Over the past forty or so years, considerable advances have been made in our understanding of the effects of ocean waves on sea ice, and vice versa, with observations, experiments and theory all playing their part. Recent years have seen the development of ever more sophisticated mathematical models designed to represent the physics more accurately and incorporate new features. What is lacking is an approach to three-dimensional scattering for ice floes that is both accurate and efficient enough to be used as a component in a theory designed to model the passage of directional wave spectra through the marginal ice zone. Bennetts & Williams (J. Fluid Mech., 2010, this issue, vol. 662, pp. 5–35) have brought together a number of solution techniques honed on simpler problems to provide just such a component.

An efficient solution method is presented for linear and time-harmonic water-wave scattering by two classes of a three-dimensional hydroelastic system. In both cases, the fluid domain is of infinite horizontal extent and finite depth. The fluid surface is either open, except in a finite region where it is covered by a thin-elastic plate, which represents an ice floe, or fully covered by a plate, except in a finite region where it is open, which represents an ice polynya. The approach outlined herein permits the boundary between the ice-covered and free-surface fluid regions to be described by an arbitrary smooth curve. To solve the governing equations of the full three-dimensional linear problem, they are first projected onto the horizontal plane by using an approximation theory that combines an expansion of the vertical motion of the fluid in a finite set of judiciously chosen modes with a variational principle. This generates a system of two-dimensional partial differential equations that are converted into a set of one-dimensional integro-differential equations using matrices of Green's functions, which are solved numerically through an application of the Galerkin technique. A numerical results section justifies the consideration of an arbitrarily shaped boundary by comparing the response of differently shaped floes and polynyas over a range of relevant wavenumbers. Comparisons are made in terms of the magnitude and direction of the far-field scattering response, and also the maximum average curvature of the floe and the maximum wave elevation within the polynya.

An empirical model reduction method is performed on nonlinear transient convection patterns near the threshold, using a pseudo-inverse-based projective method called the pseudo-balanced proper orthogonal decomposition (PBPOD). These transient patterns are large-scale amplitude/phase modulations in convection rolls, obtained by prescribing selected spatial input-shape functions. For the nonlinear convection patterns modelled, PBPOD appears to be very effective. Using the nonlinear front example, PBPOD is compared with other existing methods, such as POD and linearized BPOD. The limitations of the methods are discussed. Using a complex prescribed input, complex disturbances are generated and the outputs of the open-loop responses are compared between the original and the reduced-order models. The agreement is good. A feedback-control study is performed using the nonlinear front example. The controller is built by the pseudo-balanced reduced-order model, with a low-order nonlinear estimator. Closed-loop simulations show that the nonlinear travelling fronts can be effectively damped out by the feedback-control actions.

The nonlinear dynamics of uncharged ideally polarizable spheres freely suspended in a viscous electrolyte in a uniform electric field are analysed using theory and numerical simulations. When a sphere polarizes under the action of the field, it acquires a non-uniform surface charge, which results in an electro-osmotic flow near its surface that scales quadratically with the applied field magnitude. While this so-called induced-charge electrophoresis yields no net motion in the case of a single sphere, it can drive relative motions by symmetry breaking when several particles are present. In addition, Maxwell stresses in the fluid also result in non-zero dielectrophoretic forces, which also cause particle motions. The combination of these two nonlinear electrokinetic effects, termed dipolophoresis, is analysed in detail by using numerical simulations. An efficient simulation method based on our previous analysis of pair interactions is presented and accounts for both far-field and near-field electric and hydrodynamic interactions in the thin-Debye-layer limit, as well as steric interactions using a novel contact algorithm. Simulation results in large-scale suspensions with periodic boundary conditions are presented. While the dynamics under dielectrophoresis alone are shown to be characterized by particle chaining along the field direction, in agreement with previous investigations, chaining is not found to occur under dipolophoresis, which instead causes transient particle pairings and results in a non-uniform microstructure with large number of density fluctuations, as we demonstrate by calculating pair distribution functions and particle occupancy statistics. Dipolophoresis is also found to result in significant hydrodynamic dispersion and velocity fluctuations, and the dependence of these two effects on suspension volume fraction is investigated.

