Active colloidal particles create flow around them due to non-equilibrium processes on their surfaces. In this paper, we infer the activity of such colloidal particles from the flow field created by them via deep learning. We first explain our method for one active particle, inferring the $2s$ mode (or the stresslet) and the $3t$ mode (or the source dipole) from the flow field data, along with the position and orientation of the particle. We then apply the method to a system of many active particles. We find excellent agreements between the predictions and the true values of activity. Our method presents a principled way to predict arbitrary activity from the flow field created by active particles.

Flame–flame interactions in continuous combustion systems can induce a range of nonlinear dynamical behaviours, particularly in the thermoacoustic context. This study examines the mutual coupling and synchronisation dynamics of two thermoacoustic oscillators in a model gas-turbine combustor operating within a stochastic environment and subjected to external sinusoidal forcing. Experimental observations from two flames in an annular combustor reveal the emergence of dissimilar limit cycles, indicating localised lock-in of thermoacoustic oscillators. To interpret these dynamics, we introduce a coupled stochastic oscillator model with sinusoidal forcing terms, which highlights the critical role of individual synchronisation in enabling local lock-in. Furthermore, through stochastic system identification using this phenomenological low-order model, we mathematically demonstrate that a transition towards self-sustained oscillations can be driven solely by enhanced mutual coupling under external forcing. This combined experimental and modelling effort offers a novel framework for characterising complex coupled flame dynamics in practical combustion systems.

Elastoviscoplastic (EVP) fluid flows are driven by a non-trivial interplay between the elastic, viscous and plastic properties, which under certain conditions can transition the otherwise laminar flow into complex flow instabilities with rich space–time-dependent dynamics. We discover that under elastic turbulence regimes, EVP fluids undergo dynamic jamming triggered by localised polymer stress deformations that facilitate the formation of solid regions trapped in local low-stress energy wells. The solid volume fraction $\phi$, below the jamming transition $\phi\lt\phi_J$, scales with $\sqrt {\textit{Bi}}$, where $\textit{Bi}$ is the Bingham number characterising the ratio of yield to viscous stresses, in direct agreement with theoretical approximations based on the laminar solution. The onset of this new dynamic jamming transition $\phi \geqslant \phi _J$ is marked by a clear deviation from the scaling $\phi \sim \sqrt {\textit{Bi}}$, scaling as $\phi \sim \exp {\textit{Bi}}$. We show that this instability-induced jamming transition – analogous to that in dense suspensions – leads to slow, minimally diffusive and rigid-like flows with finite deformability, highlighting a novel phase change in elastic turbulence regimes of complex fluids.

We study the mixing of passive scalars in a velocity field generated by selected-eddy simulations (SES), an approach where only a randomly selected subset of spectrally distributed modes obey Navier–Stokes dynamics. The Taylor Reynolds number varies from 140 to 400 and the Schmidt number ($Sc$) varies from 0.25 to 1. By comparing the results with direct numerical simulations (DNS), we show that most statistics are captured with as low as $0.5\,\%$ of Navier–Stokes modes in the velocity field. This includes scalar gradients, spectra, structure functions and their departures from classical scaling due to intermittency. The results suggest that all modes need not be resolved to accurately capture turbulent mixing for $Sc\leqslant 1$ scalars.

An experimental study was conducted to investigate the impingement of a vortex ring onto a porous wall by laser-induced fluorescence and particle image velocimetry. The effects of different Reynolds numbers (${{Re}}_{\it\Gamma } = 700$ and $1800$) and hole diameters ($d_{h}^{*} = 0.067$, $0.10$, $0.133$ and $0.20$) on the flow characteristics were examined at a constant porosity ($\phi = 0.75$). To characterise fluid transport through a porous wall, we recall the model proposed by Naaktgeboren, Krueger & Lage (2012, J. Fluid Mech., vol. 707, 260–286), which shows rough agreement with the experimental results due to the absence of vortex ring characteristics. This highlights the need for a more accurate model to correlate the losses in kinetic energy ($\Delta E^{*}$) and impulse ($\Delta I^{*}$) resulting from the vortex ring–porous wall interaction. Starting from Lamb’s vortex ring model and considering the flow transition from the upstream laminar state to the downstream turbulent state caused by the porous wall disturbance, a new model is derived theoretically: $\Delta E^{*} = 1 - k(1 - \Delta I^{*})^2$, where $k$ is a parameter dependent on the dimensionless core radius $\varepsilon$, with $k = 1$ when no flow state change occurs. This new model effectively correlates $\Delta E^{*}$ and $\Delta I^{*}$ across more than 70 cases from current and previous experiments, capturing the dominant flow physics of the vortex ring–porous wall interaction.

