammoniac-eau sont prometteuses dans les domaines de la climatisation solaire et de la valorisation des rejets thermiques pour l’industrie. Pour permettre à ces machines de devenir compétitives par rapport aux systèmes à compression mécanique de vapeur, l’amélioration de leur efficacité et la baisse de leur coût sont nécessaires. C’est dans ce contexte que s’inscrit ce travail de thèse.L’étude se concentre sur l’absorbeur qui est un des composants les plus critiques de la machine à absorption en matière de taille, de coût et d’efficacité. L’objectif est d’étudier numériquement et expérimentalement les transferts couplés de masse et de chaleur dans l’absorbeur dans le but de prédire et d’améliorer ses performances.Deux absorbeurs à film tombant sont étudiés, dans lesquels la solution pauvre et la vapeur entrent en haut et le fluide caloporteur entre en bas. Le premier est un échangeur à plaques soudées et le deuxième est un échangeur à plaques et joints avec des dimensions et des profils de plaques différents.L’analyse expérimentale de ces deux absorbeurs est réalisée dans des conditions réelles de fonctionnement sur un prototype instrumenté de machine à absorption ammoniac-eau de 5 kW. Ce dispositif permet une analyse globale des débits de vapeur absorbés, des flux thermiques évacués et des efficacités d’absorption. Une analyse plus locale est aussi réalisée à l’aide de mesures de températures à l’intérieur des canaux de refroidissement dans l’absorbeur plaques et joints. Les résultats montrent une importante corrélation entre la puissance frigorifique produite par la machine à absorption et les performances de l’absorbeur. Mais ce prototype étant une machine réelle, les variables d’entrée de l’absorbeur ne peuvent pas être contrôlées. Un modèle numérique est donc nécessaire pour dissocier l’impact des différentes variables sur les performances de l’absorbeur.Un modèle 1D d’un absorbeur à film tombant est donc développé. Il est basé sur des bilans de masses, d’espèces et d’énergies, des équations de transferts de masse et de chaleur et des conditions d’équilibre à l’interface liquide-vapeur. Les résistances aux transferts de masse sont considérées dans les phases liquide et vapeur et des corrélations empiriques sont utilisées pour calculer les coefficients de transfert de masse et de chaleur.Ce modèle est validé expérimentalement avec les données globales aux bornes des deux absorbeurs et avec les mesures de températures le long des canaux du fluide de refroidissement puisqu’une différence maximale de 15% est observée. Il permet donc l’analyse détaillée des phénomènes de transferts de masse et de chaleur le long de l’absorbeur et facilite l’étude du procédé d’absorption.Enfin, une étude de sensibilité paramétrique est réalisée avec ce modèle pour discuter des résultats expérimentaux et pour identifier les pistes d’amélioration des performances de l’absorbeur et donc de la machine à absorption. / Ammonia-water absorption chillers are promising both for solar air conditioning and for industry processes. To become competitive compared to electric compression chillers, their efficiency needs to be improved and their cost has to be decreased. This thesis study takes place in this context.The focus is put on the absorber, which is one of the most critical component of absorption chillers in terms of compactness, cost and efficiency. The purpose is to study numerically and experimentally coupled heat and mass transfers which occur in the absorber in order to predict and improve its overall performances.Two falling film absorbers are analysed. In both of them, the poor solution and the vapour enter at the top and the coolant fluid enters at the bottom of the absorber. The first absorber is a brazed plate heat exchanger and the second is a gasketed plate-and-frame heat exchanger with different geometric dimensions and plates corrugations.The experimental study of these two absorbers is performed in real working conditions on an instrumented ammonia-water absorption chiller prototype of 5 KW. Thanks to this device, a global analysis of vapour absorbed mass flow rates, absorbed heat fluxes and mass effectiveness is achieved. A local analysis is also performed thanks to temperature measures inside channels of coolant fluid in the gasketed plate-and-frame heat exchanger. Results show a strong correlation between the absorption chiller cooling capacity and the absorber performances. However, since this prototype is a real chiller, absorber inlet variables cannot be controlled. Thus, a numerical model is necessary to dissociate the impact of these variables on the absorber performances.A 1D numerical model of the absorber is developed. It is based on mass, species and enthalpy balances, mass and heat transfer equations and equilibrium conditions at the vapour/solution interface. Mass transfer resistances in both liquid and vapour phases are considered while heat and mass transfer coefficients are calculated using empirical correlations.This model is validated experimentally with global data at the inlet and the outlet of the absorber and temperature measures along the absorber coolant fluid channels. A maximal relative error of 15 % is observed. Therefore, a detailed analysis of combined heat and mass transfers along the absorber and the absorption process study is performed thanks to this model.A parametric study is also performed with this model to discuss experimental results and find ways to improve the absorber performances and thus the absorption chiller performances.

