Search

Honeycombs are widely used to laminarize fluid streams by inhibiting the lateral components of the fluctuating velocity. However, they also produce additional turbulence by themselves due to the formation of large-scale instabilities and the breakup of the individual velocity profiles stemming from the honeycomb cells. In the present research, we use 2D-planar particle image velocimetry to study how honeycomb-generated turbulence is affected by a downstream grid. It is found that placing a grid near the honeycomb discharge drastically enhances flow uniformity by separating the strong jets stemming from the individual honeycomb cells into many smaller jets that are much more rapidly dissipated. The results show that using a grid reduces the integral length scale by up to a factor 10, and the axial and lateral energy spectra reveal that the grid primarily limits the energy contained in eddies with lower wave numbers. Furthermore, the grid can reduce the magnitude of peak turbulence intensity by as much as 95% and leads to a large reduction of the correlation length, as long as it is positioned upstream of the onset of the large-scale honeycomb-induced instabilities. A downstream grid is highly beneficial for both a laminar and turbulent honeycomb discharge and is most effective when there is a slight offset between the grid and honeycomb. Even though longer honeycombs generally produce more turbulence than short ones due to the larger length-scale of the shear layers, these effects are almost entirely decoupled when using a honeycomb-grid combination. Finally, a honeycomb-grid combination effectively inhibits both axial and lateral turbulence.

Air–water mixing associated with high turbulence is a common hydraulic phenomenon, where the transport of entrained air bubbles and the dissipation of turbulent energy are key processes affecting the air–water mass transfer. Herein hydraulic jumps were selected as a seminal case of self-air-entrainment in transitional water flow, and the air–water flow properties were investigated at high Froude numbers from 10.5 to 13.5. The analysis of air concentration distributions and bubble size spectra enabled deviation of the rates of bubble diffusion and turbulence dissipation. The study was facilitated with re-analyses of earlier experimental data for smaller Froude numbers between 3.8 and 10. The spatial evolution of the diffusion and dissipation rates suggested that the diffusion of air bubbles and dissipation of turbulent structures were not independent processes, and, at high Froude numbers, the mix of air and water was subject to re-aeration of the turbulent shear layer by the roller.

Microscale atmospheric simulations of the planetary boundary layer are frequently used for wind energy forecasting, emergency response, mountain meteorology, air pollution modeling, and numerous other applications involving atmospheric flows over complex terrain. These models are typically configured using local observations and are limited to resolving only microscale flow features. Mesoscale meteorology, regional influences, and large-scale turbulence are not resolved with the traditional microscale modeling techniques. This dissertation details a series of developments to the Weather Research and Forecasting (WRF) model that enable mesoscale to microscale (i.e. multiscale) simulations. These multiscale simulations dynamically downscale meteorological conditions through a series of nested domains with increasingly high resolution, which allows mesoscale meteorology, regional influence, and large-scale flow features to influence microscale simulations.Over steep terrain, the WRF model develops numerical errors that are due to grid deformation of the terrain-following coordinates. An alternative gridding technique, the immersed boundary method (IBM), has been implemented into the WRF model (Lundquist et al. 2012; Bao et al. 2018). Use of an IBM allows for microscale simulations over highly complex terrain (i.e. urban or mountainous). Here, an IBM and the WRF model’s grid-nesting framework have been modified to seamlessly work together, which allows for a microscale large-eddy simulation over complex terrain with an IBM to be nested within a traditional mesoscale WRF simulation (Wiersema et al. 2020). Additionally, grid configurations are controlled using the vertical grid nesting method of Daniels et al. (2016) and turbulence development at intermediate resolutions is improved using the cell perturbation method of Munoz-Esparza et al. (2015). Multiscale simulations are extremely challenging to configure due to the sensitivity of each nested domain to its configuration and to the configuration of its parent domain(s). This dissertation also begins to investigate the model sensitivity to grid resolution and surface boundary condition, which is integral information for modelers configuring the nested domains of future multiscale simulations.Multiscale simulations are demonstrated for the prediction of transport and mixing of a tracer gas (SF6) released in the central business district of Oklahoma City during the Joint Urban 2003 field campaign. The simulations use either 5 or 6 nested domains with horizontal resolutions that range from several kilometers for the outermost domain to 2~m for the innermost domain. The multiscale simulations are compared with microscale-only simulations and with observations of wind speed, wind direction, and SF6 concentrations. The microscale-only simulations use idealized lateral boundary conditions and are configured using local meteorological observations from the field campaign. The multiscale simulation, which is configured independent of local observations, shows similar model skill predicting wind speed and wind direction, and improved skill predicting SF6 concentrations and turbulence kinetic energy when compared with the microscale-only simulations. Additionally, the multiscale simulation includes the effects of large-scale flow features and turbulence that the microscale-only simulations are incapable of resolving, which is shown to have a dramatic effect on predictions of transport and mixing. The analysis of simulations in this dissertation demonstrates the potential for multiscale simulations to improve predictions of transport and mixing over highly complex terrain and enable microscale simulations where local observations are not available.

