In this work, we carried out direct numerical simulations in large channel domains and studied the kinematics and dynamics of fully localised turbulent bands at Reynolds number $Re=750$. Our results show that the downstream end of the band features fast streak generation and travels into the adjacent laminar flow, whereas streaks at the upstream end decay continually and more slowly. This asymmetry is responsible for the transverse growth of the band. We particularly investigated the mechanism of streak generation at the downstream end, which drives the growth of the band. We identified a spanwise inflectional instability associated with the local mean flow near the downstream end, and our results strongly suggest that this instability is responsible for the streak generation and ultimately for the growth of the band. Based on our study, we propose a possible self-sustaining mechanism of fully localised turbulent bands at low Reynolds numbers in channel flow.

In three-dimensional turbulence there is on average a cascade of kinetic energy from the largest to the smallest scales of the flow. While the dominant idea is that the cascade occurs through the process of vortex stretching, evidence for this is debated. Here we show theoretically and numerically that vortex stretching is in fact not the main contributor to the average cascade. The main contributor is the self-amplification of the strain-rate field, and we provide several arguments for why its role must not be conflated with that of vortex stretching. Numerical results, however, indicate that vortex stretching plays a more important role during fluctuations of the cascade about its average behaviour. We also resolve a paradox regarding the differing role of vortex stretching on the energy cascade and energy dissipation rate dynamics.

Convection in a rapidly rotating cylindrical annulus subject to azimuthal variations in outer boundary heat flux is investigated experimentally. The motivation for this problem stems from the influence of the laterally inhomogeneous lower mantle on the geodynamo. The absence of axial ($z$) gradients of boundary temperature ensures that the condition of quasi-geostrophy, often used to model convection outside the tangent cylinder in spherical shells, is realized in a cylindrical annulus even in strongly driven convection. Experiments are performed with water from below onset of convection to highly supercritical states (measured by the flux Rayleigh number, $Ra\sim 10^{10}$) and for boundary heat flux heterogeneity $q^{\ast }$ (defined by the ratio of the azimuthal variation to the mean boundary heat flux) in the range 0–2. The power requirement for onset of convection reduces substantially with increasing $q^{\ast }$, in line with earlier studies of the onset in rotating spherical shells. For strongly driven convection at $q^{\ast }>1$, the long-time structure is that of localized coherent cyclone–anticyclone vortex pairs, which produce narrow downwellings between them. However, shorter-time averages of the flow reveal the presence of small-scale motions, which may have an important role in magnetic field generation. For a twofold heat flux heterogeneity of $q^{\ast }\approx 2$, convection within the annulus fully homogenizes at ${\sim}30$ times the onset Rayleigh number, and no coherent vortices remain. Finally, the measured heat flux variation on the inner boundary is considerably larger compared with that on the outer boundary, which provides a plausible mechanism for inner-core heterogeneity in the Earth.

We analyse the entrainment and energetics of equal and opposite axisymmetric turbulent air plumes in a vertically confined space at a Rayleigh number of $1.24\times 10^{7}$ using theory and direct numerical simulation. On domains of sufficiently large aspect ratio, the steady state consists of turbulent plumes penetrating an interface between two layers of approximately uniform buoyancy. As described by Baines & Turner (J. Fluid Mech., vol. 37(1), 1969, pp. 51–80), upon penetrating the interface the flow in each plume becomes forced and behaves like a constant-momentum jet, due to a reduction in its mean buoyancy relative to the local environment. To observe the behaviour of the plumes we partition the domain into sub-domains corresponding to each plume. Domains of relatively small aspect ratio produce a single primary mean-flow circulation between the sub-domains that is maintained by entrainment into the plumes. At larger aspect ratios the mean flow between the sub-domains bifurcates, indicating the existence of a secondary circulation within each layer associated with entrainment into the jets. The largest aspect ratios studied here exhibit an additional, tertiary, circulation in the vicinity of the interface. Consistency between independent calculations of an effective entrainment coefficient allows us to identify aspect ratios for which the flow can be modelled using plume theory, under the assumption of a two-layer stratification. To study the flow’s energetics we use a local definition of available potential energy (APE). For plumes with Gaussian velocity and buoyancy profiles, the theory we develop suggests that the kinetic energy dissipation is split equally between the jets and the plumes and, collectively, accounts for almost half of the input of APE at the boundaries. In contrast, 1/4 of the APE dissipation and background potential energy (BPE) production occurs in the jets, with the remaining 3/4 occurring in the plumes. These bulk theoretical predictions agree with observations of BPE production from simulations to within 1 % and form the basis of a similarity solution that models the vertical dependence of APE dissipation and BPE production. Unlike results concerning the dissipation of buoyancy variance and the strength of the circulations described above, the model for the flow’s energetics does not involve an entrainment coefficient.