We investigate flow-induced choking in soft Hele-Shaw cells comprising a fluid-filled gap in between a rigid plate and a confined block of elastomer. Fluid injected from the centre of the circular rigid plate flows radially outwards, causing the elastomeric block to deform, before exiting through the cell rim. The pressure in the fluid deforms the elastomer, increasing the size of the gap near the inlet, and decreasing the gap near the cell rim, because of volume conservation of the solid. At a critical injection flow rate, the magnitude of the deformation becomes large enough that the flow is occluded entirely at the rim. Here, we explore the influence of elastomer geometry on flow-induced choking and, in particular, the case of a thick block with radius smaller than its depth. We show that choking can still occur with small-aspect-ratio elastomers, even though the confining influence of the back wall that bounds the elastomer becomes negligible; in this case, the deformation length scale is set by the radial size of the cell rather than the depth of the block. Additionally, we reveal a distinction between flow-induced choking in flow-rate-controlled flows and flow-rate-limiting behaviour in pressure-controlled flows.

This study uses the diffusion analogy (Miyake, Sci. Rep., 5R-6, 1965, Univ. of Washington, Seattle, USA) to predict the full growth behaviour of internal boundary layers (IBLs) induced by a roughness change for neutrally – and especially stably – stratified boundary layers with finite thickness. The physics of the diffusion analogy shows that the streamwise variation of the IBL thickness is dictated by $\sigma _w/U$ at the interface, where $\sigma _w$ and $U$ represent wall-normal Reynolds stress and mean streamwise velocity, respectively. The existing variants of the model, summarised by Savelyev & Taylor (2005, Boundary-Layer Meteorol., vol. 115, pp. 1–25), are tailored to IBLs confined within the constant shear stress layer. To extend the applicability of the model to the outer region, we investigate the relation between $\sigma _w/U$ and $U/U_\infty$ in the outer region across varying stratification, where $U_\infty$ is the free-stream velocity. Our analysis reveals that wind tunnel data from a number of facilities collapse onto a master curve when $\sigma _w/U$ is premultiplied by a height-independent parameter, which is a function of the ratio of Monin–Obukhov length to the boundary layer thickness. The scaled $\sigma _w/U$ decreases inversely with $U/U_\infty$ in the surface layer, transitioning to a linear decrease as $U/U_\infty$ increases. The new model, which integrates these findings, along with the effects of streamline displacement and acceleration, captures the complete characteristics of IBLs as they develop within turbulent boundary layers of finite thickness.

Turbulent flows exhibit large intermittent fluctuations from inertial to dissipative scales, characterised by multifractal statistics and breaking the statistical self-similarity. It has recently been proposed that the Navier–Stokes turbulence restores a hidden form of scale invariance in the inertial interval when formulated for a dynamically (nonlinearly) rescaled quasi-Lagrangian velocity field. Here we show that such hidden self-similarity extends to the large-eddy-simulation (LES) approach in computational fluid dynamics (CFD). In particular, we show that classical subgrid-scale models, such as implicit or explicit Smagorinsky closures, respect the hidden scale invariance at all scales – both resolved and subgrid. In the inertial range, they reproduce the hidden scale invariance of Navier–Stokes statistics. These properties are verified very accurately by numerical simulations and, beyond CFD, turn LES into a valuable tool for fundamental turbulence research.

Viscous fingering is a hydrodynamic instability typically occurring when a less viscous fluid displaces a more viscous one and which deforms the interface between the two fluids into finger-shaped intrusions. For miscible fluids, the fingering pattern is usually followed visually by adding a passive dye into one of the two fluids. The reverse displacement of a less viscous fluid by a more viscous one is classically stable, featuring a planar interface. Here, we show experimentally that in some cases, the dye can actively modify the viscosity of a polymer solution and trigger fingering in the reverse displacement. This dye-induced destabilisation is shown to be due to double-diffusive effects triggering a non-monotonic viscosity profile with a maximum because the dye diffuses faster than the polymer.