Using helical-wave decomposition (HWD), a solenoidal vector field can be decomposed into helical modes with different wavenumbers and polarities. Here, we first review the general formulation of HWD in an arbitrary single-connected domain, along with some new development. We then apply the theory to a viscous incompressible turbulent channel flow with system rotation, including a derivation of helical bases for a channel domain. By these helical bases, we construct the inviscid inertial-wave (IW) solutions in a rotating channel and derive their existing condition. The condition determines the specific wavenumber and polarity of the IW. For a set of channel turbulent flows rotating about a streamwise axis, this channel-domain HWD is used to decompose the flow data obtained by direct numerical simulation. The numerical results indicate that the streamwise rotation induces a polarity-asymmetry and concentrates the fluctuating energy to particular helical modes. At large rotation rates, the energy spectra of opposite polarities exhibit different scaling laws. The nonlinear energy transfer between different helical modes is also discussed. Further investigation reveals that the IWs do exist when the streamwise rotation is strong enough, for which the theoretical predictions and numerical results are in perfect agreement in the core region. The wavenumber and polarity of the IW coincide with that of the most energetic helical modes in the energy spectra. The flow visualizations show that away from the channel walls, the small vortical structures are clustered to form very long columns, which move in the wall-parallel plane and serve as the carrier of the IW. These discoveries also help clarify certain puzzling problems raised in previous studies of streamwise-rotating channel turbulence.

A recently developed particle-level numerical method is used to simulate flexible fibre suspensions in Newtonian simple shear flow. In this method, the flow is computed on a fixed regular ‘lattice’ using the lattice Boltzmann method, where each solid particle, or fibre in this case, is mapped onto a Lagrangian frame moving continuously through the domain. The motion and orientation of the fibre are obtained from Newtonian dynamics equations. The effect of fibre stiffness on the rheology of flexible fibre suspensions is investigated and a relation for the relative viscosity is obtained. We show that fibre stiffness (bending ratio, BR) has a strong impact on rheology in the range BR < 3. The relative viscosity increases significantly as BR decreases. These results show that the primary normal stress difference has a minimum value at BR ~ 1. The primary normal stress difference for slightly deformable fibres reaches a minimum and increases significantly as BR decreases below one. The results are explained based on Batchelor's relation for non-Brownian suspensions.

We introduce a new numerical method for the study of scalar mixing in two-dimensional advection fields. The position of an advected material strip is computed kinematically, and the associated convection–diffusion problem is solved using the computed local stretching rate along the strip, assuming that the diffusing strip thickness is smaller than its local radius of curvature. This widely legitimate assumption reduces the numerical problem to the computation of a single variable along the strip, thus making the method extremely fast and applicable to any large Péclet number. The method is then used to document the mixing properties of a chaotic sine flow, for which we relate the global quantities (spectra, concentration probability distribution functions (PDFs), increments) to the distributed stretching of the strip convoluted by the flow, possibly overlapping with itself. The numerical results indicate that the PDF of the strip elongation is log normal, a signature of random multiplicative processes. This property leads to exact analytical predictions for the spectrum of the field and for the PDF of the scalar concentration of a solitary strip. The present simulations offer a unique way of discovering the interaction rule for building complex mixtures which are made of a random superposition of overlapping strips leading to concentration PDFs stable by self-convolution.

In this study, a comprehensive inviscid temporal stability analysis of a compressible round jet is performed for Mach numbers ranging from 1 to 10. We show that in addition to the Kelvin–Helmholtz instability modes, there exist for each azimuthal wavenumber three other types of modes (counterflow subsonic waves, subsonic waves and supersonic waves) whose characteristics are analysed in detail using a WKBJ theory in the limit of large axial wavenumber. The theory is constructed for any velocity and temperature profile. It provides the phase velocity and the spatial structure of the modes and describes qualitatively the effects of base-flow modifications on the mode characteristics. The theoretical predictions are compared with numerical results obtained for an hyperbolic tangent model and a good agreement is demonstrated. The results are also discussed in the context of jet noise. We show how the theory can be used to determine a priori the impact of jet modifications on the noise induced by instability.

It has long been recognized that the osmotic transport characteristics of membranes may be strongly influenced by the presence of unstirred concentration boundary layers adjacent to the membrane. Previous experimental as well as theoretical works have mainly focused on the case where the solutions on both sides of the membrane remain well mixed due to an external stirring mechanism. We investigate the effects of concentration boundary layers on the efficiency of osmotic pumping processes in the absence of external stirring, i.e. when all advection is provided by the osmosis itself. This case is relevant in the study of intracellular flows, e.g. in plants. For such systems, we show that no well-defined boundary-layer thickness exists and that the reduction in concentration can be estimated by a surprisingly simple mathematical relation across a wide range of geometries and Péclet numbers.