Experiments were carried out in high Reynolds number turbulent boundary layers over rough surfaces of diverse geometries. Roughness configurations varied in element height, distribution (random versus ordered), shape and spacing. Rough surfaces comprising of two superposed element geometries were also tested. All flows were free of transitional effects with $Re_{\unicode[STIX]{x1D703}}$ upwards of 40 000 and $\unicode[STIX]{x1D6FF}/k_{g}$ ratios above 73. The wall-pressure spectrum and turbulent velocity profiles revealed that the roughness element spacing has the greatest impact on the turbulent structures in the boundary layer. The high-frequency scaling on shear friction velocity, $U_{\unicode[STIX]{x1D708}}$, (Meyers et al.J. Fluid Mech., vol. 768, 2015, pp. 261–293) was validated and $U_{\unicode[STIX]{x1D708}}$ was shown to be the viscous contribution to the overall surface drag. An empirical formula for the pressure drag on roughness elements was developed to reflect the finding that the pressure drag is a function of only two variables: sparseness ratio $(\unicode[STIX]{x1D706})$ and roughness Reynolds number $(k_{g}^{+})$. Results also suggest that the viscous contribution to drag approaches a constant non-zero value at high Reynolds numbers, and ‘fully rough-wall flow’ may occur at higher $k_{g}^{+}$ than previously thought.

Depth-integrated mathematical models for simulating waves and currents from deep to shallow water are presented. These models are derived from Euler’s equations in the $\unicode[STIX]{x1D70E}$-coordinate system, mapping the total water depth in Cartesian coordinates onto a specified range of $\unicode[STIX]{x1D70E}$-coordinates. The horizontal velocity is approximated as a truncated infinite series of products of prescribed shape functions of $\unicode[STIX]{x1D70E}$ and unknown functions of horizontal coordinates and time. Adopting the method of weighted residuals, the new models are obtained by minimizing the residuals of the horizontal momentum equations with either the Galerkin method or the subdomain method. These models’ linear and nonlinear water wave properties are investigated. The new models are implemented numerically. A hierarchy of numerical models with different degree of polynomial approximation is developed and checked against several benchmarked experiments and a new set of experiments of self-focusing wave groups. For both the Galerkin and subdomain models, excellent agreements are observed for both the free surface elevations and the velocity profiles. The new models are superior to the existing Boussinesq-type models for their applicability to a wide range of physical scenarios, including the interactions between a wave package of multiple frequency components and a linearly sheared current. The new Galerkin models have similar characteristics and accuracy as the Green–Naghdi models, but the new models are more efficient computationally. Finally, for the same degree of polynomial approximation the subdomain models perform better than the Galerkin models and require less computational time.

Helical vortices have been studied for more than a century to understand basic aspects of fluid motion. Helical vortices appear both in nature, e.g. as tornadoes, and in many industrial applications associated with mixing and in wakes behind rotors. Owing to the complexity of the equations governing the self-induced motion of helical vortices, it has up to now not been possible to obtain closed-form solutions describing all aspects of the motion. An important issue concerns the difference between the self-induced motion of the helical structure and the movement of fluid particles located on the helix. Here, we revisit the equations governing both the motion of the helical vortex structure and the motion of material fluid elements on the axis of the helix, and for both cases derive closed-form solutions for the resulting velocities. As a part of the paper, we also devise potential applications of the achieved knowledge.

