Existing studies of charged aerosol deposition have largely focused on isolated upper-airway or local truncated in vitro models, leaving whole-lung scale effects unresolved. Using a physiologically inspired multiscale airway-on-chip spanning conducting to acinar regions, we investigated the electrostatic contributions along with gravity, impaction, and diffusion. Our results show that charge significantly reshapes deposition patterns, enhancing early bronchiolar capture while reducing delivery to the distal acinus.

Pressure-driven flow of a dilute polymer solution has been numerically observed to support elastic turbulence which is organized around the interactions of localized versions of two-dimensional arrowhead traveling waves. Here, we isolate these spanwise-localized arrowheads for the first time. We find symmetric and asymmetric states, and identify a process in which these localized states split to spawn multiple arrowheads. These arrowheads have small velocities perpendicular to the flow suggesting they may be poor mixers.

The size of a detaching pendant drop is set by Tate’s law and depends only weakly on the nozzle radius. Here, we show that simple geometric confinement can trigger early detachment at significantly reduced volumes by introducing an additional capillary force. A minimal scaling law collapses the data across geometries, providing a robust and passive route to tune drop size without external actuation.

In this paper, we have identified and characterized a novel route to self-propulsion in which spontaneous shape changes give rise to symmetry-breaking in autophoretic colloids. By means of theoretical predictions and numerical simulations, we show that a deformable autophoretic filament with a uniform chemical profile, i.e., that is otherwise immotile, can achieve self- propulsion via a buckling instability. Our findings provide physical insight into the design of reconfigurable synthetic microswimmers and bio-inspired materials for applications such as cargo transport, drug delivery, or tissue scaffolding.

We consider hydrogel swelling while confined between two rigid walls in cases of (i) slip and (ii) no slip boundary conditions on the walls. Using the framework of large deformation poroelasticity, we find a fully nonlinear solution numerically in case (i) and a linear solution analytically in case (ii), and show that in case (ii) the hydrogel exerts substantially more force on the walls. This has application in the design of hydrogel-based actuators.

Brownian dynamics simulations of a bead-spring chain in a turbulent flow show that hydrodynamic interactions (HI) modify the stretching of polymers, owing to hydrodynamic shielding and conformation-dependent drag. HI delays the turbulence-induced migration between coiled and stretched states and alters the distribution of extension. Stiff chains stretch more while highly elastic chains stretch less, in the presence of HI, resulting in a steeper coil-stretch transition. These effects cannot be reproduced by adding HI to a dumbbell, because of its inability to form a physical coil, implying that dumbbell-based descriptions of polymer solutions must incorporate an extension-dependent drag.

Dispersing a second phase throughout a carrier fluid significantly alters the system’s bulk rheology, a critical consideration for the transport and mixing of multiphase systems comprising elastoviscoplastic (EVP) carrier fluids ubiquitous in industrial processes. This work investigates the role of elasticity and yield stress of the carrier fluid on the bulk rheology of droplet- and bubble-laden suspensions at dilute to semi-dilute concentrations under simple shear. Analysis of the stress budget and the spatial and size distribution of the dispersed phase elucidates the interplay between the carrier fluid’s EVP rheology and dispersed phase dynamics that influence the bulk rheology.

We investigate the flow of discontinuous shear-thickening fluids with S-shaped rheology in confined and disordered geometries. We demonstrate that, due to the emergence of highly viscous structures, such fluids tend to limit the flow rate at a given pressure gradient. In channel geometries containing obstacles, these structures form in the most constricted regions. In porous media, they also occur in the narrowest pore throats, but tend to be arranged transversely to the flow direction, which effectively creates a viscous barrier. We then study the formation and the spanning of these barriers, which exhibit properties similar to those of critical systems.