The present work examines the hydrodynamics of the inner-shelf region, focusing on tidally-driven alongshore flows over coral reef shelves. This study draws on field data collected in O’ahu, Hawai’i using fixed and mobile assets to develop new modes of observational research.First, a theoretical model is developed to describe how autonomous underwater vehicle (AUV)-based water velocity measurements are influenced by a surface wave field. The model quantifies a quasi-Lagrangian, wave-induced velocity bias as a function of the local wave conditions, and the vehicle’s depth and velocity using a first-order correction to the linear wave solution. The theoretical bias is verified via field experiments over a range of wave and current conditions. The analysis considers velocity measurements made using a REMUS-100 AUV, but the findings apply to any small AUV (vehicle size ≪ wavelength) immersed in a wave field. The observed wave-induced biases [O(1–5) cm/s] can be significant, and can be comparable to steady flow velocities for inner-shelf regions.Second, a new approach to estimate lateral turbulent Reynolds stresses (u′v′) in wavy coastal environments using acoustic Doppler current profilers (ADCPs) is described. The performance of the proposed method is evaluated via comparisons with independent acoustic Doppler velocimeter (ADV)-based stress estimates at two sites, and the vertical structure of the tidally-averaged turbulent Reynolds stresses is examined for an unstratified, tidally-driven flow over a rough coral reef seabed in weak waves. Observations and analysis indicate that lateral stresses are sustained by the cross-shore gradient of the mean alongshore flow, and driven by bottom-generated turbulence. Scaling considerations suggest that cross-shore transport by lateral turbulent mixing could be relevant to coral reef shelves with steep cross-reef slopes and rough bottoms.Finally, a tidally-driven alongshore flow over a forereef shelf is examined using AUV-based spatial velocity measurements along with time series data of the alongshore pressure gradient. Ensemble phase averages of AUV-based velocities reveal characteristics akin to an oscillatory boundary layer, with the nearshore flow leading the offshore flow in phase and with a corresponding velocity magnitude attenuation near the shallower regions of the reef. Analysis of the depth-averaged alongshore momentum equation indicates that the cross-shore structure and evolution of the alongshore flow is well described by a balance between local acceleration, barotropic pressure gradient, and bottom drag. This primary balance allows the estimation of a spatially-averaged drag coefficient as a function of cross-shore distance over depths spanning from 24 to 6 m. Seabed roughness data suggest that larger scales, with wavelengths of O(10 m), are more relevant than smaller meter-scale roughness for drag.

In terms of multiple temporal and spatial scales, massive data from experiments, flow field measurements, and high-fidelity numerical simulations have greatly promoted the rapid development of fluid mechanics. Machine Learning (ML) provides a wealth of analysis methods to extract potential information from a large amount of data for in-depth understanding of the underlying flow mechanism or for further applications. Furthermore, machine learning algorithms can enhance flow information and automatically perform tasks that involve active flow control and optimization. This article provides an overview of the past history, current development, and promising prospects of machine learning in the field of fluid mechanics. In addition, to facilitate understanding, this article outlines the basic principles of machine learning methods and their applications in engineering practice, turbulence models, flow field representation problems, and active flow control. In short, machine learning provides a powerful and more intelligent data processing architecture, and may greatly enrich the existing research methods and industrial applications of fluid mechanics.

The study of complex multiscale flows [@Groen2014-A], like for example the motion of small-scale turbulent eddies over large aerodynamic structures [@Jofre2022-A], microconfined high-pressure supercritical fluids for enhanced energy transfer [@Bernades2022-A], or hydrodynamic focusing of microorganisms in wall-bounded flows [@Palacios2022-A], greatly benefits from the combination of interconnected theoretical, computational and experimental approaches. This manifold methodology provides a robust framework to corroborate the phenomena observed, validate the modeling assumptions utilized, and facilitates the exploration of wider parameter spaces and extraction of more sophisticated insights. These analyses are typically encompassed within the field of Predictive Science & Engineering [@Njam2009-A], which has attracted attention in the Fluid Mechanics community and is expected to exponentially grow as computational studies transition from (mostly) physics simulations to active vectors for scientific discovery and technological innovation with the advent of Exascale computing [@Alowayyed2017-A]. In this regard, the computational flow solver presented aims at bridging the gap between studying complex multiscale flow problems and utilizing present and future state-of-the-art supercomputing systems in academic environments. The solver presented is named RHEA, which stands for open-source Reproducible Hybrid-architecture flow solver Engineered for Academia, and is available as an open-source Git repository at https://gitlab.com/ProjectRHEA/flowsolverrhea.