This study investigated the cylindrically divergent Rayleigh–Taylor instability (RTI) on a liquid–gas interface and its dependence on initial conditions. A novel hydrophobic technique was developed to generate a two-dimensional water–air interface with controlled initial conditions. The experimental configuration utilised high-pressure air injection to produce uniform circumferential acceleration. Amplitude measurements over time revealed that the cylindrical RTI growth depends strongly on the azimuthal wavenumber. Experimental results demonstrated that surface tension significantly suppresses the liquid–gas cylindrical RTI, even inducing a freeze-out and oscillatory perturbation growth – a phenomenon observed for the first time. Spectrum analysis of the interface contours demonstrated that the cylindrical RTI evolves in a weakly nonlinear regime. Linear and weakly nonlinear models were derived to accurately predict the time-varying interface amplitudes and high-order modes. The linear model was further used to determine conditions for unstable, freeze-out and oscillatory solutions of the cylindrically divergent RTI. These findings offer valuable insights into manipulating hydrodynamic instabilities in contracting/expanding geometries using surface tension.

Turbulence closures are essential for predictive fluid flow simulations in both natural and engineering systems. While machine learning offers promising avenues, existing data-driven turbulence models often fail to generalise beyond their training datasets. This study identifies the root cause of this limitation as the conflation of generalisable flow physics and dataset-specific behaviours. We address this challenge using symbolic regression, which yields interpretable, white-box expressions. By decomposing the learned corrections into inner-layer, outer-layer and pressure-gradient components, we isolate universal physics from flow-specific features. The model is trained progressively using high-fidelity datasets for plane channel flows, zero-pressure-gradient turbulent boundary layers (ZPGTBLs), and adverse pressure-gradient turbulent boundary layers (PGTBLs). For example, direct application of a model trained on channel flow data to ZPGTBLs results in incorrect skin friction predictions. However, when only the generalisable inner-layer component is retained and combined with an outer-layer correction specific to ZPGTBLs, predictions improve significantly. Similarly, a pressure-gradient correction derived from PGTBL data enables accurate modelling of aerofoil flows with both favourable and adverse pressure gradients. The resulting symbolic corrections are compact, interpretable, and generalise across configurations – including unseen geometries such as aerofoils and Reynolds numbers outside the training set. The models outperform baseline Reynolds-averaged Navier–Stokes closures (e.g. the Spalart–Allmaras and shear stress transport models) in both a priori and a posteriori tests. These results demonstrate that explicit identification and retention of generalisable components is key to overcoming the generalisation challenge in machine-learned turbulence closures.

The linear stability of a thermally stratified fluid layer between horizontal walls, where non-Brownian thermal particles are injected continuously at one boundary and extracted at the other – a system known as particulate Rayleigh–Bénard (pRB) – is studied. For a fixed volumetric particle flux and minimal thermal coupling, reducing the injection velocity stabilises the system when heavy particles are introduced from above, but destabilises it when light particles are injected from below. For very light particles (bubbles), low injection velocities can shift the onset of convection to negative Rayleigh numbers, i.e. heating from above. Particles accumulate non-uniformly near the extraction wall and in regions of strong vertical flow, aligning with either wall-impinging or wall-detaching zones depending on whether injection is at sub- or super-terminal velocity. The increase of the volumetric particle flux always enhances these effects.

In this work, we study the effect of flow curvature, or angular momentum, on the propagation and trapping characteristics of near-inertial waves (NIWs) in a curved front. The curved front is idealised as a baroclinic vortex in cyclogeostrophic balance. Motivated by ocean observations, we employ a Gaussian base flow, which by construction possesses a shield of oppositely signed vorticity surrounding its core, and we consider both cyclonic and anticyclonic representations of this flow. Following two main assumptions, i.e. that (i) the horizontal wavelength of the NIW is smaller than the length scale of the background flow (the WKBJ approximation), and (ii) the vertical wavelength of the NIW is smaller than the radial distance of interest, we derive the NIW dispersion relation and discuss the group velocity and direction of energy propagation. We show that the curvature can (i) increase the critical depth and horizontal extent of the trapping region, (ii) reduce NIW activity at the centre of the anticyclonic vortex core and enhance it in the cyclonic shield surrounding the core for high curvatures, (iii) lead to NIW trapping in the anticyclonic shield surrounding the cyclonic core, and (iv) increase the available band of NIW frequencies that are trapped. The solutions from the ray-tracing method are supported by numerical solutions of the governing equations linearised about the cyclogeostrophic base state. Finally, these methods are applied to an idealised model of oceanic mesoscale Arctic eddies showing an increase in the critical depth of trapping. Our results – while applied to polar eddies – equally apply at lower latitudes in both oceans and atmospheres, highlighting the potential importance of flow curvature in controlling the propagation of NIW energy.