The eigenspectrum of bounded viscoelastic shearing flows comprises a continuous spectrum (CS) whose eigenvalues form continuous curves or line segments in the complex plane, in addition to a discrete spectrum of isolated eigenvalues. While the Oldroyd-B model possesses only two line-segment CS, we show that the more realistic FENE-P model admits up to six distinct CS. Our analytical predictions provide a framework for interpreting numerically computed spectra of viscoelastic shearing flows.

Transonic buffet is a shock-wave/boundary-layer interaction on wings involving coexisting and self-sustained 2D chordwise shock motion and 3D separation-driven spanwise buffet-cell dynamics. Using implicit LES and spectral modal analysis of infinite swept wings up to aspect ratio 3, we show that the 2D shock mode is insensitive to sweep, while sweep transforms a quasistationary low-frequency 3D mode at unswept conditions into a spanwise-travelling mode. The 3D mode shifts monotonically to higher Strouhal numbers with increasing sweep while retaining a fixed spanwise wavelength, and pronounced buffet cells are shown to arise only when mean flow separation at the shock is sufficiently strong.

Complex mutual interaction between two convective oscillators are studied. A variety of regular and irregular modes are found. The dynamics of large plates strongly depend on the depth of immersion and include convective pendulum mode with regular antiphase oscillations, irregular fluctuations, full stops and synchronized periodic movements. The dynamics of plates of relatively small size is fundamentally different. It is characterized by the modes with a pronounced intermittent character. The very specific behavior was observed during random walks, in which the plates perform small-scale chaotic oscillations, without breaking away from each other, as if they are on a flexible bundle.

Bubble dissolution in porous media controls key applications including geological carbon sequestration, groundwater remediation, and energy engineering. The classic Epstein-Plesset model for bubble dissolution in open space is invalid in porous medium. We reveal how porous structure fundamentally reshapes dissolution, and derive analytical solutions for three typical bubble morphologies (single-pore, strip-shaped, and block-shaped). Analytical solutions are well verified by experiments and numerical simulations. Our new theory offers critical theoretical support for optimizing subsurface gas storage and gaseous pollutant removal technologies.

Using theory and simulation we analyze the retraction of a highly slender Newtonian liquid sheet with surface covered by a surfactant monolayer surrounded by air primarily in the Stokes limit with 1/Oh=0 where Oh is the Ohnesorge number. As the two surfaces of the sheet remain planar for long times after retraction is initiated, a control volume analysis is used to analytically calculate the maximum film thickness and retraction velocity. The role of finite inertia is also studied and it is shown that rim formation is suppressed if Oh $(1+B_0{\mathrm{\Gamma }}_0)\gg L_0$ where $B_0$ and $L_0$ are the Boussinesq-Scriven number and initial sheet aspect ratio and ${\mathrm{\Gamma }}_0$ the initial surfactant concentration.

When the solute concentration is different in the two sides of a semipermeable membrane, solvent flows from the solute-depleted side to the solute-enriched side. More generally, when a vesicle is placed in a solute-concentration gradient, it experiences inward osmosis on the low-concentration side and outward osmosis on the high-concentration side. This paper investigates the resulting motion of the vesicle down the gradient.

Our experiments bridge classic raindrop shape theory and the behavior of contaminated drops in atmospheric and industrial aerosols. With two cameras, we reconstruct the three-dimensional shape, orientation, and oscillation of levitated nanofluid drops. Low nanoparticle loading can destabilize the interface whereas higher loading stabilizes it. Surfactant shifts the stability thresholds by sequestering particles.

Efficient atomization of multicomponent fuel droplets is central to cleaner and more efficient combustion technologies. This study investigates the bubble-driven breakup of acoustically levitated microemulsion droplets under laser heating, revealing alternative pathways to atomization in stable, practically relevant emulsion fuels. Using high-speed imaging, distinct fragmentation modes are identified that are governed by heating intensity, bubble growth dynamics, and hydrodynamic instabilities, including Faraday and Rayleigh–Taylor mechanisms. The findings provide new physical insight into atomization processes relevant to cleaner combustion and advanced spray technologies.