Aquest treball analitzarà turbulència microconfinada per a fluxos supercrítics realitzant computacions \textit{Direct Numerical Simulation} en alta pressió ($P/P_c = 2$) usant N$_2$ en condicions transcrítiques imposades per una diferència de temperatura entre les parets inferior (${T/T_c}=0.75$) i superior (${T/T_c}=1.5$) amb un Reynolds de fricció de $Re_\tau = 100$ per a la paret inferior. En específic, aquest estudi se centra a analitzar les petites escales del flux usant els invariants del gradient de velocitats. A més, els resultats seran comparats amb un cas en condicions de baixa pressió per caracteritzar les semblances i les diferències en el flux turbulent d'un fluid supercrític i subcrític. Finalment, s'analitzen tots els termes de l'equació de vorticitat per a fluxos compressibles per entendre els mecanismes que dominen la física del fluid al problema estudiat. L'anàlisi conclou que l'existència d'un parell baroclínic localitzat a prop de la paret calenta/superior, generat per la combinació d'una força externa en el sentit de marxa del fluid i el gradient de densitat de la regió pseudoboiling perpendicular a la paret, provoca un increment en la rotació del fluid en comparació amb el cas de baixa pressió. Este trabajo analizará turbulencia microconfinada para flujos supercríticos realizando computaciones usando \textit{Direct Numerical Simulations} en alta presión ($P/P_c = 2$) usando N$_2$ en condiciones transcríticas impuestas por una diferencia de temperatura entre las paredes inferior (${T/T_c}=0.75$) y superior (${T/T_c}=1.5$) con un Reynolds de fricción de $Re_\tau = 100$ para la pared inferior. En específico, este estudio se centra en analizar las pequeñas escalas del flujo empleando los invariantes del gradiente de velocidades. Además, los resultados serán comparados con un caso con condiciones de baja presión para caracterizar las semejanzas y diferencias en el flujo turbulento de un fluido supercrítico y subcrítico. En este estudio se ha encontrado que las pequeñas escalas turbulentas cerca de la pared caliente tienen un comportamiento distinto comparado con los otros casos, el movimiento y forma de los vórtices tienen un comportamiento más extremo, llegando a conseguir más disipación en la región viscosa, así como llegar a observar movimientos de compresión y no solo de dilatación. Finalmente, se analizan todos los términos de la ecuación de vorticidad para flujos compresibles para entender los mecanismos que dominan la física del fluido en el problema estudiado. El análisis concluye que la existencia de un esfuerzo de torsión baroclínico localizado cerca de la pared caliente/superior, generado por la combinación de una fuerza externa en el sentido de marcha del fluido y el gradiente de densidad de la región pseudo-boiling perpendicular a la pared, provoca un incremento en la rotación del fluido en comparación al caso de baja presión. This work, consequently, analyzes supercritical microconfined turbulence by computing direct numerical simulations of high-pressure ($P/P_c = 2$) N$_2$ at transcritical conditions imposed by a temperature difference between the bottom (${T/T_c}=0.75$) and top (${T/T_c}=1.5$) walls for a friction Reynolds number of $Re_\tau = 100$ (bottom wall). In particular, the study focuses on analyzing the small-scale flow structures based on the invariants of the velocity-gradient tensor. Additionally, the results are compared against an equivalent case at low-pressure conditions to carefully characterize the semblances and differences between the turbulent flow at supercritical and subcritical operating conditions. According to this study, turbulent scales near hot walls with high pressure exhibit different behaviours than those with low pressure. As a result, the vortical motion shapes change and become more extreme, resulting in a significantly higher dissipation rate within the viscous layer, as well as observing compressive motions along stretching-dominated regions. Finally, a decomposition of the vorticity transport equation has been performed to provide insight on the fundamental mechanisms governing the flow physics of the problem under study. The analysis concludes that the existence of a baroclinic torque localized in the vicinity of the hot/top wall, which is generated by the combination of the external force driving the flow in the streamwise direction and the density gradient across the pseudo-boiling region in the wall-normal direction, yields a significant increment of flow rotation in comparison to the low-pressure turbulent case also considered.