The present experimental study addresses the flow of a yield stress fluid with some elasticity (Carbopol gel) in a square duct. The behaviour of two fluids with lower and higher yield stress is investigated in terms of the friction factor and flow velocities at multiple Reynolds numbers $Re^{\ast }\in$ (1, 200) and, hence, Bingham numbers $Bi\in$ (0.01, 0.35). Taking advantage of the symmetry planes in a square duct, we reconstruct the entire 3-component velocity field from two-dimensional particle image velocimetry (PIV). A secondary flow consisting of eight vortices is observed to recirculate the fluid from the core towards the wall centre and from the corners back to the core. The extent and intensity of these vortices grows with increasing $Re^{\ast }$ or, alternately, as the plug size decreases. The second objective of this study is to explore the change in flow in the presence of particles. To this end, almost neutrally buoyant finite-size spherical particles with a duct height, $2H$, to particle diameter, $d_{p}$, ratio of 12 are used at two volume fractions $\unicode[STIX]{x1D719}=5$ and 10 %. Particle tracking velocimetry is used to measure the velocity of these refractive-index-matched spheres in the clear Carbopol gel, and PIV to extract the fluid velocity. Additionally, simple shadowgraphy is also used to qualitatively visualise the development of the particle distribution along the streamwise direction. The particle distribution pattern changes from being concentrated at the four corners, at low flow rates, to being focussed along a diffused ring between the centre and the corners, at high flow rates. The presence of particles induces streamwise and wall-normal velocity fluctuations in the fluid phase; however, the primary Reynolds shear stress is still very small compared to turbulent flows. The size of the plug in the particle-laden cases appears to be smaller than the corresponding single-phase cases. Similar to Newtonian fluids, the friction factor increases due to the presence of particles, almost independently of the suspending fluid matrix. Interestingly, predictions based on an increased effective suspension viscosity agrees quite well with the experimental friction factor for the concentrations used in this study.

The stability of forced planar liquid jets in a still gaseous environment is explored using nonlinear simulation and spatial linear stability analysis. Harmonic modulation of the transverse component of the inlet velocity leads to an excitation of sinuous modes in the jet. Two forcing amplitudes, 1 % and 5 %, are investigated. While for 1 % forcing, the interfacial disturbance retains a sinuous shape throughout the domain, for 5 % forcing, an increasing downstream deviation from the sinuous wave shape is found. Both forcings lead to sufficient mean flow correction to render linear stability analysis on a base flow unfeasible. Hence, an analysis on the time-averaged mean flow is performed. A correction scheme is introduced, to account for the spreading of the interface position in the mean flow. Comparison of eigenfunctions and growth rates with their counterparts from the nonlinear simulation shows an excellent agreement for 1 % forcing. For 5 % forcing, agreement of the eigenfunctions deteriorates significantly and growth rates are falsely predicted, resulting in a breakdown of the stability model. Subsequent analysis reveals a strong interaction of the fundamental wave with the second higher harmonic wave for 5 % forcing and a reversed energy flow from the coherent motion to the mean flow. These findings provide an explanation for the failure of the linear stability model for large forcing amplitudes. The study extends the applicability of mean flow stability analysis to convectively unstable planar liquid/gas jets and supports previous findings on the limits of mean flow stability, involving pronounced influence of higher harmonic modes.

The characteristics of turbulent boundary layers (TBLs) subjected to adverse pressure gradients are analysed through well-resolved large-eddy simulations. The geometries under study are the NACA0012 and NACA4412 wing sections, at $0^{\circ }$ and $5^{\circ }$ angle of attack, respectively, both of them at a Reynolds number based on inflow velocity and chord length of $Re_{c}=400\,000$. The turbulence statistics show that adverse pressure gradients (APGs) have a significant effect on the mean velocity, velocity fluctuations and turbulent kinetic energy budget, and this effect is more prominent on the outer region of the boundary layer. Furthermore, the effect of flow history is assessed by means of an integrated Clauser pressure-gradient parameter $\overline{\unicode[STIX]{x1D6FD}}$ (Vinuesa et al., Flow Turbul. Combust., vol. 99, 2017, pp. 565–587), through the study of cases with matching local values of $\unicode[STIX]{x1D6FD}$ and the friction Reynolds number $Re_{\unicode[STIX]{x1D70F}}$ to isolate this effect. Our results show a noticeable effect of the flow history on the outer region, however the differences in the near-wall peak of the tangential velocity fluctuations appear to be mostly produced by the local APG magnitude. The one-dimensional power-spectral density shows energetic small scales in the outer region of APG TBLs, whereas these energetic scales do not appear in zero-pressure-gradient (ZPG) TBLs, suggesting that small scales near the wall are advected towards the outer layer by the APG. Moreover, the linear coherence spectra show that the spectral outer peak of high-Reynolds-number ZPG TBLs is highly correlated with the near-wall region (Baars et al., J. Fluid Mech., vol. 823, 2017, R2), unlike APG TBLs which do not show such a correlation. This result, together with the different two-dimensional spectra of APG and high-Reynolds-number ZPG TBLs, suggests different energisation mechanisms due to APG and increase in Reynolds number. To the authors’ knowledge, this is the first in-depth analysis of the TBL characteristics over wings, including detailed single-point statistics, spectra and coherence.