We derive a new theoretical estimate of the root mean square velocity of the electro-vortex flow (EVF) – a current-driven MHD flow – for high Reynolds number regime using an inertia-Lorentz balance in the vorticity transport equation. This estimate accounts for the dimension of the current collector, an important parameter that governs the EVF. There is an excellent agreement between the theory and numerical simulations performed using the custom-built code in OpenFOAM. We also explain the EVF characteristics using these results. Our numerical results reveal a distinct flow feature stemming from the domain finiteness at relatively higher current collector radii.

Most electrohydrodynamic droplet models assume passive ion transport and field-induced polarization. Here we incorporate chemically powered active interfacial ion transport into a Navier–Stokes – Poisson–Nernst–Planck – Cahn–Hilliard (NS-PNP-CH) phase-field framework using an energy–dissipation–input formulation. The resulting persistent charge asymmetry reorganizes electric fields and stresses, enabling controlled deformation, breakup and recoalescence, as well as droplet separation under shear.

We show that the classical Rayleigh–Bénard instability and nonequilibrium electroconvection become strongly intertwined at the limiting current in charge-selective systems. The resulting interaction lowers the instability threshold and removes the short-wave singularity characteristic of electroconvective instability.

River confluences play an important role in riverine systems and riverine ecology. Eddy resolving simulations are used to investigate in a systematic way the effect of increasing bed discordance of the minor tributary in a confluence of simplified geometry. Increasing the bed discordance reduces the coherence and number of streamwise oriented vortices forming on the minor tributary side of the mixing interface and promotes the formation of a near-bed intrusion of mixed fluid into the minor tributary side of the main channel near the confluence apex. This is the main mechanism that is responsible for the increase in the rates of mixing between the two streams with increasing bed discordance.

Inertia–gravity waves generated by tidal flow over topography transfer energy and momentum, shaping ocean and atmospheric dynamics. We derive a general solution including non-traditional Coriolis effects, finite depth effects, and non-hydrostatic terms. Non-traditional effects broaden where waves can exist and enhance conversion, especially at low latitudes and given weak stratification. These results expand predictions for dynamics on icy sattelites and other planetary applications.

This work investigates the dynamics of thick liquid films flowing down vertical fibres with insoluble surfactants. A one-dimensional long-wave model, validated against the full two-dimensional system, reveals three stability regimes depending on the Marangoni number ($Ma$): Rayleigh–Plateau dominated at low $Ma$, complete stabilization at intermediate $Ma$, and Marangoni-induced instability at high $Ma$. Nonlinear analysis shows these behaviors result from the competition between Marangoni convection and the difference between interface velocity and wave speed, providing new insight into surfactant-controlled film stability.

Turbulent puffs in pipe flow persist for a prolonged duration before suddenly transitioning to laminar flow via viscous exponential decay. Prior to the onset of relaminarization, the configuration of sectional streamlines indicates the existence of multiple saddles and nodal points near the wall. During relaminarization, they move from the near-wall region and may undergo saddle-node bifurcations that destroy saddle-node pairs. In such cases, the saddle/nodal distance follows the Riccati equation.

Freeze–thaw cycles drive complex interactions between objects and moving solid–liquid interfaces. Using experimental model systems of oil-in-water emulsions and polystyrene particle suspensions, we reveal the occurence of hysteresis: Solid particles drift from their initial positions after one freeze-thaw cycle, while deformable oil droplets largely return to their initial positions with reversible shape changes. Our theoretical model captures these trends, highlighting the complexity of freeze–thaw dynamics.

In recent years, the importance of previously overlooked electrostatic effects has opened a new perspective in the field of dynamic wetting. While spontaneous charging has been investigated in sliding drops, here, we show that it can substantially affect capillary wetting instabilities. The capillary breakup of a liquid bridge wetting a surface causes electrostatic charging that slows down the breakup dynamics and leads to spontaneous motion of satellite drops. Our results highlight the central importance of electrostatics in dewetting processes beyond sliding drops.

