O projeto estrutural de túneis envolve o conhecimento da variável convergência (fechamento radial) para definição da pressão atuante sobre o revestimento e do nível de descompressão do solo circundante (maciço). A adoção de modelos contínuos em elementos finitos consegue discretizar de forma mais adequada o modelo mecânico estrutural considerando a interação com o solo. Dessa forma, esta dissertação recai sobre a variação da razão de cobertura (profundidade) de um túnel, a fim de determinar a profundidade limite que o modelo numérico segue o padrão de tensões geostáticas crescentes com a profundidade versus um padrão geostático uniforme profundo. A validação do modelo numérico ocorre junto a soluções analíticas desenvolvidas por Park (2004) e Kirsch (1898) para túneis em maciços elásticos superficiais e profundos respectivamente. Nos exemplos analisados, quando a profundidade do eixo do túnel é dez vezes maior que seu diâmetro as convergências se aproximam daquelas dadas pelo GEOMEC 91 (Bernaud, 1991) e o túnel pode ser considerado profundo. A curva de assentamentos (recalques) foi considerada consequência importante a ser notada em túneis superficiais que mostraram um recalque máximo até duas vezes maior quando se reduz a profundidade Para esse estudo foram desenvolvidos scripts com profundidade variável adicionando a superfície de plasticidade de Drucker-Prager (utilizada em solos friccionais). O modelo numérico desenvolvido para túneis superficiais é validado junto a túneis reais instrumentados que tenham sido avaliados a partir de algum outro modelo contínuo, são eles: Heathrow Express, Green Park, Metrôs de Barcelona, Brasília e São Paulo com resultados variando entre 1,20% e 6,78%. Eventualmente, uma breve análise do desenvolvimento de tensões no revestimento para túneis circulares, elípticos e ferradura é realizada a fim de encontrar zonas de tração. Enquanto túneis em seção ferradura desenvolveram tração para qualquer profundidade devido ao formato retilíneo, túneis circulares precisaram se encontrar com baixa profundidade a fim de desenvolver zonas descomprimidas decorrentes da ovalização de túneis rasos. / The structural design of tunnels involves the knowledge of the variable convergence (radial closure) for the definition of the acting pressure upon the lining and for the decompression level of the involving soil (rockmass). The adoption of continuous models using finite element analysis (FEA) is able to assess the structural mechanical model accurately, concerning the soil interaction. This way, this work studies the covering ratio (depth) of tunnels, in a way to determine the threshold point which divides the geostatic growing stresses versus the deep uniform stresses pattern. The validation of the numerical model under an elastic rockmass is done by comparison with analytical solutions developed by Park (2004) and Kirsch (1898) for shallow and deep tunnels respectively. Regarding the analyzed examples, when the depth of the tunnel axis is ten times greater than its diameter; the convergences approach to the ones given by GEOMEC 91 (Bernaud, 1991) and the tunnel may be considered deep. The settlement through (ground movement) is considered an important consequence to be noticed upon shallow tunnels; these tunnels have led to a maximum settlement twice deeper when the covering ratio is shallow (H/D = 1). For this study, several scripts have been developed concerning the variable depth and adding the Drucker-Prager plasticity criterion for frictional soils. The numerical model developed for shallow tunnels is validated with other continuous models relied on real instrumentation of subway network tunnels, such as: Heathrow Express, Green Park, Barcelona, Brasília and São Paulo with the solutions ranging between 1,20% and 6,78%. Eventually, a brief analysis of the stress development in the lining for circular, ellipsoid and horseshoe tunnels is done in search of tension zones. In one hand, horseshoe tunnels develop tension for any depth due to its straight walls; in the other hand, circular tunnels need to have a small covering ratio in order to develop decompression zones caused by the shallow tunnels ovalization.

Climate changes alter species composition in both aquatic and terrestrial ecosystems. For establishment of species in a fish community, dispersal barriers, habitat structure and interactions between species act as a filter. The aim of this study was to examine changes in the fish community in Lake Tåkern, Sweden, by repeating a test-fishing in a lake 40 years after a previous test-fishing. Special interest was directed towards the possible establishment of bream (Abramis brama) and (Cyprinus carpio) which recently had been found in the lake. In addition, a decline of the low oxygen tolerant species crucian carp (Carassius carassius) and tench (Tinca tinca) was expected. Test-fishing was performed in June and August 2019 and combined with analyses of Environmental-DNA from samples in September 2019. A comparison with the average catch of 1978-1979 together showed that all cyprinid species had decreased in 2019. The hypoxic tolerant crucian carp decreased 16-fold in numbers, which was expected due to fewer events of cold winters and hypoxia. One bream was caught from the test-fishing and was also detected through eDNA-analysis. The ecological status of the fish community was classified as moderate but showed a high proportion of piscivore percids contra cyprinids, compared to six other shallow lakes in Sweden with similar characteristics. The high proportion of piscivore percids is probably contributing to the stability of the present clear-water state of the lake.

