George Batchelor founded JFM in 1956 so that all the diverse aspects of fluid mechanics research could be encompassed in a single high-quality journal. In this endeavour he succeeded admirably. The Journal has enjoyed a central position in the field, first under Batchelor's leadership, and since 1996 under the guidance of David Crighton. In the nearly 50 years since it was founded, JFM has stood for the highest standards of scholarly work and its clear and logical presentation. Over that period of time the subject has remained vibrant while the science and the scientists have evolved. Through this editorial, we wish to raise with the readership a number of issues that we deem to be important and explain some of the action that will be taken.

Over the last couple of decades the percolation time for JFM papers – the average time from submission to publication – has been about 15 or 16 months, which is similar to that for other major journals. There are also occasional aberrations that have resulted in an individual paper percolating over a period of years! We aim to reduce the average percolation time to 12 months and the number of aberrations to near zero. The percolation time for a paper is the collective responsibility of the authors, the referees, the editors and the Press. JFM has direct control over the last two of these, and to some extent the second, and we shall reduce delays by instituting internal reforms and new guidelines for associate editors and referees, while maintaining the customary rigour of the process. We hope that authors will do their parts by replying to referees reports promptly, yet carefully and with a willingness to fine-tune their expositions.

Over the years the average paper in JFM has grown in length by a factor of more than two. From 1970 to 1990 the average length increased by 70% and this trend has continued. Undoubtedly, there is a negative correlation between a paper's length and its readership. We have asked the associate editors (and hence in turn the referees) to make detailed recommendations for the streamlining of papers. There remains no page limit for JFM papers, but the associate editors will exert a firmer hand in attempting to reverse the trend. The ultimate test for a paper's proper length is the density of ideas and results.

It is perceived, wrongly, that JFM publishes only papers of an ‘archival’ nature and does not view short papers favourably. To dispel this myth we are immediately creating a fast track for papers that are 10 or fewer printed pages. These will be handled by a designated associate editor, currently K. R. Sreenivasan, who will expedite the refereeing process and aim at the publication of these papers in 3–4 months from their initial submission. To accomplish this the papers must be prepared using the JFM latex style file and be submitted via e-mail. The papers will be recycled to the authors at most once; if further revisions are necessary, then they will revert to the regular track. These short papers will not be Letters, nor Rapid Communications, but normal JFM papers that will appear in the journal unsegregated from regular-track papers. There will be no compromise on the required standards of scientific quality and clarity of writing.

Fluid mechanics is an extremely vital subject in which traditional topics remain of interest while new ones enter the domain. In an effort to encourage the appearance in JFM of some of the latest topics of investigation, we will be appointing some new associate editors with special expertise who can attract such papers.

We calculate the radiation of acoustic waves into the atmosphere by surface gravity waves on the ocean surface. We show that because of the phase speed mismatch between surface gravity waves and acoustic waves, a single surface wave radiates only evanescent acoustic waves. However, owing to nonlinear terms in the acoustic source, pairs of ocean surface waves can radiate propagating acoustic waves if the two surface waves propagate in almost equal and opposite directions. We derive an analytic expression for the acoustic radiation by a pair of ocean surface waves, and then extend the result to the case of an arbitrary spectrum of ocean surface waves. We present some examples for both the two-dimensional and three-dimensional regimes. Of particular note are the findings that the efficiency of acoustic radiation increases at higher wavenumbers, and the fact that the directionality of the acoustic radiation is often independent of the shape of the spectrum.

