Steady small-amplitude thermal convection in a fluid-saturated. infinitely extended porous layer is investigated theoretically in the wavenumber range 1/√2−1. It was shown that the point of multiple bifurcation Ra0 = (3/√2 + 2)π2, α0 = 2−0.25 leads to secondary bifurcation when the wavenumber decreases.

As a result a new branch of a stable, complicated, three-dimensional flow in the square cell was discovered for α close to α0. This branch joins two adjacent branches of three-dimensional flows emanating from the trivial solution and causes their stability transition at the branching points.

The problems of tsunami generation are treated by standard integral-transform and modified stationary-phase methods to yield asymptotic approximations to the surface disturbances. The effects of asymmetry are considered in a one-dimensional ocean. Series representations are used to produce sets of normal-mode oscillations in a two-dimensional ocean, and the magnitudes of the wave front and wave train are discussed in relation to the asymmetry of the generating region.

In this paper, the analysis on secondary flow in curved elliptic pipes of Topakoglu & Ebadian (1985) has been extended up to a point where the rate-of-flow expression is obtained for any value of flatness ratio of the elliptic cross-section. The analysis is based on the double expansion method of Topakoglu (1967). Therefore, no approximation is involved in any step other than the natural limitation of the finite number of calculated terms of the expansions. The obtained results are systematically plotted against the curvature of centreline of the curved pipe for different values of Reynolds number.

An array of long, vertically uniform salt fingers in an environment with salt input from above, fresh input from below, a vertically constant, stabilizing temperature gradient and negligible salt diffusion is found to be unstable to perturbations with vertical structure. The maximum growth rate and the form of the instability are derived for fingers with widths that yield maximum buoyancy flux in the unperturbed state. The dependence of the instability on the magnitude of the imposed salt difference is obtained for the heat–salt system. A direct (non-oscillatory) mode with a vertical scale of the order of the buoyancy-layer thickness is the most unstable when the amplitude of the vertical velocity of the fingers is large. The instability is due to the shear flow between rising and sinking fluid in adjacent fingers and is relatively unaffected by the perturbation buoyancy. When the driving is weaker, the dominant instability involves the same processes as for the basic fingers, i.e. perturbation buoyancy, viscosity and diffusion, and the mode becomes oscillatory in time. All of the most unstable modes derived here have a vertical scale of the order of the buoyancy-layer thickness. Both the direct and the oscillatory modes have net horizontal flows that vary with the vertical coordinate and time and in finite amplitude could cause the fingers to incline toward the horizontal. The oscillatory mode involves pairs of fingers so the emerging behaviour could include a kind of period doubling.

Eckart's (1952) second-order, self-adjoint partial differential equation for the free-surface displacement of monochromatic gravity waves in water of variable depth h is derived from a variational formulation by approximating the vertical variation of the velocity potential in the average Lagrangian by that for deep-water waves. It is compared with the ‘mild-slope equation’, which also is second order and self-adjoint and may be obtained by approximating the vertical variation in the average Lagrangian by that for uniform, finite depth. The errors in these approximations vanish for either κh ↓ 0 or κh ↑ ∞ (κ ≡ ω2/g). Both approximations are applied to slowly modulated wavetrains, following Whitham's (1974) formulation for uniform depth. Both conserve wave action; the mild-slope approximation conserves wave energy, but Eckart's approximation does not (except for uniform depth). The two approximations are compared through the calculation of reflection from a gently sloping beach and of edge-wave eigenvalues for a uniform slope (not necessarily small). Eckart's approximation is inferior to the mild-slope approximation for the amplitude in the reflection problem, but it is superior in the edge-wave problem, for which it provides an analytical approximation that is exact for the dominant mode and in error by less than 1.6% for all higher modes within the range of admissible slopes. In contrast, the mild-slope approximation requires numerical integration (Smith & Sprinks 1975) and differs significantly from the exact result for the dominant mode for large slopes.