Na prática de fundações por estacas helicoidais submetidas a esforços de tração (também conhecidas como ancoragens helicoidais) frequentemente são observadas discrepâncias entre valores de capacidade de carga previstos por métodos teóricos e medidos em provas de carga. Entre outros fatores, estas diferenças ocorrem pelo fato de nenhum dos métodos teóricos existentes considerarem adequadamente o efeito da instalação no comportamento deste tipo de ancoragem. Durante a penetração por rotação das hélices da estaca/ancoragem no terreno, os parâmetros de resistência e de rigidez do solo penetrado, e as tensões radiais em torno da estaca são modificados. O nível de degradação do solo que suportará as cargas aplicadas na ancoragem é dependente do tipo e características do solo, e do número e geometria das hélices. Diante deste cenário de incertezas provenientes do efeito da instalação, duas distintas abordagens probabilísticas são apresentadas nesta dissertação. A primeira abordagem tem o intuito de verificar o efeito da variabilidade de distintas variáveis influentes no comportamento de ancoragens helicoidais. Para este fim, foi utilizado o método de expansão em Polinômio de Caos, baseado na modelagem numérica de ancoragens helicoidais em areia, calibrada com os resultados medidos em modelos testados em centrifuga. Foi verificado nesta parte o nível de influência de cada parâmetro de entrada do solo no comportamento carga x deslocamento da ancoragem. No segundo enfoque, foi proposto e avaliado um modelo probabilístico para simular a curva carga x deslocamento, obtida em provas de carga. Nesta parte, foi realizada uma análise de confiabilidade para verificação do estado limite de serviço de ancoragens helicoidais baseada em resultados de provas de carga executadas no Brasil. São também apresentadas nesta análise as incertezas provenientes da variabilidade do fator de torque utilizado na verificação da capacidade de carga durante a instalação da ancoragem. / Discrepancies between predicted and measured results of uplift capacity of helical piles (or helical anchors) have been commonly observed in the practical experience. Among other factors, these differences occur because the existing theoretical methods do not consider adequately the installation effect on the behavior of this type of anchor. During the anchor installation, the penetration by rotation of the helical plates into the ground modifies the soil strength and stiffness parameters, and the radial stresses around the pile. The degradation degree of the soil penetrated is dependent on the type and characteristics of the soil, and also on the number and geometry of helices. Against this background of uncertainty due to the installation effect, two different probabilistic approaches are presented in this dissertation. The first approach is used to verify the effect of the variability of different influential variables on the behavior of helical anchors. For this purpose, the probabilistic method named Sparse Polynomial Chaos Expansion was used, based on the numerical modeling of helical anchors in sand, calibrated with measured results obtained from reduced models tested in centrifuge. As a result, the level of influence of each soil input parameter on the anchor load-displacement behavior is presented. In the second approach, a probabilistic model to simulate the load-displacement curve obtained in load tests was proposed and evaluated. For this part, a reliability analysis was performed to check the serviceability limit state of helical anchors, based on the results of field load tests performed in Brazil. In addition, the uncertainties arising from the variability of the torque factor were evaluated. The torque factor is frequently used to verify the uplift capacity of helical anchors during installation.