Deformation-induced lateral migration of a bubble slowly rising near a vertical plane wall in a stagnant liquid is numerically and theoretically investigated. In particular, our focus is set on a situation with a short clearance c between the bubble interface and the wall. Motivated by the fact that numerically and experimentally measured migration velocities are considerably higher than the velocity estimated by the available analytical solution using the Faxén mirror image technique for a/(a + c) ≪ 1 (here a is the bubble radius), when the clearance parameter ϵ(=c/a) is comparable to or smaller than unity, the numerical analysis based on the boundary-fitted finite-difference approach solving the Stokes equation is performed to complement the experiment. The migration velocity is found to be more affected by the high-order deformation modes with decreasing ϵ. The numerical simulations are compared with a theoretical migration velocity obtained from a lubrication study of a nearly spherical drop, which describes the role of the squeezing flow within the bubble–wall gap. The numerical and lubrication analyses consistently demonstrate that when ϵ ≤ 1, the lubrication effect makes the migration velocity asymptotically μVB12/(25ϵγ) (here, VB1, μ and γ denote the rising velocity, the dynamic viscosity of liquid and the surface tension, respectively).

The flow over a circular cylinder controlled by a two-dimensional synthetic jet positioned at the mean rear stagnation point has been experimentally investigated in a water channel at the cylinder Reynolds number Re = 950. This is an innovative arrangement and the particle-image-velocimetry measurement indicates that it can lead to a novel and interesting phenomenon. The synthetic-jet vortex pairs induced near the exit convect downstream and interact with the vorticity shear layers behind both sides of the cylinder, resulting in the formation of new induced wake vortices. The present vortex synchronization occurs when the excitation frequency of the synthetic jet is between 1.67 and 5.00 times the natural shedding frequency at the dimensionless stroke length 99.5. However, it is suggested that the strength of the synthetic-jet vortex pair plays a more essential role in the occurrence of vortex synchronization than the excitation frequency. In addition, the wake-vortex shedding is converted into a symmetric mode from its original antisymmetric mode. The symmetric shedding mode weakens the interaction between the upper and lower wake vortices, resulting in a decrease in the turbulent kinetic energy produced by them. It also has a significant influence on the global flow field, including the velocity fluctuations, Reynolds stresses and flow topology. However, their distributions are still dominated by the large-scale coherent structures.

We study both experimentally and numerically the steady zonal flow generated by longitudinal librations of a spherical rotating container. This study follows the recent weakly nonlinear analysis of Busse (J. Fluid Mech., vol. 650, 2010, pp. 505–512), developed in the limit of small libration frequency–rotation rate ratio and large libration frequency–spin-up time product. Using particle image velocimetry measurements as well as results from axisymmetric numerical simulations, we confirm quantitatively the main features of Busse's analytical solution: the zonal flow takes the form of a retrograde solid-body rotation in the fluid interior, which does not depend on the libration frequency nor on the Ekman number, and which varies as the square of the amplitude of excitation. We also report the presence of an unpredicted prograde flow at the equator near the outer wall.

Various studies have suggested that the no-slip condition may not hold for Newtonian liquids flowing over (for the most part) non-wetting surfaces. This paper describes an experimental study of steady Poiseuille flow at various Reynolds numbers up to 0.12 of four different aqueous monovalent electrolyte solutions through naturally hydrophilic and hydrophobically coated fused-silica channels with a depth of 33 μm. The slip lengths for these flows were estimated using a local method based on a new particle velocimetry technique that determines velocities at three different wall-normal distances within the first 400 nm next to the wall. These results are corrected using direct measurements of the near-wall particle distribution, which is highly non-uniform as expected due to repulsive electric double-layer interactions between the 100 nm tracer particles and the wall. In all cases, the slip lengths were not more than 23 nm and for all but one case, zero within their uncertainties. As illustrated here, the standard assumption of uniformly distributed tracers can significantly increase slip length estimates obtained using local methods and near-wall velocity data.