Experimental results are presented for the instantaneous release of a constant volume of air into water in a long horizontal tube of square cross-section. The tube is closed at both ends and the volume of air is confined at one of the ends before it is released. The resulting motion, after the rapid formation of an air-cavity front, may be divided into three phases: initially the front of the air cavity moves at constant speed, later its speed decreases monotonically, and finally its speed executes a long series of erratic stops and starts before coming entirely to rest. The transition from the first to the second phase is observed to occur when a disturbance due to the tube end overtakes the cavity front. The final phase is dominated by surface-tension effects, complicated by surface contaminants. A simple model of the flow, based on Benjamin's (1968) theory of steady cavity flow and the classical theory of hydraulic jumps, is developed. With correction for surface tension, the model results compare well with the experimental results for the first two phases.

Model experiments are performed in a long, thin-walled, fluid-filled, latex tube in which the fluid is locally suddenly stopped, starting from a steady flow, thus simulating the wave phenomenon generated by the final closure of the aortic valve. The resulting waveform is determined as it propagates upstream. The effect of a local step-wise change in compliance close to the valve, representing the aortic sinus section, is investigated. The observed phenomena are analysed by means of a quasi-one-dimensional model, solved by the method of characteristics, taking into account the influence of nonlinearities, wall shear stress, viscoelastic wall properties and wave reflections. The theoretical computations are well confirmed by the experimental results. The pressure jump, induced by the valve closing, appeared to be slightly affected by nonlinearities. The decrease of the pressure jump while propagating upstream and the gradual pressure increase that follows the pressure jump are caused by the effect of wall shear stress. The local change in compliance generates the expected wave reflections and has a strong influence on the rise-time of the wave front. The experiments confirmed the prediction that wall viscoelasticity is the dominant factor in the gradual decay of the slope of the wave.

We present results of the calculation of superharmonic normal mode perturbations to the exact nonlinear deep-water capilary wave solution of Crapper (1957). By using the method of Longuet-Higgins (1978a), we are able for the first time to consider all waveheights up to and including the maximum for two-dimensional perturbations. We find agreement with the recent asymptotic analysis of Hogan, Gruman & Stiassnie (1988). Superharmonic instabilities are found at various waveheights less than the maximum.

It has been shown (Evans 1976) that the power absorbed by a general, axisymmetric body depends solely upon the added-mass and damping coefficients. These coefficients are fundamental properties of the body, representing the component of the force on the body proportional to the acceleration and velocity of the body respectively in the radiation problem, where the body is forced to oscillate in the absence of incoming waves.

In the present paper these coefficients are determined by solution of the radiation problem, for a mouth-upward cylindrical duct situated on the sea bed and fitted with a piston undergoing forced oscillations. The added-mass and damping coefficients are then used to study the power absorption properties of the duct when the power take-off is modelled by a linear-spring–dashpot system attached to the piston. Curves of the added mass, damping coefficients and absorption length (a measure of the power absorbed) as functions of wavenumber are presented, for different duct diameters and different depths of submergence.

This paper examines the sound generation process when a supersonic aerofoil cuts through a steady jet flow. It is shown that the principal sound is generated by the leading edge of the aerofoil when it interacts with the streaming jet. To the leading order in terms of the jet velocity, no trailing-edge sound is generated. This is not the result of the cancellation of a trailing-edge sound by that from vortex shedding through the imposition of the Kutta condition. Instead, the null acoustic radiation from the trailing edge is entirely because, to the leading order, there is no interaction between the trailing edge and the jet. The effect of the trailing edge is to diffract sound waves generated by the leading edge. It is shown that the diffracted field (as well as the incident field) is regular at the trailing edge and the issue of satisfying the Kutta condition does not arise during the diffraction process. Thus, there is no extra vortex shedding from the trailing edge owing to its interaction with the flow, apart from those resulting from the discontinuity across the aerofoil, generated by the flow-leading edge interaction. This is in sharp contrast to the case of subsonic aerofoils where the removal of the singularity in the diffracted field at the trailing edge through the imposition of the Kutta condition results in vortex shedding from the sharp edge and energy exchange between the sound field and the vortical wake.