We address the differences observed due to heat release between reacting and non-reacting versions of otherwise identical shear flows under conditions for which buoyancy effects are negligible. The differences considered here result from density changes produced by exothermic reaction, and are shown to be similar to those produced by free-stream density differences in non-reacting flows. The piecewise linear variations of temperature with mole fraction allow the density changes due to exothermic reaction to be related to an equivalent non-reacting flow, in which the temperature of one of the fluids is raised to an effective value determined by the peak temperature and overall stoichiometry. This leads to a general equivalence principle by which the scaling laws for non-reacting flows can be extended to predict effects of heat release by exothermic reaction. This equivalence principle is then applied to axisymmetric turbulent jets, where it leads to a generalized momentum diameter d+ in which the scaling laws for burning and non-burning jets become identical – it effectively extends the classical momentum diameter d* of Thring & Newby (1953) and Ricou & Spalding (1961) to exothermic reacting flows. The resulting predicted effects of heat release, in both the near and far fields, show good agreement with experimental data from momentum-dominated turbulent jet diffusion flames. The equivalence principle is then applied to planar turbulent jets, for which it also accurately predicts the observed effects of combustion heat release.

Coagulation and growth of aerosol particles subject to isotropic turbulence has been explored using direct numerical simulations. The computations follow the trajectories of 262 144 initial particles as they are convected by the turbulent flow field. Collision between two parent particles leads to the formation of a new daughter particle with the mass and momentum (but not necessarily the energy) of the parent particles. The initially monodisperse population of particles will develop a size distribution over time that is controlled by the collision dynamics. In an earlier study, Sundaram & Collins (1997) showed that collision rates in isotropic turbulence are controlled by two statistics: (i) the radial distribution of the particles and (ii) the relative velocity probability density function. Their study considered particles that rebound elastically; however, we find that the formula that they derived is equally valid in a coagulating system. However, coagulation alters the numerical values of these statistics from the values they attain for the elastic rebound case. This difference is substantial and must be taken into consideration to properly predict the evolution of the size distribution of a population of particles. The DNS results also show surprising trends in the relative breadth of the particle size distribution. First, in all cases, the standard deviation of the particle size distribution of particles with finite Stokes numbers is much larger than the standard deviation for either the zero-Stokes-number or infinite-Stokes-number limits. Secondly, for particles with small initial Stokes numbers, the standard deviation of the final particle size distribution decreases with increasing initial particle size; however, the opposite trend is observed for particles with slightly larger initial Stokes numbers. An explanation for these phenomena can be found by carefully examining the functional dependence of the radial distribution function on the particle size and Stokes number.

A finite volume method is used to study the generation, propagation and interaction of internal waves in a linearly stratified fluid. The internal waves were generated using single and multiple momentum sources. The full unsteady equations of motion were solved using a SIMPLE scheme on a non-staggered grid. An open boundary, based on the Sommerfield radiation condition, allowed waves to propagate through the computational boundaries with minimum reflection and distortion. For the case of a single momentum source, the effects of viscosity and nonlinearity on the generation and propagation of internal waves were investigated.

Internal wave–wave interactions between two wave rays were studied using two momentum sources. The rays generated travelled out from the sources and intersected in interaction regions where nonlinear interactions caused the waves to break. When two rays had identical properties but opposite horizontal phase velocities (symmetric interaction), the interactions were not described by a triad interaction mechanism. Instead, energy was transferred to smaller wavelengths and, a few periods later, to standing evanescent modes in multiples of the primary frequency (greater than the ambient buoyancy frequencies) in the interaction region. The accumulation of the energy caused by these trapped modes within the interaction region resulted in the overturning of the density field. When the two rays had different properties (apart from the multiples of the forcing frequencies) the divisions of the forcing frequencies as well as the combination of the different frequencies were observed within the interaction region.

The model was validated by comparing the results with those from experimental studies. Further, the energy balance was conserved and the dissipation of energy was shown to be related to the degree of nonlinear interaction.

The behaviour of internal waves at a caustic level, turning point and critical layer have been investigated numerically. At a caustic reflection, a triad interaction was formed within the reflection region and the internal wave energy was transferred to lower frequencies (subharmonics). This resulted in a local subharmonic instability. One of the excited internal waves penetrated the caustic level and propagated downwards. This downward propagating wave then produced a second caustic where further reflection could take place. At a turning point, nonlinear interaction between the incident and reflected waves transferred energy to higher frequencies (evanescent trapped waves) which resulted in a superharmonic instability. At the critical level, energy was transferred to the mean flow. As the degree of nonlinearity increased, more energy was found to be transferred and overturning resulted due to a shear instability.