Stochastic resonance (SR) is universal phenomenon, where noise amplifies a weak periodic signal in bistable nonlinear systems, with wide applications in biology, climate science, engineering etc., although in fluid dynamics it remains underexplored. Recently, we unexpectedly found SR above non-modal elastic instability onset in an inertialess viscoelastic channel flow, where it emerges on the top of a chaotic streamwise velocity power spectrum $E_u$ due to its interaction with white-noise spanwise velocity power spectrum $E_w$ and weak elastic waves. These three conditions necessary for SR emergence differ from those required for the classical SR emergence mentioned above. Here, we consider SR in an inertialess viscoelastic channel flow with a smoothed inlet causing order of magnitude lower noise intensity than in our former studies. Our observations reveal that SR appears at the same conditions mentioned above, where SR is found just upon the instability onset in a lower subrange of a transition regime, in contrast, here, SR persists across all flow regimes – transition, elastic turbulence and drag reduction. Furthermore, we provide experimental evidence that SR, presented by a sharp peak in $E_u$, manifests as either a standing or propagating wave in the $x$-direction, with a rather uniform amplitude of streamwise velocity fluctuations and zero propagation velocity in the $z$-direction. These findings reveal a new mechanism underpinning the transition to a chaotic channel flow of viscoelastic fluids and establish SR as a robust framework for understanding complex flow dynamics. This work opens new avenues for exploring SR in other nonlinear systems and practical applications such as mixing enhancement and flow control in industrial and biological contexts.

Ultra-thin liquid sheets generated by impinging two liquid jets are crucial high-repetition-rate targets for laser ion acceleration and ultra-fast physics, and serve widely as barrier-free samples for structural biochemistry. The impact of liquid viscosity on sheet thickness should be comprehended fully to exploit its potential. Here, we demonstrate experimentally that viscosity significantly influences thickness distribution, while surface tension primarily governs shape. We propose a thickness model based on momentum exchange and mass transport within the radial flow, which agrees well with the experiments. These results provide deeper insights into the behaviour of liquid sheets and enable accurate thickness control for various applications, including atomization nozzles and laser-driven particle sources.

We study the dispersion of bubble swarms rising in initially quiescent water using three-dimensional Lagrangian tracking of deformable bubbles and tracer particles in an octagonal bubble column. Two different bubble sizes (3.5 mm and 4.4 mm) and moderate gas volume fractions ($0.52\,\%{-}1.20\,\%$) are considered. First, we compare the dispersion inside bubble swarms with that for single-bubble cases, and find that the horizontal mean squared displacement (MSD) in the swarm cases exhibits oscillations around the asymptotic scaling predicted for a diffusive regime. This occurs due to wake-induced bubble motion; however, the oscillatory behaviour is heavily damped compared to the single-bubble cases due to the presence of bubble-induced turbulence (BIT) and bubble–bubble interactions in the swarm. The vertical MSD in bubble swarms is nearly an order of magnitude faster than in the single-bubble cases, due to the much higher vertical fluctuating bubble velocities in the swarms. We also investigate tracer dispersion in BIT, and find that concerning the time to transition away from the ballistic regime, larger bubbles with a higher gas void fraction transition earlier than tracers, consistent with Mathai et al. (2018, Phys. Rev. Lett., vol. 121, 054501). However, for bubble swarms with smaller bubbles and a lower gas void fraction, they transition at the same time. This differing behaviour is due to the turbulence being more well-mixed for the larger bubble case, whereas for the smaller bubble case, the tracer dispersion is highly dependent on the wake fluctuations generated by the oscillating motion of nearby bubbles.