The transport of bacteria in river systems is a phenomenon which occurs on a multitude of length scales ranging from the size of individual microbes up to the size of an entire estuary. At the same time the understanding of the spreading of microbial populations after a localised contamination event such as a combined sewer overflow is crucial for the prediction of the water quality downstream of the source, which is in turn essential to managing public health. It is well-established that microbial populations in fluvial systems may preferably be found on the surface of small particles rather than solely freely suspended in the water body. The attachment to particles provides an environment beneficial to the survival of bacteria due to the improved access to nutrients and the shielding from environmental stressors, but also alters their dispersion characteristics as the transport of bacteria is then coupled to the trajectories of heavy particles. The importance in the distinction between the particle-attached and the freely-suspended mode of transport has been recognised in the mechanistic modelling of bacteria fate and transport. However, due to the multiscale nature of the problem, the mechanisms which govern the transport of particles in river-like flows are never resolved explicitly, and hence, the models profoundly rely upon the availability of accurate descriptions thereof. The associated problem of particles settling in a turbulent carrier flow is an active topic of research by itself, and is rich in emerging phenomena such as the emergence of spatial inhomogeneities or non-trivial modifications of the settling characteristics compared to quiescent environments. In particular, the transient settling of particles in horizontal open channels, which serves as an abstraction of particle-attached bacteria transport in rivers, has hitherto received only little attention in the literature. As a consequence, the knowledge on the impact of its defining features such as boundedness, anisotropy and vertical inhomogeneity on the settling characteristics is limited and needs to be addressed to enable the formulation of reliable models thereof. The aim of this thesis is to fill the knowledge gap on the transport characteristics of heavy particles in turbulent horizontal open channel flows, and to identify phenomena which may be of importance in the context of bacteria transport modelling. For this purpose, the incompressible Navier--Stokes equations and the momentum balance equations for dispersed particles are solved using direct numerical simulations and the immersed boundary method. This approach resolves all relevant scales of turbulence and the microscopic flow around each particle explicitly, and thus, describes the particle-fluid interaction from fundamental principles of physics without the need of additional modelling. Apart from the contaminated particles, which are introduced near the free surface of the flow, the simulation domain includes approximately 100,000 fully resolved particles at the bottom of the domain, which form a realistic sediment bed, and enable the examination of the interaction between contaminated particles and mobile sediments. Concerning the parameter space, the value of the friction Reynolds number is varied within the range $Re_{\tau} \in [241,838]$, while the contaminant parameter space is chosen such that the resulting relative turbulence intensities---defined as the ratio between the friction velocity and the undisturbed terminal velocity---lie within the range $I_{\tau} \in [0.47,2.88]$. Moreover, two types of sediment bedforms are investigated in order to assess their effect on contaminant transport, namely a macroscopically flat bed and a bed featuring ripples. The analysis of the simulation data shows that the settling velocity of the contaminant particles is enhanced in the ensemble-averaged sense, yet, the time from beginning of the settling until the initial deposition is prolonged when compared to the ratio between the channel height and the terminal velocity. The enhancement is demonstrated to be a result of the preferential sampling of turbulent sweep events, which also implies that the streamwise component of the particle velocity is increased compared to the mean fluid velocity at the same position. A closer examination of the spatial organisation of contaminated particles reveals that they tend to accumulate in large-scale high-speed velocity streaks in the outer region of turbulence. Due to this focusing mechanism, the mean-squared lateral displacement of the settling particles stagnates in the lower half of the channel such that contaminants are not further dispersed in cross-stream direction until shortly before deposition. The same behaviour could be reproduced using a time-invariant exact coherent flow state resembling a hairpin vortex as a proxy for turbulence, and an extended parameter sweep in this setup suggests that this transport barrier effect persists even at high relative turbulence intensities. It is speculated that this phenomenon might confine contaminated particles to a region close to the river bank over a considerable downstream distance in the aftermath of a combined sewer overflow event, which might seriously impact decisions regarding public health measures. Near the sediment bed, the barrier effect of the large-scale motions is inactive and contaminants are found to disperse laterally at a rate which presumably depends on the Shields parameter. The interaction between the sediment and the contaminants is distinct for the two bed topologies under investigation. In the case of macroscopically flat beds, the contaminated particles are transported towards sediment ridges which are in turn known to be a result of the action of large-scale fluid motions, and the mixing of contaminants and sediment particles is restricted to the thin layer of sediment near the interface. In contrast, the presence of ripples leads to a capturing effect where contaminated particles are preferentially deposited in the trough of the ripple, and subsequently buried by a thick layer of sediment due to the propagation of the bed feature. This mechanism temporarily immobilises a large share of all contaminated particles until the displacement of the ripple has sufficiently progressed for them to be eroded on the windward side. During the immobilisation, the associated bacteria are shielded from solar radiation to a substantial degree, which likely has a significant impact on their inactivation, especially in shallow waters. Moreover, the cyclic nature of this phenomenon may provide one of many explanations for bacteria storages which are known to exist in river sediments and may cause bursts in fecal bacteria indicator levels even in absence of immediate contamination events. It is concluded that direct numerical simulation can be a valuable tool for the analysis of bacteria transport, and recommendations are made on how the conjectures compiled in this thesis can be targeted in laboratory experiments to examine their relevance.

