This study quantifies the mixing that results from a pulsed jet in cross-flow in the near jet region. By large eddy simulation computations, it also helps to understand the physical phenomena involved in the formation of the pulsed jet in cross-flow. The boundary conditions of the jet inlet are implemented via a Navier–Stokes characteristic boundary condition coupled with a Fourier series development. The signals used to pulse the jet inlet are a square or a sine wave. A new way of characterizing the mixing is introduced with the goal of easily interpreting and quantifying the complicated mixing process involved in a pulsed jet in cross-flow flow fields. Different flow configurations, pulsed and non-pulsed, are computed and compared, keeping the root mean square value of the signal constant. This comparison not only allows the characterization of the mixing but also illustrates some of the properties of the mixing characterization.

The St Venant equations in conjunction with a phenomenological law for erosion are used to explore the nonlinear dynamics of cyclic steps – linearly unstable bedform patterns which emerge when uniform flow over an erodible bed becomes supercritical. The instability saturates by blocking the overlying flow and creating hydraulic jumps just downstream of the steepest part of the steps. Near onset, steadily migrating, nonlinear step patterns are constructed and shown to suffer a short-wavelength secondary instability that ‘roughens’ the bed and renders the staircase patterns less regular and time-dependent. An eddy viscosity is needed to regularize both the onset of the primary steps and the secondary instabilities. Further beyond the critical Froude number, the steps block the flow sufficiently to arrest erosion significantly, creating complicated patterns mixing migrating steps and stationary bedforms. The reduction in flux also stabilizes roll waves – a second, hydrodynamic instability of uniform supercritical flow. It is further shown that roll waves are purely convective instabilities, whereas cyclic steps can be absolute. Thus, in the finite geometries of the laboratory or field, it may be difficult to excite roll waves. On the other hand, the complicated spatiotemporal patterns associated with the cyclic-step instability should develop naturally. The complicated patterns resulting from the secondary instability do not appear to have been observed experimentally, calling into question the validity of the model.

The linear stability of a uniform flow in an infinitely wide erodible channel is investigated with respect to disturbances of the bed that are periodic in both the transverse and the longitudinal directions. A rotational flow and sediment transport model, originally developed to study the formation of two-dimensional dunes and antidunes, is straightforwardly extended to cover variations in the lateral direction. Sediment is assumed to be transported as bed load, disregarding the role of suspension. Following a standard linearization procedure, a dispersion relationship is obtained that expresses the growth rate and the celerity of the sand wave as a function of the streamwise and spanwise wavenumbers and of the relevant flow and sediment parameters. Regions of instabilities in the space of the parameters are found, which can be associated with bed forms of different kinds, spanning from dunes and antidunes to alternate bars. Therefore, the present theory allows for a unified view of the formation of two- and three-dimensional bed forms in rivers in terms of the relevant flow and sediment parameters.

An integral model predicting the mean flow and mixing properties of inclined plane and round turbulent buoyant jets in a motionless environment of uniform density is proposed. The escaping masses from the main buoyant jet flow are simulated, and the model can be successfully applied to initial discharge inclinations from 90 to with respect to the horizontal plane. This complementary approach introduces a concentration coefficient, which is calibrated using experimental evidence. The present model has incorporated the second-order approach and, regarding the jet-core region, a jet-core model based on the advanced integral model for the production of more correct transverse profiles of the mean axial velocities and mean concentrations than the common Gaussian or top-hat profiles. The partial differential equations for momentum and tracer conservation are written in orthogonal and cylindrical curvilinear coordinates for inclined plane and round buoyant jets, respectively, and they are integrated under the closure assumptions of (a) quasi-linear spreading of the mean flow and mixing fields, and (b) known transverse profile distributions. The integral forms are solved by employing the Runge–Kutta algorithm. Since the most important contribution in the present model is the simulation of the escaping masses, the model has been called the escaping mass approach (EMA). Herein EMA is applied to predict the mean flow properties (trajectory characteristics, mean axial velocities and mean concentrations) for inclined plane and round buoyant jets. The results predicted are compared with experimental data available in the literature, and the accuracy obtained is more than satisfactory. The performance of the EMA is up to 56 % better than using classical integral procedures. EMA can be used for design purposes and for environmental impact assessment studies.