While Ostwald ripening is widely studied, its reaction-controlled regime on unpinned substrates lacks a rigorous analytical foundation. Here, the authors establish a theoretical framework by deriving the kinetic equations and explicit bubble growth rates under constant contact angle conditions. Validated by Lattice Boltzmann simulations, their theory reveals a striking “reversed volume ripening” on heterogeneous substrates, where a smaller-volume bubble completely consumes a larger one. This fundamentally proves that ripening is governed by curvature-driven chemical potential rather than volume.

Stretch an elastic sheet between your hands and let go; it snaps back. Thin water films do the same, rupturing almost as soon as they form. The authors had a simple idea: instead of changing the liquid or coating the surface, just hold the film edges. Two laser-engraved grooves on a plain acrylic plate pin a film of pure water over more than thirty centimeters. When the grooves end, the film ruptures and drips in a steady rhythm. The stability comes not from chemistry, but from geometry.

We investigated the dynamics of a Newtonian liquid jet issued from a long circular nozzle into a gaseous environment. While the jet contracts at high Reynolds number, it swells at low Reynolds number. To analyze this, we derived an integral momentum balance for the flow in both the nozzle and jet. The swell at low Reynolds number is associated with an excess integral wall shear stress near nozzle exit relative to perfect Poiseuille flow. Numerical simulations revealed self-similar behavior of the flow within the nozzle, which is explained from the stick-slip transition at the nozzle lip and the subsequent development of a boundary layer along the jet interface.

Cerebrospinal-fluid motion and solute transport in the subarachnoid space are strongly influenced by trabeculae. Most existing flow and transport descriptions rely on homogenized porous-media models whose accuracy under physiological oscillatory-flow conditions remains uncertain. This study benchmarks such macroscopic models against direct numerical simulations in a canonical fibrous channel. The main advance is to show that the unsteady Brinkman equation accurately captures the flow field, whereas current quasi-steady transport models fail to represent key unsteady shear-enhanced dispersion mechanisms.

Flexible appendages performing metachronal rowing can efficiently transport fluid, but the role of elasticity in impulsively driven systems remains unclear. This study numerically investigates the transient flow induced by an array of wall-mounted flexible blades under impulsive metachronal rowing. Two regimes emerge depending on the ratio of natural to rowing frequency, corresponding to linear response and deformation saturation. Maximum transport occurs near Ca = 1, where tip-shed vortices are optimally positioned and captured by neighboring blades, reinforcing thrust. These insights provide guidance for bio-inspired propulsion and microfluidic transport.

Single-beam acoustic tweezers based on focused ultrasound provide a compact and biocompatible platform for single cell trapping, yet stable three-dimensional trapping is often hindered by acoustic bulk streaming at high frequencies. Here, we develop a unified theoretical– numerical framework to quantify the competition between acoustic radiation force and streaming-induced drag force across viscous-to-inertial flow regimes. We derive pressure-scaling laws for streaming velocity and show trapping performance varies non-monotonically with focal pressure, contrary to conventional expectations. These findings offer practical guidelines for optimizing high-frequency acoustic tweezers for robust cell trapping.

We show that grooved surfaces fully wetted by a relatively inviscid lubricant may exhibit a large apparent slip length. Exploring this singular limit, we map the key asymptotic regimes defined by the encapsulation height and the submerged ridge area fraction. Our theory bridges classical superhydrophobic models with a newly predicted giant-slip regime. This transition is described by an exterior flow problem where the thin lubricant films wetting the ridges are effectively replaced by a Navier-slip condition.