Previous studies claimed that the non-monotonic effects of wettability came mainly from the heterogeneity of geometries or flow conditions on multiphase displacements in porous media. For macroscopic homogeneous porous media, without permeability contrast or obvious preferential flow pathways, most pore-scale evidence showed a monotonic trend of the wettability effect. However, this work reports transitions from monotonic to non-monotonic wettability effects when the dimension of the model system rises from two-dimensional (2-D) to three-dimensional (3-D), validated by both the network modelling and the microfluidic experiments. The mechanisms linking the pore-scale events to macroscopic displacement patterns have been analysed through direct simulations. For 2-D porous media, the monotonic effect of wettability comes from the consistent transition pattern for the full range of capillary numbers $Ca$, where the capillary fingering mode transitions to the compact displacement mode as the contact angle $\theta$ decreases. Yet, it is indicated that the 3-D porous geometries, even though homogeneous without permeability contrast or obvious preferential flow pathways, introduce a different $Ca$–$\theta$ phase diagram with new pore-scale events, such as the coupling of capillary fingering with snap-off during strong drainage, and frequent snap-off events during strong imbibition. These events depend strongly on geometric confinements and capillary numbers, leading to the non-monotonicity of wettability effects. Our findings provide new insights into the multiphase displacement dependent on wettability in various natural porous media and offer design principles for engineering artificial porous media to achieve desired immiscible displacement behaviours.

We investigate the dynamics of a pair of rigid rotating helices in a viscous fluid, as a model for bacterial flagellar bundle and a prototype of microfluidic pumps. Combining experiments with hydrodynamic modelling, we examine how spacing and phase difference between the two helices affect their torque, flow field and fluid transport capacity at low Reynolds numbers. Hydrodynamic coupling reduces the torque when the helices rotate in phase at constant angular speed, but increases the torque when they rotate out of phase. We identify a critical phase difference, at which the hydrodynamic coupling vanishes despite the close spacing between the helices. A simple model, based on the flow characteristics and positioning of a single helix, is constructed, which quantitatively predicts the torque of the helical pair in both unbounded and confined systems. Finally, we show the influence of spacing and phase difference on the axial flux and the pump efficiency of the helices. Our findings shed light on the function of bacterial flagella and provide design principles for efficient low-Reynolds-number pumps.

We study the homogeneous isotropic turbulence of a shear-thinning fluid modelled by the Carreau model, and show how the variable viscosity affects the multiscale behaviour of the turbulent flow. We show that Kolmogorov theory can be extended to such non-Newtonian fluids, provided that the correct choice of average is taken when defining the mean Kolmogorov scale and dissipation rate, to properly capture the effect of the variable viscosity. Thus the classical phenomenology à la Kolmogorov can be observed in the inertial range of scale, with the energy spectra decaying as $k^{-5/3}$, with $k$ being the wavenumber, and the third-order structure function obeying the $4/5$ law. The changing viscosity instead strongly alters the small scale of turbulence, leading to an enhanced intermittent behaviour of the velocity field.

We present the measurements of the decay of stationary turbulence at Reynolds numbers based on the Taylor microscale $Re_{\lambda }=493, 599, 689$ produced in a large-scale von Kármán flow using stereoscopic particle image velocimetry. First, steady-state conditions were established, after which the impellers were simultaneously and abruptly stopped, and the turbulent decay was measured over 10–20 impeller rotation periods. A total of 258 decay experiments were performed. The temporal evolution of the ensemble-averaged turbulent kinetic energy (TKE) showed excellent agreement over all $Re_{\lambda }$ and exhibited two distinct phases: a short, initial transition phase where the TKE remained almost constant due to the inertia of the flow and lasted approximately $0.4$ impeller rotations, followed by a classical power-law decay. To extract the decay exponent $n$, a curve-fitting function based on a one-dimensional energy spectrum was used, and successfully captured the entire measured decay process. A value $n=1.62$ was obtained based on ensemble-averaged TKE. However, different decay exponents were found for individual velocity components: $n=1.38$ for the axial component consistent with various reports in the literature and Loitsiansky’s prediction ($n=1.43$), and $n=1.99$ for the radial and circumferential components indicating saturation/confinement effects. Similarly, the longitudinal integral length scale in the axial direction grew as $L\propto t^{2/7}$, whereas it remained nearly constant in the radial direction. Finally, the evolution of the ensemble-averaged velocity gradients showed that after the impellers were stopped, the mean flow pattern persisted for a short time before undergoing a large-scale reversal before the onset of the turbulent decay.