Data from direct numerical simulations (DNS) of disperse bubbly flow in an upward vertical channel are used to develop a new second-moment closure for bubble-induced turbulence (BIT) in the Euler–Euler framework. The closure is an extension of a BIT model originally proposed by Ma et al. (Phys. Rev. Fluids, vol. 2, 2017, 034301) for two-equation eddy-viscosity models and focuses on the core region of the channel, where the interfacial term and dissipation term are in balance. Particular attention in this study is given to the treatment of the pressure–strain term for bubbly flows and the form of the interfacial term to account for BIT. For the latter, the concept of an effective BIT source is proposed, which leads to a significant simplification of the modelling work for both the pressure–strain correlation and the interfacial term itself. The anisotropy of bubbly flow is analysed with the aid of the anisotropy-invariant map obtained from the DNS data, and a parameter governing this issue is established. The complete second-moment closure is tested against reference data for different bubbly channel flows and a case of a bubble column. The agreement achieved with the DNS data is very good and the performance of the new model is better than obtained with the standard procedure. Furthermore, the model is shown to be robust and to fulfil the requirements of realizability.

This paper presents a theoretical and experimental study of the long-standing fluid mechanics problem involving the temporal resolution of a large localised initial disturbance into a sequence of solitary waves. This problem is of fundamental importance in a range of applications, including tsunami and internal ocean wave modelling. This study is performed in the context of the viscous fluid conduit system – the driven, cylindrical, free interface between two miscible Stokes fluids with high viscosity contrast. Owing to buoyancy-induced nonlinear self-steepening balanced by stress-induced interfacial dispersion, the disturbance evolves into a slowly modulated wavetrain and further into a sequence of solitary waves. An extension of Whitham modulation theory, termed the solitary wave resolution method, is used to resolve the fission of an initial disturbance into solitary waves. The developed theory predicts the relationship between the initial disturbance’s profile, the number of emergent solitary waves and their amplitude distribution, quantifying an extension of the well-known soliton resolution conjecture from integrable systems to non-integrable systems that often provide a more accurate modelling of physical systems. The theoretical predictions for the fluid conduit system are confirmed both numerically and experimentally. The number of observed solitary waves is consistently within one to two waves of the prediction, and the amplitude distribution shows remarkable agreement. Universal properties of solitary wave fission in other fluid dynamics problems are identified.

The effects of flow topology on the subgrid-scale (SGS) kinetic energy flux in compressible isotropic turbulence is studied. The eight flow topological types based on the three invariants of the filtered velocity gradient tensor are analysed at different scales, along with their roles in the magnitude and direction of kinetic energy transfer. The unstable focus/compressing (UFC), unstable node/saddle/saddle (UN/S/S) and stable focus/stretching (SFS), are the three predominant topological types at all scales; they account for at least 75 % of the flow domain. The UN/S/S and SFS types make major contributions to the average SGS flux of the kinetic energy from large scales to small scales in the inertial range. The unstable focus/stretching (UFS) topology makes a contribution to the reverse SGS flux of kinetic energy from small scales to large scales. In strong compression regions, the average contribution of the stable node/saddle/saddle (SN/S/S) topology to the SGS kinetic energy flux is positive and is predominant over those of other flow topologies. In strong expansion regions, the UFS topology makes a major contribution to the reverse SGS flux of the kinetic energy. As the turbulent Mach number increases, the increase of volume fraction of the UFS topological regions leads to the increase of the SGS backscatter of kinetic energy. The SN/S/S topology makes a dominant contribution to the direct SGS flux of the compressible component of the kinetic energy, while the UFS topology makes a dominant contribution to the reverse SGS flux of the compressible component of the kinetic energy.