Made available in DSpace on 2015-03-26T13:28:14Z (GMT). No. of bitstreams: 1 texto completo.pdf: 5185340 bytes, checksum: 2affa702a631cd03bd10eb05ee16522a (MD5) Previous issue date: 2013-01-28 / In the latest years, several studies about shear connectors in perforated flat steel plates have been conducted in order to assess its application in composite structures of steel and concrete, including experimental and numerical approaches. Nevertheless, the behavior of these connectors remains a subject of investigation. This work presents the development of a numerical model capable of simulating the behavior of steel-concrete connection with Crestbond shear connectors using the Finite Element Method. The numerical analysis was performed using the ABAQUS software, considering the elastoplastic behavior for steel and the Concrete Damage Plasticity model for concrete. In order to validate the numerical model, experimental results obtained by Veríssimo (2007) and Oliveira (2007) were used for comparisons. In numerical simulations were evaluated the influence of concrete constitutive model, and the influence of the steelconcrete connection representation using two strategies: Tie restrictions and contact interactions between surfaces. These strategies proved to be able to adequately represent the phenomena observed in laboratory tests, especially those that used contact interactions between surfaces. The numerical model is suitable for simulating the structural behavior of Crestbond connectors, being able to represent the interaction between steel and concrete, to characterize the maximum load resisted by the connection and to characterize the deformation of the connector, providing a useful tool for several other studies. / Nos últimos anos, vários estudos sobre conectores de cisalhamento em chapas planas de aço perfuradas têm sido conduzidos, com vistas à sua aplicação em estruturas mistas de aço e concreto, incluindo abordagens experimentais e numéricas. Apesar disso, o comportamento desses conectores ainda permanece como objeto de investigações. Neste trabalho, apresenta-se o desenvolvimento de um modelo numérico capaz de simular o comportamento da ligação aço-concreto com conectores de cisalhamento Crestbond, utilizando o Método dos Elementos Finitos. A análise numérica foi feita por meio do software ABAQUS, considerando-se o comportamento elastoplástico para o aço e a aplicação do Modelo de Dano Plástico para o concreto. Para validação do modelo numérico, foram utilizados resultados experimentais obtidos por Veríssimo (2007) e por Oliveira (2007). Nas simulações numéricas foram avaliadas a influência do modelo constitutivo do concreto e a influência do modo de representação da ligação aço-concreto através de duas estratégias: utilização de restrições tipo Tie e utilização de interações de contato entre superfícies. Estas estratégias mostraram-se capazes de representar adequadamente os fenômenos observados nos ensaios de laboratório, principalmente quando se utilizaram interações de contato entre superfícies. O modelo numérico desenvolvido é adequado para a simulação do comportamento estrutural dos conectores Crestbond, sendo capaz de representar a interação entre aço e concreto, de caracterizar a carga máxima resistida pela conexão e de caracterizar as deformações do conector, constituindo uma ferramenta útil para vários outros estudos.

Desde meados de 2007, um aumento significativo de unidades construídas pelo sistema Paredes de concreto tem influenciado o mercado da construção habitacional no Brasil. Em abril de 2012, foi publicada uma norma brasileira que trata exclusivamente desse sistema. Além disso, o desafio em minimizar o déficit habitacional estimula a utilização desse método alternativo, pois quando aplicado adequadamente proporciona alta produtividade e menor custo em relação a outros métodos construtivos. Por essas razões, é importante conhecer quais modelos de cálculo são seguros e pertinentes para a análise estrutural de edificações construídas pelo sistema Paredes de Concreto. Neste trabalho são avaliados, por meio da comparação de resultados, quatro modelos de cálculo, dos quais dois são numéricos, utilizados para o pré-dimensionamento e dimensionamento de paredes de concreto armado, moldadas no local. Este tipo de estrutura tem sido bastante utilizada na construção de edifícios. Deste modo utilizou-se como estudo de caso, um projeto adaptado de um edifício real com quinze pavimentos tipo, para a aplicação dos modelos de cálculo e algumas normas de projeto. Os dois modelos numéricos foram desenvolvidos no software SAP2000, baseado