If a moving body were made to vanish within a fluid, its boundary-layer vorticity would be released into the fluid at all locations simultaneously, a phenomenon we call global vorticity shedding. We approximate this process by studying the related problem of rapid vorticity transfer from the boundary layer of a body undergoing a quick change of cross-sectional and surface area. A surface-piercing foil is first towed through water at constant speed, , and constant angle of attack, then rapidly pulled out of the fluid in the spanwise direction. Viewed within a fixed plane perpendicular to the span, the cross-sectional area of the foil seemingly disappears. The rapid spanwise motion results in the nearly instantaneous shedding of the boundary layer into the surrounding fluid. Particle image velocimetry measurements show that the shed layers quickly transition from free shear layers to form two strong, unequal-strength vortices, formed within non-dimensional time , based on the foil chord and forward velocity. These vortices are connected to, and interact with, the foil’s tip vortex through additional streamwise vorticity formed during the rapid pulling of the foil. Numerical simulations show that two strong spanwise vortices form from the shed vorticity of the boundary layer. The three-dimensional effects of the foil removal process are restricted to the tip of the foil. This method of vorticity transfer may be used for quickly introducing circulation to a fluid to provide forcing for biologically inspired flow control.

The wake of a sphere rolling along a wall at low Reynolds number is investigated numerically and experimentally. Two successive transitions are identified in this flow, as the Reynolds number is increased. The first leads to the periodic shedding of planar symmetric hairpin vortices. The second and previously unknown transition involves a loss of planar symmetry and a low-frequency lateral oscillation of the wake, exhibiting a surprising 7:3 resonance with the hairpin vortex shedding. The two transitions are characterized by dye visualizations and quantitative information obtained from numerical simulations, such as force coefficients and wake frequencies (Strouhal numbers). Both transitions are found to be supercritical. Further increasing the Reynolds number, the flow becomes progressively more disorganized and chaotic. Overall, the transition sequence for the rolling sphere is closer to the one for a non-rotating sphere in a free stream than to that of a non-rotating sphere close to a wall.

The ability of wavelet multi-resolution analysis to detect and track the energy-containing motions that govern the dynamics of a fluid flow offers a unique hierarchical framework for modelling and simulating turbulence. In this paper, the role of the wavelet thresholding level in wavelet-based modelling and simulation of turbulent flows is systematically examined. The thresholding level controls the relative importance of resolved energetic structures and residual unresolved background flow and, thus, the achieved turbulence resolution. A fully adaptive eddy capturing approach is developed that allows variable-fidelity numerical simulations of turbulence to be performed. The new method is based on wavelet filtering with time-varying thresholding. The thresholding level automatically adapts to the desired turbulence resolution during the simulation. The filtered governing equations supplemented by a localized dynamic energy-based closure model are solved numerically using the adaptive wavelet collocation method. The approach is successfully tested in the numerical simulation of both linearly forced and freely decaying homogeneous turbulence.

The evolution of barotropic vortices over a topographic, axisymmetric mountain in a homogeneous rotating fluid is studied experimentally. The aim is to identify the main physical processes observed in (i) a horizontal plane of motion, perpendicular to the rotation axis of the system, and (ii) a vertical plane across the diameter of the mountain. The vortices are monopolar cyclones initially generated near or over the topography. Initially, the vortices drift towards the mountain due to the -effect associated with the topographic slope. On arriving, they turn around the obstacle in an anticyclonic direction, whilst anticyclonic vorticity is generated over the summit. The long-term vorticity distribution is dominated by the original cyclone elongated around the topographic contours and the generated anticyclone over the tip of the topography. In the vertical plane an oscillatory uphill–downhill flow is generated, which is directly related to the drift of the cyclone around the mountain. Depending on the vortex characteristics, the period of the oscillation ranges from 4 to 10 times the rotation period of the system. The horizontal and vertical flow fields are reproduced numerically by using a shallow-water formulation, which allows a detailed view of the vertical motions, hence facilitating the interpretation of the experimental results. In addition, the cyclone–anticyclone pair over the mountain is compared with analytical solutions of topographically trapped waves. A general conclusion is that vertical motions persist for several days (or rotation periods), which implies that this mechanism might be potentially important for the vertical transport over seamounts.