This study investigates the effect of compliant walls on the turbulent heat transfer in channel flows over viscous-hyperelastic walls. We show that the compliant wall leads to an increase not only of the momentum transfer but also of the heat transfer, and that the heat transfer enhancement is favorable compared to the momentum one

We present a thermodynamically consistent continuum theory for magnetic-particle transport and capture in high-gradient fields. Derived from a free-energy functional, the model couples magnetism, mass transport, and flow without empirical shutoff rules, so that shielding, anisotropic deposition, and boundary-layer confinement emerge naturally from particle-field feedback. A Mason number phase diagram then organizes capture into thermodynamic, transitional, and dynamic regimes, providing a predictive basis for High-Gradient-Magnetic-Separation design and optimization.

We measure particle motion in a two-dimensional fluid-saturated granular bed, below a submerged oscillating plate, to test mechanisms responsible for the change in shape of its surface. As the plate moves upward, some grains in a region of the bed are mobilized and dragged by the fluid both vertically and horizontally through a matrix of fixed particles. The measured particle streamlines are fit over many cycles to those in the experiment, using the pressure field in the bed that results from the solution of Darcy’s equation. When different horizontal and vertical permeabilities are used in the Darcy flow relations, we find relatively good agreement with the predictions of the mixture flow.

In unstably stratified mixing layers laden with heated particles, analysis of the interphase kinetic energy transfer shows that the contribution of mean power supplied by heated particles to the fluid is stronger than that of the fluctuating part at the early stage, especially near the mixing interfaces. The inclined streaks of particles clustering in the vertical direction become stronger as the flow evolves. As a result, the fluctuating power increases with flow evolution especially near the upper stream, which could be interpreted as the enhancement of updraft buoyancy production and perturbations generated by the inclined streaks.

Sedimentation of nonspherical particles is typically governed by orientation-dependent interactions. We show that cube-shaped particle suspensions recover classical Stokesian scaling laws associated with spheres. This behavior emerges from geometric symmetry, which suppresses anisotropy and promotes isotropic microstructure. The results identify symmetry, not sphericity, as the key determinant of collective sedimentation dynamics.

We report the first direct numerical simulations using a recently introduced exact director-based formulation of the equations of motion for inertial spheroids in turbulence. The results reveal that particle shape and finite inertia induce complex, nonmonotonic trends in preferential clustering. Gravity markedly alters these dynamics, enhancing small-scale clustering for rod- and disk-like particles while suppressing clustering for nearly spherical particles. We confirm that at weak inertia, rod-like spheroids tend to align with the flow’s major stretching direction, whereas disk-like spheroids align with its major shrinking direction.

Dense suspensions of non-Brownian particles encountered in industrial processes like concrete mixing, chocolate refining, and ceramic processing are well known to show abrupt jamming transition under shear flow, yet, the effects of particle size distribution on this transition remain poorly understood. Using Discrete Element Method-based simulations in two dimensions, the authors show that polydisperse suspensions undergo critical rigidity transition preceding shear jamming, with scaling exponents consistent with percolation theory. Remarkably, the order parameter, susceptibility, and the microstructural properties of polydisperse suspensions are found to be identical to those of the statistically equivalent bidisperse systems

Particle-laden turbulence is commonly modeled through two-way coupling, but inter-particle collisions can become significant when inertial particles accumulate near the walls. By comparing the two-way and four-way coupled point-particle direct numerical simulations in channel flow, this work shows that inter-particle collisions can substantially weaken near-wall particle accumulation, enhance turbulence attenuation, and promote drag reduction. These effects arise from the enhanced particle dispersion and the amplified slip-velocity fluctuations, highlighting the importance of four-way coupling for accurately modeling particle-laden turbulent flows.

Twin-swirl flows can generate various vortex breakdown structures depending on the swirling direction, strength, and the momentum ratio between the swirling streams. The present study uses scale-resolving simulations to analyze the mechanism of radial pressure gradient formation, role of entrainment in mixing, and identification of coherent structures in such flows. We show that while the centripetal acceleration dominates the radial pressure gradient formation with a single swirler, the contribution of advection and turbulence components are also significant in twin-swirl flows. We also show that the entrainment velocity is better estimated by the rms components than the mean velocity.