River confluences - locations where rivers join one another - are fundamental components of natural drainage networks. Differences in topography, geology, soils, land use, and human activities within watersheds upstream of confluences can produce differences in thermal or chemical properties of river flows and in the materials transported by these flows. Mixing is initiated along the mixing interface (MI) that develops between the two incoming streams with different properties. Therefore, the understanding of fluvial processes at confluences is important for determining river mixing both at and downstream of individual confluences and at the scale of drainage networks. The primary goal of this thesis is to describe the main mechanisms that control mixing and transport at river confluences and the role played by the complex flow structures in the flow and how they change with planform geometry and other flow and geometrical parameters. The study is carried out using Computational Fluid Dynamics modeling based on the state of the art Detached Eddy Simulation approach and High Performance Computing. By starting with a mixing layer between parallel streams with simple geometry, the model is validated based on laboratory experiment data. Moreover, some hypotheses regarding the growth of the mixing layer are amended with the extensive data provided by the model, which is a valuable supplement to the experiment. By performing a detailed parametric study in very long and wide domains for simplified cases one can focus on the spatial development of the MI and the large scale coherent structures forming within and in the vicinity of the MI without the complications of other factors. More specifically, the influence of velocity and density difference of the two streams, flow depth, inflow conditions and angles between the two streams on the spatial development of the MI is analyzed. The data resulting from these simulations conducted in simple geometries is a unique set of data which can be used to test and improve theoretical models used to predict global parameters describing flow and mixing at natural river confluences. In particular, this research uses for the first time well resolved Large Eddy Simulation based techniques to investigate how density differences between the incoming streams affect the spatial development of the mixing interface and mixing downstream of the confluence apex. In order to investigate flow dynamics, mixing processes and effects of temperature stratifications at natural river confluences with discordant bed, a series of simulations is performed for the confluence of the Ebro and Segre Rivers in Spain, which is one of the most studied confluences in Europe. With the detailed survey data of the confluence bed and flow conditions data provided, the goal is to understand the main mechanisms responsible for mixing at a confluence with a large bed discordance and how the velocity ratio between the two incoming streams affects mixing. Besides, more insights are provided that if temperature stratification effects affect significantly flow structure and mixing based on real conditions recorded at a natural confluence. The study provides a comprehensive set of flow data in the confluence including velocity, temperature distribution etc. It serves as important supplement to the field measurements, which are generally more difficult to obtain. It also allows estimating scale effects between field conditions and conditions at which laboratory experiments of confluence flow and mixing are conducted.

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第22264号 / 理博第4578号 / 新制||理||1657(附属図書館) / 京都大学大学院理学研究科地球惑星科学専攻 / (主査)准教授 松四 雄騎, 教授 千木良 雅弘, 教授 岩田 知孝 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DGAM