It is now well-known that there is an exact formula relating the far-field jet noise spectrum to the convolution product of a propagator (that accounts for the mean flow interactions) and a generalized Reynolds stress autocovariance tensor (that accounts for the turbulence fluctuations). The propagator depends only on the mean flow and an adjoint vector Green’s function for a particular form of the linearized Euler equations. Recent numerical calculations of Karabasov, Bogey & Hynes (AIAA Paper 2011-2929) for a Mach 0.9 jet show use of the true non-parallel flow Green’s function rather than the more conventional locally parallel flow result leads to a significant increase in the predicted low-frequency sound radiation at observation angles close to the downstream jet axis. But the non-parallel flow appears to have little effect on the sound radiated at to the downstream axis. The present paper is concerned with the effects of non-parallel mean flows on the adjoint vector Green’s function. We obtain a low-frequency asymptotic solution for that function by solving a very simple second-order hyperbolic equation for a composite dependent variable (which is directly proportional to a pressure-like component of this Green’s function and roughly corresponds to the strength of a monopole source within the jet). Our numerical calculations show that this quantity remains fairly close to the corresponding parallel flow result at low Mach numbers and that, as expected, it converges to that result when an appropriately scaled frequency parameter is increased. But the convergence occurs at progressively higher frequencies as the Mach number increases and the supersonic solution never actually converges to the parallel flow result in the vicinity of a critical- layer singularity that occurs in that solution. The dominant contribution to the propagator comes from the radial derivative of a certain component of the adjoint vector Green’s function. The non-parallel flow has a large effect on this quantity, causing it (and, therefore, the radiated sound) to increase at subsonic speeds and decrease at supersonic speeds. The effects of acoustic source location can be visualized by plotting the magnitude of this quantity, as function of position. These ‘altitude plots’ (which represent the intensity of the radiated sound as a function of source location) show that while the parallel flow solutions exhibit a single peak at subsonic speeds (when the source point is centred on the initial shear layer), the non-parallel solutions exhibit a double peak structure, with the second peak occurring about two potential core lengths downstream of the nozzle. These results are qualitatively consistent with the numerical calculations reported in Karabasov et al. (2011).

The Taylor–Goldstein (T–G) equation is extended to include the effects of small-scale turbulence represented by non-uniform vertical and horizontal eddy viscosity and diffusion coefficients. The vertical coefficients of viscosity and diffusion, and , respectively, are assumed to be equal and are expressed in terms of the buoyancy frequency of the flow, , and the dissipation rate of turbulent kinetic energy per unit mass, , quantities that can be measured in the sea. The horizontal eddy coefficients, and , are taken to be proportional to the dimensionally correct form, , found appropriate in the description of horizontal dispersion of a field of passive markers of scale . The extended T–G equation is applied to examine the stability and greatest growth rates in a turbulent shear flow in stratified waters near a sill, that at the entrance to the Clyde Sea in the west of Scotland. Here the main effect of turbulence is a tendency towards stabilizing the flow; the greatest growth rates of small unstable disturbances decrease, and in some cases flows that are unstable in the absence of turbulence are stabilized when its effects are included. It is conjectured that stabilization of a flow by turbulence may lead to a repeating cycle in which a flow with low levels of turbulence becomes unstable, increasing the turbulent dissipation rate and so stabilizing the flow. The collapse of turbulence then leads to a condition in which the flow may again become unstable, the cycle repeating. Two parameters are used to describe the ‘marginality’ of the observed flows. One is based on the proximity of the minimum flow Richardson number to the critical Richardson number, the other on the change in dissipation rate required to stabilize or destabilize an observed flow. The latter is related to the change needed in the flow Reynolds number to achieve zero growth rate. The unstable flows, typical of the Clyde Sea site, are relatively further from neutral stability in Reynolds number than in Richardson number. The effects of turbulence on the hydraulic state of the flow are assessed by examining the speed and propagation direction of long waves in the Clyde Sea. Results are compared to those obtained using the T–G equation without turbulent viscosity or diffusivity. Turbulence may change the state of a flow from subcritical to supercritical.

The characteristics of the boundary layer separating a turbulence region from an irrotational (or non-turbulent) flow region are investigated using rapid distortion theory (RDT). The turbulence region is approximated as homogeneous and isotropic far away from the bounding turbulent/non-turbulent (T/NT) interface, which is assumed to remain approximately flat. Inviscid effects resulting from the continuity of the normal velocity and pressure at the interface, in addition to viscous effects resulting from the continuity of the tangential velocity and shear stress, are taken into account by considering a sudden insertion of the T/NT interface, in the absence of mean shear. Profiles of the velocity variances, turbulent kinetic energy (TKE), viscous dissipation rate (), turbulence length scales, and pressure statistics are derived, showing an excellent agreement with results from direct numerical simulations (DNS). Interestingly, the normalized inviscid flow statistics at the T/NT interface do not depend on the form of the assumed TKE spectrum. Outside the turbulent region, where the flow is irrotational (except inside a thin viscous boundary layer), decays as , where is the distance from the T/NT interface. The mean pressure distribution is calculated using RDT, and exhibits a decrease towards the turbulence region due to the associated velocity fluctuations, consistent with the generation of a mean entrainment velocity. The vorticity variance and display large maxima at the T/NT interface due to the inviscid discontinuities of the tangential velocity variances existing there, and these maxima are quantitatively related to the thickness of the viscous boundary layer (VBL). For an equilibrium VBL, the RDT analysis suggests that (where is the Kolmogorov microscale), which is consistent with the scaling law identified in a very recent DNS study for shear-free T/NT interfaces.