Quantized vortex filaments in Bose superfluids act as line-like sources for microhydrodynamic (low Reynolds number) normal-fluid motion on scales that standard turbulence grids cannot resolve. We develop a two-level multiscale framework that couples a filtered normal-fluid solver to an explicit microhydrodynamic Stokes Linear Response (LRT), incorporating these effects self-consistently into both vortex dynamics and the resolved normal-fluid equations. The approach enables efficient superfluid-turbulence computations for laboratory, cryogenic, and astrophysical settings.

Unstable periodic orbits provide a dynamical-systems view of coherent structures in wall-bounded turbulence. In large-eddy simulations of plane Couette flow, we identify an orbit in which streamwise rolls trigger streak instability, generate spanwise vortices, and stretch them toward the wall. The associated Lyapunov vectors localize in high-strain shear layers, linking vortex dynamics to orbital instability.

Estimating minimum pressure from particle measurements is important for vortex flows, especially when particles are sparse, noisy, or absent near vortex cores. While physics-informed neural networks have been used for flow-field assimilation, their accuracy for minimum-pressure recovery under controlled particle density, particle distribution, incomplete observations, and noise in spatial coordinates and velocity measurements has not been systematically quantified. This work fills that gap using analytical two-/three-dimensional vortices and a turbulent flow of interacting counter-rotating vortices of unequal strength obtained from Large-Eddy Simulations.

An accurate inflow description for atmospheric boundary-layer (ABL) large-eddy simulation (LES) is critical, yet conventional mapping of homogeneous turbulence to inhomogeneous ABLs can distort spatial correlations and disrupt turbulence continuity. In this article, an adaptive energy-preserving mapping (AEPM) strategy is proposed, which preserves target energy profiles while maintaining spatial-correlation properties. An $a$ $priori$ numerical test and three LES cases (neutral/unstable building flows and a flat-plate boundary-layer flow) demonstrate the robustness of the AEPM method and its capability to reproduce realistic inlet and downstream turbulence statistics.

We identify the essential role of temporal filtering in enabling data-driven closure models that reproduce turbulent dynamics in the energy-containing range. The scope of the present study differs fundamentally from conventional subgrid-scale modeling, which relies on the universality of small-scale dynamics. While the model constructed on training data preprocessed with a temporal low-pass filter stably and accurately captures the chaotic dynamics of the largest eddies in turbulence, the one trained without temporal low-pass filtering exhibits steady or periodic behavior and fails to capture the chaotic dynamics of the energy-containing range.

Twin pulsed jets are highly efficient at transferring energy, making them particularly attractive for applications ranging from underwater propulsion and maneuvering to the filling of heart cavities. However, little is known about their dynamic interactions within a hemispherical elastic cavity. Distinct flow regimes are identified here based on formation time and jet spacing. These regimes include short-time decay, decay at the lower wall, wall rebound, and wall rebound with secondary vortices. These findings enhance our understanding of the complex flow patterns generated by certain medical devices and pathological conditions, as well as their effects on cardiac function and performance.

Separation from curved surfaces is known to be a viscous phenomenon, governed by boundary layer dynamics. In his 1904 seminal paper, where he introduced the boundary layer, Prandtl hinted at the possibility of having an inviscid separation criterion, “from external conditions” outside the boundary layer. Here, we present a candidate for such a long-sought criterion. We show that there is a unique separation angle that minimizes the curvature in the outer potential flow. Moreover, this minimizing angle coincides with the experimentally observed one in the flow over a circular cylinder in the subcritical regime, where the averaged flow characteristics force are independent of Reynolds number.

Direct numerical simulations and full analytical theory reveal how the geometry of obstacles determines the onset of vortex nucleation in quantum fluids such as Bose–Einstein condensates and related superfluid systems. In particular, the flows around obstacles with sharp corners such as walls and wells display a time-irreversible transition when the velocity exceeds a critical value which is well below the sound speed. This work paves the way for the study of skin friction in superfluid systems.