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第22756号 / 工博第4755号 / 新制||工||1744(附属図書館) / 京都大学大学院工学研究科都市社会工学専攻 / (主査)教授 細田 尚, 教授 戸田 圭一, 准教授 音田 慎一郎 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Liquid wastes discharged from industrial outfalls have been researched for many years in the past. Majority of past studies, initiated in 1960s, were experimental studies mainly focused on basics of discharges such as key geometrical properties. Eventually, more robust experimental studies were performed to measure the mixing properties of effluent discharges with various jet configurations and ambient water conditions. Discharges could be as a means of submerged diffusers or surface channels and receiving water could vary from a homogenous calm ambient to a very complex stratified turbulent cross flow ambient. Depending on the bathymetric and economic situation around an outfall project, submerged discharges are preferred designs for most of ocean outfalls. It is the reason that majority of past studies have evaluated the mixing characteristics of submerged jets. Since early 1990s, the numerical modelling has emerged to support complex fluid mechanic problems. Later in 1990s and early in 2000s, the use of computational fluid dynamic (CFD) tools emerged in predicting the jet properties for the effluent discharges. Since then different numerical models have been developed for different applications. Similar to experimental studies, most of numerical studies have been focused on the submerged dense jet discharges. The current study intends to stay focused on the numerical modelling of such jets too; however, to cover the gaps in the literature. To achieve this, a thorough literature review was performed on the past CFD studies of over past 20 years to better understand what was done and what the gaps are. The results of this thorough review revealed that although there has been a great progress in the CFD studies in the field of effluent discharges, there are some applications that have not been investigated before, yet. It was found that there are some discharge inclinations that were not studied numerically before. Four discharge angles of 60°,75°, 80° and 85° were selected in this study, as previous studies mostly focused on 30° and 45°. The higher inclinations are more suitable for deep water outfalls where terminal rise height of the jet does not attach to the ambient water surface. The numerical model OpenFOAM was used in this study which is based on the Finite Volume Method (FVM) applying LRR turbulence model closure. LRR turbulence models was proved to be a capable choice for effluent discharge modelling. The second gap identified in the comprehensive literature review completed was the submerged dense effluent discharge into shallow water with surface attachment (for both inclined and vertical discharges). There was no previous numerical study of such jets identified. Three different regimes were identified: full submergence, plume contact and centerline impingement regimes (i.e. FSR, PCR and CIR). Key geometrical and dilution properties of these jets at surface contact (Xs, Ss) and return point (Xr, Sr) were extracted numerically and compared to those available from experiments. Two discharge angles (30° and 45°) were investigated based on the available experimental data. Five Reynolds-averaged Navier-Stokes (RANS) turbulence models were examined in this study: realizable k-ε and k-ω SST models (known as two-equation turbulence models), v2f (four equations to model anisotropic behavior) and LRR and SSG turbulence models (known as Reynolds stress models - six equations to model anisotropic behavior). Vertical dense effluent discharges are popular in the design of outfall systems. Vertical jets provide the opportunity to be efficient for a range of ambient currents, where the jet will be pushed away not to fall on itself. This research work investigates worst case scenario in terms of mixing and dilution of such jets: vertical dense effluent discharges with no ambient current and in shallow water where jet impacts the surface. This scenario provides a conservative design criteria for such outfall systems. The numerical modelling of such jets has not been studied before and this research work provides novel, though preliminary, insights in simulations of vertical dense effluent discharges in shallow waters. Turbulent vertical discharges with Froude numbers ranging from 9 to 24 were simulated using a Reynolds stress model (RSM), based on the results from inclined dense discharges to characterize the geometrical (i.e., maximum discharge rise Zm and lateral spread Rsp) and dilution μmin properties of such jets. Three flow regimes were reproduced numerically, based on the experimental data: deep, intermediate and impinging flow regimes.

Topic of this master`s thesis is founding of the hightway bridge, which is located between Lipník and Bělotín. Thesis is aimed for creating shallow foundation and foundation on piles and choosing which option is better. All evaluations have been made without specialized software and than compared with GEO5 results. Better solution will be completed with procedure for construction and drawing documentation.

Tato diplomová práce se zabývá matematickým modelováním vodních vln v blízkosti pobřeží pomocí parciálních diferenciálních rovnic. Cílem této práce je formulace pohybových rovnic a jejich následné numerické řešení s grafickou interpretací dosažených výsledků.

This research involves shake table testing of 1g scale models that mimic the coupled seismic response of a structure on a shallow mat foundation and foundation soil (known as soil-foundation-structural-interaction or SFSI). In previous research, SFSI effects have been quantified through analytical models, numerical analyses, and limited field data. This research works towards increasing the amount of empirical data through scale model shake table testing. A suite of earthquake time histories is considered in evaluating a nominal 10th scale soil-structure model using a flexible wall barrel on a 1-D shake table. San Francisco Young Bay Mud (YBM) is used as the prototype soil and long period narrow building as the prototype structure. Foundation embedment depth, fundamental mode of the structure, and seismic loading function are varied to generate a large database of SFSI results under controlled conditions. The foundation level response is compared to free-field responses to determine the magnitude of the SFSI. The results confirm the effects of foundation embedment on the peak ground motion and the spectral acceleration at the predominant period of the structure. The foundation level accelerations are deamplified compared to free-field results. Results also confirm the legitimacy of the testing platform and program by comparing the data to previous experimental study.