In Ablowitz, Fokas & Musslimani (J. Fluid Mech., vol. 562, 2006, pp. 313–343) a novel formulation was proposed for water waves in three space dimensions. In the flat-bottom case, this formulation consists of the Bernoulli equation, as well as of a non-local equation. The variable-bottom case, which now involves two non-local equations, was outlined but not explored in the above paper. Here, the variable-bottom formulation is addressed in more detail. First, it is shown that in the weakly nonlinear, weakly dispersive regime, the above system of three equations can be reduced to a system of two equations. Second, by combining the novel non-local formulation of the above authors with conformal mappings, it is shown that in the two-dimensional case, it is possible to obtain a system of two equations without any asymptotic approximations. Furthermore, for the weakly nonlinear, weakly dispersive regime, the nonlinear equations are simpler than the equations obtained without conformal mappings, since they contain lower order derivatives for the terms involving the bottom variable.

An experimental and theoretical investigation of the air trapping by a blunt, locally spherical body impacting onto the free surface of water is conducted. In the parameter regime previously studied theoretically by Hicks & Purvis (J. Fluid Mech., vol. 649, 2010, pp. 135–163), excellent agreement between experimental data and theoretical modelling is obtained. Earlier predictions of the radius of the trapped air pocket are confirmed. A boundary element method is used to consider air cushioning of an impact of an axisymmetric body into water. Efficient computational methods are obtained by analytically integrating the boundary integral equation over the azimuthal variable. The resulting numerically computed free-surface profiles predict an annular touchdown region in excellent agreement with the experiments.

When droplets of dilute suspensions are left to evaporate the final dry residue is typically the familiar coffee-ring stain, with nearly all material deposited at the initial triple line (Deegan et al., Nature, vol. 389, 1997, pp. 827–829). However, aqueous poly(ethylene oxide) (PEO) droplets only form coffee-ring stains for a very narrow range of the experimental parameters molecular weight, concentration and drying rate. Instead, over a wide range of values they form either a flat disk or a very distinctive tall central monolith via a four-stage deposition process which includes a remarkable bootstrap-building step. To predict which deposit will form, we present a quantitative model comparing the effects of advective build-up at the triple line to diffusive flux and use this to calculate a dimensionless number . By experimentally varying concentration and flux (using a low-pressure drying chamber), the prediction is tested over nearly two orders of magnitude in both variables and shown to be in good agreement with the boundary between disks and monoliths at .

The interplay between incompressibility and stratification can lead to non-conservation of horizontal momentum in the dynamics of a stably stratified incompressible Euler fluid filling an infinite horizontal channel between rigid upper and lower plates. Lack of conservation occurs even though in this configuration only vertical external forces act on the system. This apparent paradox was seemingly first noticed by Benjamin (J. Fluid Mech., vol. 165, 1986, pp. 445–474) in his classification of the invariants by symmetry groups with the Hamiltonian structure of the Euler equations in two-dimensional settings, but it appears to have been largely ignored since. By working directly with the motion equations, the paradox is shown here to be a consequence of the rigid lid constraint coupling through incompressibility with the infinite inertia of the far ends of the channel, assumed to be at rest in hydrostatic equilibrium. Accordingly, when inertia is removed by eliminating the stratification, or, remarkably, by using the Boussinesq approximation of uniform density for the inertia terms, horizontal momentum conservation is recovered. This interplay between constraints, action at a distance by incompressibility, and inertia is illustrated by layer-averaged exact results, two-layer long-wave models, and direct numerical simulations of the incompressible Euler equations with smooth stratification.