no Método dos Elementos Finitos, considerando análise elástica linear. O primeiro discretiza as paredes com elementos de casca e o segundo com elementos de barra. A interação entre o solo e a estrutura não foi considerada neste trabalho. Normas de projeto foram comparadas na etapa de pré-dimensionamento quanto às resistências últimas à compressão com objetivo de avaliar a recente norma brasileira. A verificação ao dimensionamento das paredes quanto às tensões normais de compressão e tração e tensões de cisalhamento foi realizada baseada nas expressões da NBR 16055:2012. Por fim, verificou-se o Estado Limite de Serviço das deformações excessivas globais, referentes à movimentação lateral do edifício e a estabilidade global do edifício estudado, pela NBR 6118:2007. / Since 2007, a significant use of the system known as Concrete Wall has been influenced the Brazilian residential construction market. In April 2012, a Brazilian code exclusively devoted to this system was published. Besides that, the challenge in reducing housing deficit estimulate the use of this alternative method, because when properly applied it provides high productivity and lower costs compared with other construction methods. For these reasons it is important to evaluate which designing models are safe and appropriate to structural analysis of buildings constructed in Concrete Wall system. In this work it is evaluated, through results comparison, four designing models, of which two are numerical, used to pre-design and to design reinforced concrete walls casted in place. This kind of structure has been widely used in buildings construction. Thus, an adapted project from a real building with fifteen floors was used as a case study in order to apply the designing models and some project design codes. Both numerical models were developed in SAP2000 software, based on Finite Element Method, considering elastic linear analysis. The former used shell elements to represent the walls and the latter used frame elements. The soil-structure interaction has not been considered in this work. Project design codes were compared at the pre-design stage in terms of the ultimate compression resistance, aiming evaluate the recent Brazilian code. The check of normal stress of compression, normal stress of tension, and shear stress, was made based on NBR 16055:2012 expressions. Finally, the service limit state related to global excessive deformation due to the lateral movement of the building and the global stability was verified through NBR 6118:2007.

Apresenta-se um trabalho numérico e experimental, com o objetivo de estudar o comportamento dos pavimentos de concreto estruturalmente armados, quando submetidos a forças verticais estáticas e repetidas. Avalia-se a contribuição da armadura de flexão, na resistência de placas de concreto apoiadas sobre meio elástico. Verifica-se experimentalmente o modo de ruína de placas submetidas a carregamento monotônico e a carregamento repetido, com e sem armadura, e acompanha-se o desenvolvimento das fissuras no concreto e das deformações no aço. Comparam-se os resultados obtidos, com as recomendações da NBR 6118:2003, para o dimensionamento de estruturas submetidas à fadiga. É desenvolvido modelo numérico capaz de representar a interface do solo com a placa do pavimento, a fissuração do concreto, a contribuição da armadura e o comportamento pós-fissuração. Com auxílio do modelo numérico, validado a partir dos resultados experimentais, estuda-se a influência de alguns parâmetros importantes para o dimensionamento dos pavimentos, tais como: capacidade de suporte da fundação, espessura das placas, área de aço, dimensões das placas em planta, posição de aplicação da força e presença de juntas de transferência de deslocamento. Os resultados experimentais mostram uma significativa contribuição da armadura positiva na resistência de placas isoladas sob forças verticais centradas. Verificou-se que as forças repetidas provocam fadiga do aço e que o número de ciclos depende da deformação provocada na armadura. A partir dos resultados são traçadas diretrizes para o dimensionamento, no qual devem ser considerados os momentos positivos e os negativos, e deve ser feita a verificação da fadiga do concreto e da armadura / A numerical and experimental study on the behavior of structurally reinforced concrete pavement under vertical static and repeated load is presented. The contribution of bending reinforcement to the strength of concrete slab is assessed. It is experimentally verified the fracture of reinforced and plain concrete slabs under monotonic loads and under cyclic