We consider laminar high-Reynolds-number flow through a finite-length planar channel, where a portion of one wall is replaced by a thin massless elastic membrane that is held under longitudinal tension T and subject to a linear external pressure distribution. The flow is driven by a fixed pressure drop along the full length of the channel. We investigate the global stability of two-dimensional Poiseuille flow using a method of matched local eigenfunction expansions, which is compared to direct numerical simulations. We trace the neutral stability curve of the primary oscillatory instability of the system, illustrating a transition from high-frequency ‘sloshing’ oscillations at high T to vigorous ‘slamming’ motion at low T. Small-amplitude sloshing at high T can be captured using a low-order eigenmode truncation involving four surface-based modes in the compliant segment of the channel coupled to Womersley flow in the rigid segments. At lower tensions, we show that hydrodynamic modes increasingly contribute to the global instability, and we demonstrate a change in the mechanism of energy transfer from the mean flow, with viscous effects being destabilizing. Simulations of finite-amplitude oscillations at low T reveal a generic slamming motion, in which the flexible membrane is drawn close to the opposite rigid wall before recovering rapidly. A simple model is used to demonstrate how fluid inertia in the downstream rigid channel segment, coupled to membrane curvature downstream of the moving constriction, together control slamming dynamics.

The effects of a uniform axial magnetic field directed towards an oscillating wall in a semi-infinite viscous fluid (or Stokes layer) is investigated. The linear stability and disturbance characteristics are determined using both Floquet theory and via direct numerical simulations. Neutral stability curves and critical parameters for instability are presented for a range of magnetic field strengths. Results indicate that a magnetic field directed towards the boundary wall is stabilizing, which is consistent with that found in many steady flows.

Injection of carbon dioxide (CO2) into geological formations is widely regarded as a promising tool for reducing global atmospheric CO2 emissions. To evaluate injection scenarios, estimate reservoir capacity and assess leakage risks, an accurate understanding of the subsurface spreading and migration of the plume of mobile CO2 is essential. Here, we present a complete solution to a theoretical model for the subsurface migration of a plume of CO2 due to natural groundwater flow and aquifer slope, and subject to residual trapping. The results show that the interplay of these effects leads to non-trivial behaviour in terms of trapping efficiency. The analytical nature of the solution offers insight into the physics of CO2 migration, and allows for rapid, basin-specific capacity estimation. We use the solution to explore the parameter space via the storage efficiency, a macroscopic measure of plume migration. In a future study, we shall incorporate CO2 dissolution into the migration model and study the importance of dissolution relative to capillary trapping and the impact of dissolution on the storage efficiency.

In this study, we investigate the dynamics of a freely rising and falling cylinder. This is, in essence, a vortex-induced vibration (VIV) system comprising both transverse (Y) and streamwise (X) degrees-of-freedom (d.o.f.), but with zero spring stiffness and zero damping. This problem represents a limiting case among studies in VIV, and is an extension of recent research of elastically mounted bodies having very low spring stiffness, as well as bodies with very low mass and damping. We find that if the mass ratio (where m* = cylinder mass/displaced fluid mass) is greater than a critical value, m*crit = 0.545, the body falls or rises with a rectilinear trajectory. As the mass ratio is reduced below m*crit = 0.545, the cylinder suddenly begins to vibrate vigorously and periodically, with a 2P mode of vortex formation, as reported in the preliminary study of Horowitz & Williamson (J. Fluids Struct. vol. 22, 2006, pp. 837–843). The similarity in critical mass between freely rising and elastically mounted bodies is unexpected, as it is known that the addition of streamwise vibration can markedly affect the response and vortex formation in elastically mounted systems, which would be expected to modify the critical mass. However, we show in this paper that the similarity in vortex formation mode (2P) between the freely rising body and the elastically mounted counterpart is consistent with a comparable phase of vortex dynamics, strength of vortices, amplitudes and frequencies of motion and effective added mass (CEA). All of these similarities result in comparable values of critical mass. The principal fact that the 2P mode is observed for the freely rising body is an interesting and consistent result; based on the previous VIV measurements, this is the only mode out of the known set {2S, 2P, 2T} to yield negative effective added mass (CEA < 0), which is a condition for vibration of a freely rising body. In this paper, we deduce that there exists only one possible two degree-of-freedom elastically mounted cylinder system, which can be used to predict the dynamics of freely rising bodies. Because of the symmetry of the vortex wake, this system is one for which the natural frequencies are fNX = 2fNY. Although this seems clear in retrospect, previous attempts to predict critical mass did not take this into account. Implementing such an elastic system, we are able to predict vibration amplitudes and critical mass (m*crit = 0.57) for a freely rising cylinder in reasonable agreement with direct measurements for such a rising body, and even to predict the Lissajous figures representing the streamwise–transverse vibrations for a rising body with very small mass ratios (down to m* = 0.06), unobtainable from our direct measurements.