In this paper, a multi-layer model is developed for the purpose of studying nonlinear internal wave propagation in shallow water. The methodology employed in constructing the multi-layer model is similar to that used in deriving Boussinesq-type equations for surface gravity waves. It can also be viewed as an extension of the two-layer model developed by Choi & Camassa. The multi-layer model approximates the continuous density stratification by an -layer fluid system in which a constant density is assumed in each layer. This allows the model to investigate higher-mode internal waves. Furthermore, the model is capable of simulating large-amplitude internal waves up to the breaking point. However, the model is limited by the assumption that the total water depth is shallow in comparison with the wavelength of interest. Furthermore, the vertical vorticity must vanish, while the horizontal vorticity components are weak. Numerical examples for strongly nonlinear waves are compared with laboratory data and other numerical studies in a two-layer fluid system. Good agreement is observed. The generation and propagation of mode-1 and mode-2 internal waves and their interactions with bottom topography are also investigated.

We study transport in an idealized porous medium characterized by a spatially varying retardation factor, which models linear instantaneous chemical adsorption of a solute. Using a stochastic modelling approach, we study the impact of disorder correlation on the large-scale dispersion behaviour. We consider short, long-range and intermediate-range disorder correlations, and demonstrate that (truncated) power-law correlation causes anomalous dispersion, even in the presence of weak heterogeneity. We identify different preasymptotic and asymptotic regimes of anomalous dispersion that shed new light on the disorder and local-scale transport mechanisms leading to non-Fickian behaviour. The analytical results are complemented by numerical random walk particle tracking simulations, which are found to be in good agreement with the derived dispersion behaviour. We conclude the paper by deriving an effective transport equation for this system, which can be shown to be tied to the family of continuous-time random walk models.

The global, temporal stability of the two-dimensional, incompressible flow over a forward-facing step in a plane channel is investigated numerically. The geometry is varied systematically covering constriction ratios (step-to-inlet height) from 0.23 to 0.965. A three-dimensional linear stability analysis shows that the stability boundary is a smooth continuous function of the constriction ratio. If the critical Reynolds and wavenumbers are scaled appropriately, they approach a linear asymptotic behaviour for large step heights. The critical mode is found to be stationary and confined to the region of separated flow downstream of the step for all constriction ratios. An energy-transfer analysis reveals that the basic flow becomes unstable due to a combined effect involving lift-up and flow deceleration, leading to a critical mode exhibiting steady streaks. Moreover, the receptivity of the flow to initial as well as to structural perturbations is studied by means of an adjoint analysis.

Hypersonic, high-enthalpy flow over a rearward-facing step has been numerically investigated using computational fluid dynamics (CFD). Two conditions relevant to suborbital and superorbital flow with total specific enthalpies of and , are considered. The Mach number and unit Reynolds number per metre were 7.6, 11.0 and , respectively. The Reynolds number based on the step height was correspondingly and . The computations were carried out assuming the flow to be laminar throughout and the real gas effects such as thermal and chemical non-equilibrium are studied using Park’s two-temperature model with finite-rate chemistry and Gupta’s finite-rate chemistry models. In the close vicinity of the step, detailed quantification of flow features is emphasised. In particular, the presence of the Goldstein singularity at the lip and separation on the face of the step have been elucidated. Within the separated region and downstream of reattachment, the influence of real gas effects has been identified and shown to be negligible. The numerical results are compared with the available experimental data of surface heat flux downstream of the step and reasonable agreement is shown up to 30 step heights downstream.

The stability of a horizontal interface between a solution of reactant on top of another solution of reactant is analysed. A chemical product is generated at the interface as a result of a bimolecular chemical reaction . In general, all chemical components are assumed to have different densities and viscosities, and a transverse velocity is introduced parallel to the interface between the reactants. Although the transverse flow is known for its stabilizing effect in viscously unstable non-reactive systems in the presence of an injection velocity, it is shown here that it can actually destabilize an initially stable reactive front. An expression for the critical transverse velocity beyond which an initially stable interface is destabilized is derived in the case of an initial sharp interface for reactants of the same viscosity. The analysis is extended to a diffused profile, and purely buoyancy-driven flows are analysed first in the absence of viscosity contrast and then in the presence of transverse flows and viscosity contrast. Various possible density fingering scenarios are determined based on the relative contribution of each chemical component to the density profile. It is found that the chemical reaction can destabilize a buoyancy-stable initial interface by generating a non-monotonic density profile. Unlike the viscous fingering of a reactive interface, a symmetry in the stability characteristics with respect to density increase or decrease by chemical reaction product is observed in the case of chemically buoyancy-driven flows for identical reactants.