loads. The concrete cracking and the strain on steel are monitored. The results are compared to the brazilian code recommendations to the design of structures under fatigue. A numerical model that represents the soil-slab interface, the concrete cracking, the reinforcement contribution and the tension stiffening behavior is developed. Some important parameters to the pavement design are analyzed using this finite element model, whose reliability was verified to experimental results. These parameters are: the soil capacity; the slab thickness; reinforcement ratio; slab dimensions; load positions; and joints conditions. The tests data show an important contribution of the positive reinforcement on the strength of an isolated slab with centered load. It has been verified the fatigue of the steel. The number of cycles depends on the reinforcement strain. The results are used to propose some directions to the design. The positive and negative bending moments have to be considered and the fatigue on the concrete and on the reinforcement has to be verified

O bulbo molhado formado na irrigação por gotejamento pode ser modelado a partir informações das propriedades físico-hídricas do solo. Mas apesar do grande progresso na modelagem, a aplicação de modelos na irrigação e drenagem ainda não foi implementada em nenhum nível de maneira substancial. Assim, o objetivo do presente trabalho foi desenvolver um modelo numérico utilizando a técnica dos volumes finitos para estimar a forma e as dimensões do volume de solo molhado sob irrigação por gotejamento, e verificar a validade do modelo através da comparação com dados recolhidos em condições experimentais. O ensaio foi conduzido na área experimental do Departamento de Engenharia de Biossistemas da Escola Superior de Agricultura Luiz de Queiroz-ESALQ/USP. As sondas de TDR foram confeccionadas segundo procedimentos descritos por Souza et al. (2006). O perfil do bulbo molhado formado no solo abaixo de um emissor do tipo gotejador foi determinado para 3 tipos de material: areia, solo arenoso e solo argiloso e para 3 vazões nominais do emissor: 2, 4 e 6 L/h. Cada uma das combinações entre tipo de solo e vazão foi repetida três vezes em caixas (1000 L) diferentes, totalizando 27 medições. Um modelo numérico foi desenvolvido para simular a distribuição da água no perfil do solo abaixo de uma fonte pontual. A solução da equação diferencial para o movimento da água em solo não saturado foi realizada pela discretização do espaço-tempo utilizando a técnica dos volumes finitos. Um algoritmo em linguagem Visual Basic foi escrito para implementar o conjunto de equações e simular a evolução do bulbo molhado no tempo. Foi simulada a formação do bulbo molhado nas mesmas condições do experimento, e gráficos de isolinhas de umidade foram traçados no software Surfer. O modelo numérico proposto foi capaz de simular a formação do bulbo molhado em diferentes condições de tipo de solo e vazão do emissor. / The soil water patterns in drip irrigation can be modeled from physical and hydraulic properties of soil. But despite the great progress in modeling, the models have not been implemented at any level in irrigation and drainage. The objective of this study was to develop a numerical model using finite volume technique to estimate the shape and dimensions of the wetted soil volume under drip irrigation and verify the validity of the model by comparing data collected under experimental conditions. The trial was conducted at the experimental site of the Department of Biosystems Engineering, Escola Superior de Agricultura Luiz de Queiroz- ESALQ/USP. The TDR probes were manufactured according Souza et al. (2006). The soil water patterns under a drip emitter type was determined for three types of material: sand, sandy soil and clay soil and three nominal flows of the emitters: 2, 4 and 6 L/h. Each of the combinations of soil type and flow rate was repeated three times in boxes (1000 L), totaling 27 measurements. A numerical model was developed to simulate the distribution of water in the soil profile below a point source. The solution of the differential equation for the movement of water in unsaturated soil was carried out by the discretization of space-time using the technique of finite volume. An algorithm in Visual Basic language was written to implement the set of equations and simulate the evolution of wetted soil volume in time. Contour plots of soil water content were drawn in Surfer software. The proposed numerical model was able to simulate wetted soil volume under different conditions of soil type and flow of the emitter.