In this paper, we study the forcing of baroclinic critical levels, which arise in stratified fluids with horizontal shear flow along the surfaces where the phase speed of a wave relative to the mean flow matches a natural internal wave speed. Linear theory predicts the baroclinic critical-layer dynamics is similar to that of a classical critical layer, characterized by the secular growth of flow perturbations over a region of decreasing width. By using matched asymptotic expansions, we construct a nonlinear baroclinic critical layer theory to study how the flow perturbations evolve once they enter the nonlinear regime. A key feature of the theory is that, because the location of the baroclinic critical layer is determined by the streamwise wavenumber, the nonlinear dynamics filters out harmonics and the modification to the mean flow controls the evolution. At late times, we show that the vorticity begins to focus into yet smaller regions whose width decreases exponentially with time, and that the addition of dissipative effects can arrest this focussing to create a drifting coherent structure. Jet-like defects in the mean horizontal velocity are the main outcome of the critical-layer dynamics.

The transition to turbulence in pipe flow proceeds through several distinct stages, eventually producing aggressively expanding regions of fluctuations, ‘slugs’, surrounded by laminar flow. By examining mean-velocity profiles, fluctuating-velocity profiles and Reynolds stress profiles, the seminal study of Wygnanski & Champagne (J. Fluid Mech., vol. 59 (2), 1973, 281–335) concluded that the flow inside slugs is ‘identical’ to fully turbulent flow. Although this conclusion is widely accepted, upon closer examination of their analysis, we find that their data cannot be used to substantiate this conclusion. We resolve this conflict via new experiments and simulations wherein we pair slugs and fully turbulent flow at the same value of Reynolds number ($Re$). We conclude that the flow inside a slug is indeed indistinguishable from a fully turbulent flow but only when the two flows share the same value of $Re$. Our work highlights the rich $Re$-dependence of transitional pipe flows.

The nucleation of vapour bubbles in stretched or overheated (metastable) liquids is a complex phenomenon with a wide spectrum of applications. Several models, with different levels of detail, have been proposed to predict the key features of bubble dynamics from bubble formation up to its growth, transport and deformation. Most of them focus separately on a few of these aspects. Here, we present a thorough model based on an isothermal diffuse interface description of a two-phase liquid–vapour system endowed with thermal fluctuations, exploiting Landau and Lifshitz’s fluctuating hydrodynamic theory. The stochastic forcing allows for the spontaneous appearance of vapour clusters inside the liquid; the diffuse interface approach provides the hydrodynamic description of the subsequent growth and transport dynamics. In this work we focus on a coarse-grained version of this model, obtained through the averaging of the complete three-dimensional equations on spherical shells: the resulting stochastic equations will spatially depend on the radial distance from the vapour cluster centre. The numerical simulations give access to the mean first passage time, i.e. the time until, on average, the formation of a supercritical bubble. A rough estimate shows that the computational effort is reduced by four orders of magnitude with respect to brute-force atomistic simulations and by two orders of magnitude with respect to the full three-dimensional fluctuating model. The simulations extend up to the very long time scales, allowing us to analyse inertially driven bubble oscillations in confined systems with perfect agreement with available theoretical predictions.

Highly turbulent Taylor–Couette flow with spanwise-varying roughness is investigated experimentally and numerically (direct numerical simulations with an immersed boundary method) to determine the effects of the spacing and spanwise width $s$ of the spanwise-varying roughness on the total drag and on the flow structures. We apply sandgrain roughness, in the form of alternating rough and smooth bands to the inner cylinder. Numerically, the Taylor number is $O(10^{9})$ and the roughness width is varied in the range $0.47\leqslant \tilde{s}=s/d\leqslant 1.23$, where $d$ is the gap width. Experimentally, we explore $Ta=O(10^{12})$ and $0.61\leqslant \tilde{s}\leqslant 3.74$. For both approaches the radius ratio is fixed at $\unicode[STIX]{x1D702}=r_{i}/r_{o}=0.716$, with $r_{i}$ and $r_{o}$ the radius of the inner and outer cylinder respectively. We present how the global transport properties and the local flow structures depend on the boundary conditions set by the roughness spacing $\tilde{s}$. Both numerically and experimentally, we find a maximum in the angular momentum transport as a function of $\tilde{s}$. This can be attributed to the re-arrangement of the large-scale structures triggered by the presence of the rough stripes, leading to correspondingly large-scale turbulent vortices.