The formation and evolution of double-diffusive interleaving is experimentally investigated with the purpose of analysing the influence of the convective flow structures, at different scales, on the mean flow. Recently, Krishnamurti (J. Fluid Mech., vol. 558, 2006, p. 113) has shown that, in the case of a continuous stratification experiment, the Reynolds stresses, due to convective flow patches, are able to vertically transport horizontal momentum, maintaining the mean flow. This mechanism is similar to the turbulent wind observed in thermal convection. In this study, the interleaving is produced using the classical set-up of Ruddick & Turner (Deep-Sea Res., vol. 558, 1979, p. 903). The dam-break experiments better resemble the case of oceanic fronts, where interleaving is commonly observed. The flow structures are investigated by measuring the two-dimensional flow fields using the particle image velocimetry technique. The resulting two-dimensional vector fields reveal complex fine-scale flow structures, and convective patterns are observed inside the finger-favourable layers. Vortical structures at scales comparable with the layer thickness are embedded in these regions and seem to be responsible for sustaining the horizontal mean flow against the viscous dissipations, especially in a region close to the layer nose. A spectral analysis of the flow fields suggest that the energy balance is governed by an inverse energy cascade, which implies a transfer of energy from the smaller scales to the larger scales (mean flow).

The absolute/convective character of the linear instability of axisymmetric jets is investigated for a wide range of parallel velocity and density profiles. An adjoint-based sensitivity analysis is carried out in order to maximize the absolute growth rate of jet profiles with and without density variations. It is demonstrated that jets without counterflow may display absolute instability at density ratios well above the previously assumed threshold ρjet/ρ∞ = 0.72, and even in homogeneous settings. Absolute instability is promoted by a strong velocity gradient in the low-velocity region of the shear layer, as well as by a step-like density variation near the location of maximum shear. A new efficient algorithm for the computation of the absolute instability mode is presented.

We summarize the results of an extensive campaign of direct numerical simulations of Rayleigh–Bénard convection at moderate and high Prandtl numbers (10−1 ≤ Pr ≤ 104) and moderate Rayleigh numbers (105 ≤ Ra ≤ 109). The computational domain is a cylindrical cell of aspect ratio Γ = 1/2, with the no-slip condition imposed on all boundaries. By scaling the numerical results, we find that the free-fall velocity should be multiplied by in order to obtain a more appropriate representation of the large-scale velocity at high Pr. We investigate the Nusselt and the Reynolds number dependences on Ra and Pr, comparing the outcome with previous numerical and experimental results. Depending on Pr, we obtain different power laws of the Nusselt number with respect to Ra, ranging from Ra2/7 for Pr = 1 up to Ra0.31 for Pr = 103. The Nusselt number is independent of Pr. The Reynolds number scales as , neglecting logarithmic corrections. We analyse the global and local features of viscous and thermal boundary layers and their scaling behaviours with respect to Ra and Pr, and with respect to the Reynolds and Péclet numbers. We find that the flow approaches a saturation state when Reynolds number decreases below the critical value, Res ≃ 40. The thermal-boundary-layer thickness increases slightly (instead of decreasing) when the Péclet number increases, because of the moderating influence of the viscous boundary layer. The simulated ranges of Ra and Pr contain steady, periodic and turbulent solutions. A rough estimate of the transition from the steady to the unsteady state is obtained by monitoring the time evolution of the system until it reaches stationary solutions. We find multiple solutions as long-term phenomena at Ra = 108 and Pr = 103, which, however, do not result in significantly different Nusselt numbers. One of these multiple solutions, even if stable over a long time interval, shows a break in the mid-plane symmetry of the temperature profile. We analyse the flow structures through the transitional phases by direct visualizations of the temperature and velocity fields. A wide variety of large-scale circulation and plume structures has been found. The single-roll circulation is characteristic only of the steady and periodic solutions. For other regimes at lower Pr, the mean flow generally consists of two opposite toroidal structures; at higher Pr, the flow is organized in the form of multi-jet structures, extending mostly in the vertical direction. At high Pr, plumes mainly detach from sheet-like structures. The signatures of different large-scale structures are generally well reflected in the data trends with respect to Ra, less in those with respect to Pr.