The three-dimensional boundary layer on a swept wing can support different types of hydrodynamic instability. Here attention is focused on the so-called ‘spanwise instability’ problem which occurs when the attachment-line boundary layer on the leading edge becomes unstable to Tollmien–Schlichting waves. In order to gain insight into the interactions that are important in that problem a simplified basic state is considered. This simplified flow corresponds to the swept attachment-line boundary layer on an infinite flat plate. The basic flow here is an exact solution of the Navier–Stokes equations and its stability to two-dimensional waves propagating along the attachment line can be considered exactly at finite Reynolds number. This has been done in the linear and weakly nonlinear regimes by Hall, Malik & Poll (1984) and Hall & Malik (1986). Here the corresponding problem is studied for oblique waves and their interaction with two-dimensional waves is investigated. In fact oblique modes cannot be described exactly at finite Reynolds number so it is necessary to make a high-Reynolds-number approximation and use triple-deck theory. It is shown that there are two types of oblique wave which, if excited, cause the destabilization of the two-dimensional mode and the breakdown of the disturbed flow at a finite distance from the leading edge. First a low-frequency mode closely related to the viscous stationary crossflow mode discussed by Hall (1986) and MacKerrell (1987) is a possible cause of breakdown. Secondly a class of oblique wave with frequency comparable with that of the two-dimensional mode is another cause of breakdown. It is shown that the relative importance of the modes depends on the distance from the attachment line.

Thermosolutal convection provides a testbed for applications of nonlinear dynamics to fluid motion. If the ratio of solutal to thermal diffusivity is small and the solutal Rayleigh number RS is large, instability sets in at a Hopf bifurcation as the thermal Rayleigh number RT is increased. For two-dimensional convection in a rectangular box the fundamental mode is a single roll with point symmetry about its axis. The symmetries of periodic and steady solutions form an eighth-order group with invariant subgroups that describe pure single-roll and multiroll solutions. A systematic numerical investigation reveals a rich variety of spatiotemporal behaviour in the regime where RS [Gt ] RT−RS > 0. Point symmetry is broken and there is a branch of spatially asymmetric periodic solutions. These mixed-mode oscillations lose their temporal symmetry in a subsequent bifurcation, followed eventually by a transition to chaos. The numerical experiments can be interpreted by relating the physical form of the solutions to an appropriate bifurcation structure.

Fully three-dimensional numerical solutions are presented for Rayleigh-Bénard convection subject to stress-free boundary conditions. A motion with square planform is studied for varying Rayleigh number R and Prandtl number σ. It may be understood partly in terms of a truncated modal representation (after Lorenz 1963). Thermal layers of unusual structure are found at high R. For small σ, steady solutions exist, but are not of ‘flywheel’ type, and the heat transport depends strongly on σ. The study also verifies that laminar convective flows may be ergodic.

We present measurements of the azimuthal rotation velocity $\dot{{\it\theta}}(t)$ and thermal amplitude ${\it\delta}(t)$ of the large-scale circulation in turbulent Rayleigh–Bénard convection with modulated rotation. Both $\dot{{\it\theta}}(t)$ and ${\it\delta}(t)$ exhibit clear oscillations at the modulation frequency ${\it\omega}$. Fluid acceleration driven by oscillating Coriolis force causes an increasing phase lag in $\dot{{\it\theta}}(t)$ when ${\it\omega}$ increases. The applied modulation produces oscillatory boundary layers and the resulting time-varying viscous drag modifies ${\it\delta}(t)$ periodically. Oscillation of $\dot{{\it\theta}}(t)$ with maximum amplitude occurs at a finite modulation frequency ${\it\omega}^{\ast }$. Such a resonance-like phenomenon is interpreted as a result of optimal coupling of ${\it\delta}(t)$ to the modulated rotation velocity. We show that an extended large-scale circulation model with a relaxation time for ${\it\delta}(t)$ in response to the modulated rotation provides predictions in close agreement with the experimental results.

The present work is a comprehensive theoretical study of the heat transfer associated with a 3-singlet compound drop that is growing because of change of phase. The geometry is the same as in Part 1, i.e. a vapour bubble partially surrounded by its own liquid in another immiscible liquid. The attempt here is to gain fundamental understanding of the transport processes that take place in connection with direct-contact heat exchange. The fluid dynamics associated with its growth and translation is treated in Part 1. Here, that flow field solution is used to obtain the temperature field and hence the evaporation rate. The energy equation for the system consisting of a single compound drop is solved numerically by finite-difference methods. The results give the complete time history of evaporation of the drop. In addition, useful quantities such as the Nusselt number are given and compared with existing experimental data. Most of the results have good agreement with experimental data.