Low-frequency unsteadiness (LFU) in separated flows is often linked to drag modulation and recirculation-bubble dynamics, but its governing mechanisms remain unclear. This work introduces a reduced-order, physics-based framework that describes LFU through kinetic-energy transport within the mean recirculation region using vorticity-based quantities derived from the rotational Navier–Stokes equations. For the wake of a normal plate, the analysis reveals that Bernoulli-energy transport and Lamb-vector dynamics govern the charging and discharging processes underlying LFU.

In final states of freely decaying two-dimensional turbulence over topography, background flows follow a linear potential vorticity (PV)-streamfunction relationship, but localized vortices have remained poorly understood. We show that the vortices locked to topographic bumps and dips follow a robust, “sinh”-like relationship. We propose an empirical model—a superposition of topographic background flow and Gaussian vortices—that accurately reproduces the quasistationary final states. Linear stability analyses of these stationary vortex solutions explain the observed vortex-topography correlations across different energy levels.

Turbulent flows beneath water surfaces control key processes in the Earth system, yet linking the motion of the free surface to subsurface dynamics has relied on numerical simulations at Reynolds numbers far smaller than in natural flows. By combining particle image velocimetry with free-surface profilometry in a laboratory setting, we take a large step towards bridging the gap to real-world flows, reaching Reynolds numbers two orders of magnitude higher than DNS simulations. Surface features remain strongly correlated with subsurface turbulence, with correlations that are near instantaneous but highly nonlocal in space, persisting up to two integral length scales beneath the surface.

Bendable thin structures such as floating modular pontoons can be displaced, rotated and deformed by incoming gravity waves. We propose a diffractionless theory to calculate the second order mean yaw moment on slender elastic structures in waves. In the case of non-moored, freely drifting floaters, the mean yaw moment can rotate the structure to a preferential orientation with respect to the angle of incidence. Using our theory, we can predict this preferential orientation and how it varies with floater shape and its bending modulus.

Inertia–gravity waves generated by flow over topography transfer energy and momentum between the ocean and its boundaries. Finite depth forces the wave to be a sum of vertical modes, and when the gravest mode has a length scale comparable to the topographic source, wave conversion and drag are suppressed. We show how a relaxation of the hydrostatic approximation and acoustic approximation influence this suppression.

Identifying the origin of dangerous events and perturbations in fluid systems is a central challenge in many inverse problems. This work develops a unified framework combining linear and quadratic sensitivity analysis to create a positional embedding for one-shot localization of sources in weakly nonlinear flows, with unknown intensity. By extending classical adjoint-based approaches beyond the linear regime, the method significantly improves accuracy and efficiency in detecting sources under nonlinear interactions in fluid systems.

We demonstrate for the first time that it is possible to access in real-time (live measurements) two-dimensional instantaneous velocity fields with both high spatial resolution and high sampling frequency. Using a standard Optical Flow algorithm properly optimized, standard 4Mp snapshots can be processed live up to 460 Hz. Moreover, using the proper experimental settings, it also becomes possible to access 1 vector per pixel, leading to very high spatial resolution. Apart from considerable gain in computing time and power consumption, this approach also unlocks new experiments like very low frequency measurements, closed-loop flow control, or rare events detection.

Fluids in nature are usually turbulent and contain small particles, a phenomenon observed in paper production, rain formation, astrophysics, and the oceans, among other places. These particles are more often than not nonspherical, such as fibers in paper. Particle orientation in the flow determines how the flow drags them and, eventually, how the particles distribute in space and orient. We use a symmetry-based simplification, analogous to a classical description of neutrally buoyant spheroids, to introduce a new framework for flows with nonspherical particles, which yields a compact set of evolution equations with fewer degrees of freedom.