The collision rate of monodisperse solid particles in a turbulent gas is governed by a wide range of scales of motion in the flow. Recent studies have shown that large-scale energetic eddies are the dominant factor contributing to the relative velocity between two colliding particles (the turbulent transport effect), whereas small-scale dissipative eddies can enhance the collision rate significantly by inducing local non-uniform particle distribution (the accumulation effect). The turbulent transport effect is most noticeable when the particle inertial response time τp is of the order of the flow integral timescale and the accumulation effect is most pronounced when τp is comparable to the flow Kolmogorov time.

We study these two contributions separately through direct numerical simulations. The two effects are quantified carefully with a numerical procedure that is independent of the computation of average collision rate. This facilitates the study of not only the statistical description of the collision kernel, but also the relative contributions and modelling of the two physical effects. Simulations at several flow Reynolds numbers were performed to suggest a model for the accumulation effect. The data show that the accumulation effect scales linearly with flow Taylor microscale Reynolds number Rλ, while the theory for fully developed turbulence indicates that the maximum level of the turbulent transport effect scales with R1/2λ. Finally, an integrated model has been developed to predict the collision rate at arbitrary flow Reynolds numbers and particle inertia.

In studies of the turbulent air flow over water waves it is usually assumed that the effect of viscosity near the water surface is negligible, i.e. the Reynolds number, Re = u∗λ/v, is considered to be high. However, for short waves or low wind speeds this assumption is not valid. Therefore, a second-order turbulence closure that takes into account viscous effects is used to simulate the air flow. The model shows reasonable agreement with laboratory measurements of wave-induced velocity profiles. Next, the dependence of the dimensionless energy flux from wind to waves, or growth rate, on Re is investigated. The growth rate of waves that are slow compared to the wind is found to increase strongly when Re decreases below 104, with a maximum around Re = 800. The numerical model predictions are in good agreement with analytical theories and laboratory observations. Results of the study are useful in field conditions for the short waves in the spectrum, which are particularly important for remote sensing applications.

This paper discusses unsteady surface pressures on aircraft flaps and their correlation with far-field noise. Analyses are made of data from a 4.7% DC-10 aircraft model test, conducted in the 40 × 80 feet wind tunnel at NASA Ames Research Center. Results for various slat/wing/flap configurations and various flow conditions are discussed in detail to reveal major trends in surface pressure fluctuations. Spectral analysis, including cross-correlation/coherence, both among unsteady surface pressures and between far-field noise and near-field fluctuations, is used to reveal the most coherent motions in the near field and identify potential sources of noise related to flap flows. Dependencies of surface pressure fluctuations on mean flow Mach numbers, flap settings and slat angles are discussed. Dominant flow features in flap side edge regions, such as the formation of double-vortex structures, are shown to manifest themselves in the unsteady surface pressures as a series of spectral humps. The spectral humps are shown to correlate well with the radiated noise, indicating the existence of major noise sources in flap side edge regions. Strouhal number scaling is used to collapse the data with satisfactory results. The effects of flap side edge fences on surface pressures are also discussed. It is shown that the application of fences effectively increases the thickness of the flaps so that the double-vortex structures have more time to evolve. As a result, the characteristic timescale of the unsteady sources increases, which in turn leads to a decrease in the dominant frequency of the source process. Based on this, an explanation is proposed for the noise reduction mechanism of flap side edge fences.

Planar flow past multiple successive blades and wakes is studied for nearly aligned configurations with normal non-symmetry inducing lift. The typical blade lies relatively near the centreline of the oncoming wake from the preceding blade. The central motion over a wide parameter range is in condensed periodic boundary layers and wakes with fixed displacement, buried within surrounding incident shear flow. This is accompanied, however, by streamwise jumps in the pressure, velocity and mass flux, across the leading edge of each blade, a new and surprising feature which is supported by the combination of incident shears and a solid surface and which is related to the normal flow through the multi-blade system. The leading-edge jumps are required in order to satisfy the equi-pressure condition at the trailing edge. Computational results include separating flows and show the lift and drag, and these are followed by a short-blade analysis which captures the main flow properties explicitly. The results agree qualitatively with experiments and direct simulations for rotor blade flows. The jump feature also extends for example to a single blade immersed in the relatively large wake of an upstream blade.