The dissertation investigates buoyancy effects in turbulent bluff-body wakes that evolve in stratified fluids. The investigation utilizes high-resolution numerical simulations and employs a body-inclusive approach to describe the flow from the body into the far wake unlike the usual temporal-model approximation of most prior stratified-wake simulations. The dissertation is composed of three main parts. The first part focuses on the dynamics of vorticity that accounts for the unexpected regeneration and increase of turbulence in the near-to-intermediate wake when stratification increases in the regime of low body Froude numbers. The second part characterizes buoyancy effects on the evolution of turbulent kinetic energy in a sphere wake at moderate Froude number and an intermediate Reynolds number. The third part concerns the decay of a disk wake at relatively high Reynolds number and a wide range of Froude numbers, constitutes the major contribution of this thesis, and is summarized below.Large-eddy simulations (LES) of flow past a disk are performed at Re = UbLb/ν = 50,000 and at Fr = Ub/NLb = ∞,50,10,2; Ub is the free-stream velocity, Lb is the disk diameter, ν is the fluid kinematic viscosity, and N is the buoyancy frequency.In the axisymmetric wake in a homogeneous fluid, it is found that the mean streamwise velocity deficit (U0) decays in two stages; U0 ∝ x−0.9 during 10 < x/Lb < 65 followed by U0 ∝∼ x−2/3. Consequently, none of the simulated stratified wakes is able to exhibit the classical 2/3 decay exponent of U0 in the interval before buoyancy effects set in. The turbulent characteristic velocity, taken as K1/2 with K the turbulent kinetic energy (TKE), satisfies K1/2 ∝∼ x−2/3 after x/Lb ≈ 10. Turbulent wakes are affected by stratification within approximately one buoyancy time scale (Ntb ≈ 1) after which, provided that RehFrh2 ≥ 1, we find 3 regimes: weakly stratified turbulence (WST), intermediately stratified turbulence (IST), and strongly stratified turbulence (SST). The regime boundaries are delineated by the turbulent horizontal Froude number Frh = u′h/NLHk; here, u′h and LHk are r.m.s horizontal velocity and TKE- based horizontal wake width. WST begins when Frh decreases to O(1), spans 1 < Ntb < 5 and, while the mean flow is strongly affected by buoyancy in WST, turbulence is not. Thus, while the mean flow transitions into the so-called non-equilibrium (NEQ) regime, turbulence remains approximately isotropic in WST. The next stage of IST, identified by progressively increasing turbulence anisotropy, commences at N tb ≈ 5 once F rh decreases to O(0.1). During IST, the mean flow has arrived into the NEQ regime with a constant decay exponent, U0 ∝ x−0.18, but turbulence is still in transition. The exponent of 0.18 for the disk wake is smaller than the approximately 0.25 exponent found for the stratified sphere wake. When F rh decreases by another order of magnitude to F rh ∼ O(0.01), the wake transitions into the third regime of SST that is identified based on the asymptote of turbulent vertical Froude number (Frv = u′h/Nlv) to a O(1) constant. During SST that commences at Ntb ≈ 20, turbulence is strongly anisotropic (u′z ≪ u′h), and, both u′h and U0 satisfy x−0.18 decay signifying the arrival of the NEQ regime for both turbulence and mean flow. Turbulence is patchy and temporal spectra are broadband in the SST wake.Energy budgets reveal that stratification has a direct and positive influence on the prolongation of wake life. During the WST/early-IST stage, energy budgets show that the mean buoyancy flux acts to augment the MKE before the additional augmentation by reduced turbulent production. On the other hand, during WST/early-IST, the decay of TKE is faster than the unstratified case because of negative buoyancy flux (a sink that serves to increase turbulent potential energy) and increased dissipation and, additionally, also by the reduced production. In the late-IST/early-SST stages, production is enhanced and, additionally, there is injection from turbulent potential energy so that the TKE decay slows down. Only in the SST stage, when NEQ is realized for both the mean and turbulence, the turbulent buoyancy flux becomes negative again, acting as a sink of TKE.