Le vent joue un rôle clé dans la structure et le fonctionnement des lacs. Le brassage de la colonne d’eau est notamment impliqué dans la compétition entre les différentes espèces de phytoplancton dont les stratégies de positionnement vertical varient largement entre groupes. Par exemple, les cyanobactéries sont globalement favorisées par des conditions de brassage faible grâce à des mécanismes physiologiques de flottaison leur permettant de se maintenir à une profondeur désirée, alors que les diatomées et les algues vertes qui n’ont pas de système de flottaison dépendent d’un brassage important pour être suspendues dans la zone euphotique. Dans les lacs peu profonds, le brassage généré par le vent est généralement suffisant pour entrainer la resuspension des sédiments. Ceci a pour conséquence directe une augmentation de la turbidité qui peut directement impacter la croissance et la productivité algale. La seconde conséquence est la libération de phosphore précédemment stocké dans les sédiments. La pollution et l’usage d’engrais ont provoqué un important import de phosphore dans les lacs qui sont pour la plupart dans un état d’eutrophisation. Dans un premier temps, cette thèse a eu pour objectif de modéliser la compétition entre une cyanobactérie et une algue verte dans une colonne d’eau de lac peu profond prenant en compte la possibilité de resuspension des sédiments et de phosphore. Différents niveaux de brassage ont été testés : de quasi-stagnant à tempête. Des centaines de simulation ont été réalisées dans différentes conditions de brassage, d’eutrophisation (de oligo- à hypereutrophe) et de réchauffement climatique (actuel et +2°C) permettant d’établir des domaines de conditions de dominance d’une espèce ou de l’autre. Dans une seconde partie, la thèse a eu pour objectif d’explorer les impacts potentiels du brassage sur l’ensemble du réseau trophique pélagique. En effet, la grande majorité des études portant sur le brassage sont des études in situ focalisées sur le phytoplancton. Dans le cas présent, nous avons utilisé des mésocosmes de 15m3 équipés de batteurs à vagues dont les amplitudes, fréquences et longueurs d’ondes sont modifiables. Une première expérience en 2012 a eu pour objet de suivre durant 9 semaines la dynamique des variables physico-chimiques standards et des communautés de phytoplancton, zooplancton, bactéries et virus dans deux modalités de brassage : brassage superficiel et brassage de l’ensemble de la colonne d’eau avec resuspension des sédiments. L’année suivante, la même expérience a été conduite mais cette fois-ci couplée à un traitement de réchauffement appliqué grâce à des bâches de polyéthylène transparentes posées sur les mésocosmes. Ces deux expériences montrent un effet important du brassage sur le phytoplancton avec une augmentation du contenu en chlorophyll a et une augmentation de la productivité. Au niveau du zooplancton, la réponse dépend du sous-groupe. Dans nos expériences, uniquement les copépodes ont répondu négativement au régime de brassage avec un potentiel effet cascade sur la population de rotifers. L’expérience de 2012 suggère également un renforcement de l’activité de lyse bactérienne par les virus dans des conditions de brassage important. En 2013, aucun effet du réchauffement de l’eau de 1°C n’a été observé sur les variables mesurées. Des analyses supplémentaires en cours devraient permettre de compléter les informations obtenues et de vérifier l’absence d’effet du réchauffement. / The wind plays a key role in the structure and functioning of lakes. Water column mixing is involved in the competition between different species of phytoplankton with different vertical positioning strategies. For example, buoyant cyanobacteria are generally favored by low mixing conditions, while sinking diatoms and green algae depend on mixing to be suspended in the photic zone. In addition, in shallow lakes, wind-induced mixing is usually sufficient to cause sediment resuspension. This leads to an increase in turbidity which can directly impact the algal growth and productivity. Resuspension can also induce the release of phosphorus previously stored in sediments. This internal loading via resuspension can boost algal production and growth. The first goal of this thesis was to model the competition between the buoyant cyanobacterium Microcystis aeruginosa and a sinking green algae under different mixing conditions: from quasi-stagnant to storm-like events. Hundreds of simulations were carried out in different conditions of mixing, phosphorus concentration and water temperatures in order to establish the conditions necessary for the dominance of one species on the other. In the second part of this thesis, we explored the potential impacts of mixing on the pelagic food web. The vast majority of previous studies on mixing in shallow lakes are in situ studies focused on phytoplankton. In the summer 2012, we used 15m3 mesocosms equipped with wave-makers and followed during nine weeks the dynamics of physical and chemical variables and the dynamic of phytoplankton, zooplankton, bacteria and viruses under two mixing conditions: whole column mixing with sediment resuspension and superficial mixing without resuspension. The following year, the same experiment was conducted but this time crossed with a warming treatment. Both experiments gave similar results in regards to mixing. We observed a significant effect of mixing on phytoplankton with an increase in chlorophyll content and an increase in productivity. The response of zooplankton to mixing depended on the subgroup. In our experiments, only copepods responded negatively to mixing with a potential cascade effect on the rotifer population while bosminas remained unaffected throughout the experiments. The 2012 experience also suggests an increase in bacterial lysis activity by virus in mixed enclosures. In 2013, no effect of water warming (+1°C) was observed on the measured variables, however, additional analysis are still underway to confirm or infirm these results.