In non-premixed turbulent combustion the reactive zone is localized at the stoichiometric surfaces of the mixture and may be locally approximated by a diffusion flame. Experiments and numerical simulations reveal a characteristic structure at the edge of such a two-dimensional diffusion flame. This ‘triple flame’ or ‘edge flame’ consists of a curved flame front followed by a trailing edge that constitutes the body of the diffusion flame. Triple flames are also observed at the edge of a lifted laminar diffusion flame near the exit of burners. The speed of propagation of the triple flame determines such important properties as the rate of increase of the flame surface in non-premixed combustion and the lift-off distance in lifted flames at burners. This paper presents an approximate theory of triple flames based on an approximation of the flame shape by a parabolic profile, for large activation energy and low but finite heat release. The parabolic flame path approximation is a heuristic approximation motivated by physical considerations and is independent of the large activation energy and low heat release assumptions which are incorporated through asymptotic expansions. Therefore, what is presented here is not a truly asymptotic theory of triple flames, but an asymptotic solution of a model problem in which the flame shape is assumed parabolic. Only the symmetrical flame is considered and Lewis numbers are taken to be unity. The principal results are analytical formulas for the speed and curvature of triple flames as a function of the upstream mixture fraction gradient in the limit of infinitesimal heat release as well as small but finite heat release. For given chemistry, the solution provides a complete description of the triple flame in terms of the upstream mixture fraction gradient. The theory is validated by comparison with numerical simulation of the primitive equations.

A scale-dependent dynamic subgrid-scale model for large-eddy simulation of turbulent flows is proposed. Unlike the traditional dynamic model, it does not rely on the assumption that the model coefficient is scale invariant. The model is based on a second test-filtering operation which allows us to determine from the simulation how the coefficient varies with scale. The scale-dependent model is tested in simulations of a neutral atmospheric boundary layer. In this application, near the ground the grid scale is by necessity comparable to the local integral scale (of the order of the distance to the wall). With the grid scale and/or the test-filter scale being outside the inertial range, scale invariance is broken. The results are compared with those from (a) the traditional Smagorinsky model that requires specification of the coefficient and of a wall damping function, and (b) the standard dynamic model that assumes scale invariance of the coefficient. In the near-surface region the traditional Smagorinsky and standard dynamic models are too dissipative and not dissipative enough, respectively. Simulations with the scale-dependent dynamic model yield the expected trends of the coefficient as a function of scale and give improved predictions of velocity spectra at different heights from the ground. Consistent with the improved dissipation characteristics, the scale-dependent model also yields improved mean velocity profiles.

Sand banks and sand waves are two types of sand structures that are commonly observed on an off-shore sea bed. We describe the formation of these features using the equations of the fluid motion coupled with the mass conservation law for the sediment transport. The bottom features are a result of an instability due to tide–bottom interactions. There are at least two mechanisms responsible for the growth of sand banks and sand waves. One is linear instability, and the other is nonlinear coupling between long sand banks and short sand waves. One novel feature of this work is the suggestion that the latter is more important for the generation of sand banks. We derive nonlinear amplitude equations governing the coupled dynamics of sand waves and sand banks. Based on these equations, we estimate characteristic features for sand banks and find that the estimates are consistent with measurements.

The propagation of weakly nonlinear, long internal wave fronts in a contraction is considered in the transcritical limit as a model for the establishment of virtual controls. It is argued that the appropriate equation to describe this process is a variable coefficient Korteweg–de Vries equation. The solutions of this equation are then considered for compressive and rarefaction fronts. Rarefaction fronts exhibit both normal and virtual control solutions. However, the interaction of compressive fronts with contractions is intrinsically unsteady. Here the dynamics take two forms, interactions with the bulk of the front and interactions with individual solitary waves separating off from a front trapped downstream of the contraction.