Abstract A conceptual framework was developed to assess how changes in temperature and precipitation affect sub-surface hydrology, groundwater recharge, groundwater quantity and quality. A conceptual and statistical approach was developed to predict groundwater level variations. Daily rainfall, snowmelt and evapotranspiration values were generated with a novel conceptual hydrological model developed in this study. These values were cross-correlated with observed groundwater levels to find representative time lags and significant correlations. A statistical model linking rainfall, snowmelt, evapotranspiration and groundwater level was then developed and validated. The model simulated seasonal variations in groundwater level very accurately. A sequential approach was developed to assess surface water-groundwater interactions. The simulated surface water level estimated with the WSFS model and recharge estimated with CoupModel were linked to the groundwater flow model MODFLOW. Groundwater, surface water and snow samples were collected to study the chemical composition of groundwater in an unconfined esker aquifer in Northern Finland. Concentrations of Ca2+, Cl-, NO3-N and SiO2 and electrical conductivity were determined. Water quality in the main aquifer was found to be similar to that in the perched groundwater. Solute concentrations generally decreased during and immediately after snowmelt periods, indicating the importance of snowmelt input for groundwater quality. In the perched groundwater, NO3-N concentration increased with elevated groundwater level, indicating a nitrogen source on the land surface. The Cl- concentration in groundwater decreased when the surface water level rose higher than groundwater level. According to simulation results for the A1B climate change scenario, groundwater recharge is projected to increase in winter months due to increased snowmelt and decreased soil frost depth. The spring snowmelt peak in late spring will decrease. This will reduce aquifer storage in early spring, increasing the vulnerability to summer droughts. It is projected that flow regimes between unconfined aquifers and surface water may change, affecting water quantity and possibly quality in groundwater systems. / Tiivistelmä Tässä työssä kehitettiin konseptuaalinen viitekehysmalli, jolla voidaan arvioida kuinka muutokset lämpötilassa ja sadannassa vaikuttavat hydrologiaan, pohjaveden muodostumiseen, pohjaveden määrään ja laatuun. Tässä työssä kehitettiin ja yhdistettiin konseptuaalinen ja tilastomatemaattinen regressiomalli, jolla voidaan simuloida pohjaveden pinnankorkeuden muutoksia. Konseptuaalisella mallilla laskettiin päivittäinen sadanta, lumen sulanta ja haihdunta. Havaitut pohjaveden pinnankorkeudet korreloitiin sadannan, lumen sulannan ja haihdunnan kanssa, jotta löydettiin merkitsevät korrelaatiot tyypillisillä viiveillä. Lopuksi tilastollinen regressiomalli, joka yhdistää sadannan, lumen sulannan, haihdunnan ja pohjaveden pinnankorkeuden, kalibroitiin ja validoitiin. Kehitetyllä mallilla onnistuttiin simuloimaan vuodenaikainen pohjaveden pinnankorkeus. Yhdensuuntainen mallinnusmenetelmä kehitettiin arvioimaan pinta- ja pohjaveden vuorovaikutusta. Menetelmässä pintaveden korkeus ja pohjaveden muodostuminen linkitettiin ajan suhteen muuttuvina reunaehtoina pohjavedenvirtauksen mallinnusohjelmaan MODFLOW. Simuloidut pintaveden pinnankorkeudet saatiin Suomen ympäristökeskuksen vesistömallijärjestelmästä ja pohjaveden muodostuminen simuloitiin 1D lämmön- ja aineensiirtomallilla, CoupModel. Pudasjärven Törrönkankaan pohjavesimuodostumasta, Pudasjärvestä, Kivarinjoesta ja lumesta kerättiin näytteet ja niistä määritettiin Ca2+, Cl-, NO3-N ja SiO2 pitoisuudet sekä sähkönjohtavuus. Pitoisuudet pohja- ja salpavedessä olivat hyvin samanlaiset. Pitoisuudet yleisesti pienenivät, kun pohjaveden pinnankorkeus nousi etenkin keväisin lumien sulamisen jälkeen. Ainoastaan salpavedessä NO3-N pitoisuus kasvoi, kun pohjaveden pinnankorkeus nousi. Tämä johtuu todennäköisesti salpaveden yläpuolella olevasta NO3-N lähteestä. Cl- pitoisuus pohjavedessä pieneni, kun pintaveden korkeus nousi korkeammalle kuin pohjavesi. A1B ilmastoskenaariossa pohjaveden muodostumisen ennakoidaan lisääntyvän talvikuukausina. Tämä johtuu lumen sulannan lisääntymisestä ja roudan vähenemisestä. Keväinen lumen sulamisen huippu voi mahdollisesti pienentyä ja johtaa pohjavesivarojen pienentymiseen keväisin. A1B ilmastoskenaariossa pinta- ja pohjaveden vuorovaikutus voi myös muuttua ja siten vaikuttaa pohjaveden määrään ja mahdollisesti myös laatuun.