The relative velocities and positions of monodisperse high-inertia particle pairs in isotropic turbulence are studied using direct numerical simulations (DNS), as well as Langevin simulations (LS) based on a probability density function (PDF) kinetic model for pair relative motion. In a prior study (Rani et al., J. Fluid Mech., vol. 756, 2014, pp. 870–902), the authors developed a stochastic theory that involved deriving closures in the limit of high Stokes number for the diffusivity tensor in the PDF equation for monodisperse particle pairs. The diffusivity contained the time integral of the Eulerian two-time correlation of fluid relative velocities seen by pairs that are nearly stationary. The two-time correlation was analytically resolved through the approximation that the temporal change in the fluid relative velocities seen by a pair occurs principally due to the advection of smaller eddies past the pair by large-scale eddies. Accordingly, two diffusivity expressions were obtained based on whether the pair centre of mass remained fixed during flow time scales, or moved in response to integral-scale eddies. In the current study, a quantitative analysis of the (Rani et al. 2014) stochastic theory is performed through a comparison of the pair statistics obtained using LS with those from DNS. LS consist of evolving the Langevin equations for pair separation and relative velocity, which is statistically equivalent to solving the classical Fokker–Planck form of the pair PDF equation. Langevin simulations of particle-pair dispersion were performed using three closure forms of the diffusivity – i.e. the one containing the time integral of the Eulerian two-time correlation of the seen fluid relative velocities and the two analytical diffusivity expressions. In the first closure form, the two-time correlation was computed using DNS of forced isotropic turbulence laden with stationary particles. The two analytical closure forms have the advantage that they can be evaluated using a model for the turbulence energy spectrum that closely matched the DNS spectrum. The three diffusivities are analysed to quantify the effects of the approximations made in deriving them. Pair relative-motion statistics obtained from the three sets of Langevin simulations are compared with the results from the DNS of (moving) particle-laden forced isotropic turbulence for $St_{\unicode[STIX]{x1D702}}=10,20,40,80$ and $Re_{\unicode[STIX]{x1D706}}=76,131$. Here, $St_{\unicode[STIX]{x1D702}}$ is the particle Stokes number based on the Kolmogorov time scale and $Re_{\unicode[STIX]{x1D706}}$ is the Taylor micro-scale Reynolds number. Statistics such as the radial distribution function (RDF), the variance and kurtosis of particle-pair relative velocities and the particle collision kernel were computed using both Langevin and DNS runs, and compared. The RDFs from the stochastic runs were in good agreement with those from the DNS. Also computed were the PDFs $\unicode[STIX]{x1D6FA}(U|r)$ and $\unicode[STIX]{x1D6FA}(U_{r}|r)$ of relative velocity $U$ and of the radial component of relative velocity $U_{r}$ respectively, both PDFs conditioned on separation $r$. The first closure form, involving the Eulerian two-time correlation of fluid relative velocities, showed the best agreement with the DNS results for the PDFs.

The local implications of the well-known flux conservation equations of Morton et al. (Proc. R. Soc. Lond. A, vol. 234, 1956, p.1) for plumes and jets are considered. Given the vertical velocity distributions of a model plume or jet, the divergence-free radial velocity distributions are calculated. It is shown that in general the velocity of the plume boundary is not described by the local total fluid velocity in this way. A two-fluid model tracking the evolution of both ‘plume’ and ‘ambient’ fluid is proposed which resolves this apparent inconsistency and also provides a way of explicitly describing the mixing process within a model plume. The plume boundary acts as a phase boundary across which ambient fluid is entrained, and the plume boundary moves at the velocity of the plume fluid. The difference between the plume-fluid radial velocity and the total fluid velocity quantifies in a natural way the purely horizontal entrainment flux of ambient fluid into the plume across the phase boundary at the plume edge.

The scattering of acoustic waves by compact three-dimensional axisymmetric vortices is studied using direct numerical simulation in the case where the incoming wave is aligned with the symmetry axis and the direction of propagation of the vortices. The cases of scattering by Hill’s spherical vortex and Gaussian vortex rings are examined, and results are compared with predictions obtained by matched asymptotic expansions and the Born approximation. Good agreement is obtained for long waves, with the Born approximation usually giving better predictions, especially as the difference in scale between vortex and incoming waves decreases and as the Mach number of the flow increases. An improved version of the Born approximation which takes into account higher-order effects in Mach number gives the best agreement.