We consider a strongly nonlinear long wave model for large amplitude internal waves in a three-layer flow between two rigid boundaries. The model extends the two-layer Miyata–Choi–Camassa (MCC) model (Miyata, Proceedings of the IUTAM Symposium on Nonlinear Water Waves, eds. H. Horikawa & H. Maruo, 1988, pp. 399–406; Choi & Camassa, J. Fluid Mech., vol. 396, 1999, pp. 1–36) and is able to describe the propagation of long internal waves of both the first and second baroclinic modes. Solitary-wave solutions of the model are shown to be governed by a Hamiltonian system with two degrees of freedom. Emphasis is given to the solitary waves of the second baroclinic mode (mode 2) and their strongly nonlinear characteristics that fail to be captured by weakly nonlinear models. In certain asymptotic limits relevant to oceanic applications and previous laboratory experiments, it is shown that large amplitude mode-2 waves with single-hump profiles can be described by the solitary-wave solutions of the MCC model, originally developed for mode-1 waves in a two-layer system. In other cases, however, e.g. when the density stratification is weak and the density transition layer is thin, the richness of the dynamical system with two degrees of freedom becomes apparent and new classes of mode-2 solitary-wave solutions of large amplitudes, characterized by multi-humped wave profiles, can be found. In contrast with the classical solitary-wave solutions described by the MCC equation, such multi-humped solutions cannot exist for a continuum set of wave speeds for a given layer configuration. Our analytical predictions based on asymptotic theory are then corroborated by a numerical study of the original Hamiltonian system.

We develop a method to estimate space–time flow statistics from a limited set of known data. While previous work has focused on modelling spatial or temporal statistics independently, space–time statistics carry fundamental information about the physics and coherent motions of the flow and provide a starting point for low-order modelling and flow control efforts. The method is derived using a statistical interpretation of resolvent analysis. The central idea of our approach is to use known data to infer the statistics of the nonlinear terms that constitute a forcing on the linearized Navier–Stokes equations, which in turn imply values for the remaining unknown flow statistics through application of the resolvent operator. Rather than making an a priori assumption that the flow is dominated by the leading singular mode of the resolvent operator, as in some previous approaches, our method allows the known input data to select the most relevant portions of the resolvent operator for describing the data, making it well suited for high-rank turbulent flows. We demonstrate the predictive capabilities of the method, which we call resolvent-based estimation, using two examples: the Ginzburg–Landau equation, which serves as a convenient model for a convectively unstable flow, and a turbulent channel flow at low Reynolds number.

Superhydrophobic surfaces (SHSs) have the potential to reduce drag at solid boundaries. However, multiple independent studies have recently shown that small amounts of surfactant, naturally present in the environment, can induce Marangoni forces that increase drag, at least in the laminar regime. To obtain accurate drag predictions, one must solve the mass, momentum, bulk surfactant and interfacial surfactant conservation equations. This requires expensive simulations, thus preventing surfactant from being widely considered in SHS studies. To address this issue, we propose a theory for steady, pressure-driven, laminar, two-dimensional flow in a periodic SHS channel with soluble surfactant. We linearize the coupling between flow and surfactant, under the assumption of small concentration, finding a scaling prediction for the local slip length. To obtain the drag reduction and interfacial shear, we find a series solution for the velocity field by assuming Stokes flow in the bulk and uniform interfacial shear. We find how the slip and drag depend on the nine dimensionless groups that together characterize the surfactant transport near SHSs, the gas fraction and the normalized interface length. Our model agrees with numerical simulations spanning orders of magnitude in each dimensionless group. The simulations also provide the constants in the scaling theory. Our model significantly improves predictions relative to a surfactant-free one, which can otherwise overestimate slip and underestimate drag by several orders of magnitude. Our slip length model can provide the boundary condition in other simulations, thereby accounting for surfactant effects without having to solve the full problem.