Volcanic fissure eruptions typically start with the opening of a linear fissure that erupts along its entire length, following which, activity localises to one or more isolated vents within a few hours or days. Localisation is important because it influences the spatiotemporal evolution of the hazard posed by the eruption. Previous work has proposed that localisation can arise through a thermoviscous fingering instability driven by the strongly temperature dependent viscosity of the rising magma. Here, we explore how thermoviscous localisation is influenced by the irregular geometry of natural volcanic fissures. We model the pressure-driven flow of a viscous fluid with temperature-dependent viscosity through a narrow fissure with either sinusoidal or randomised deviations from a uniform width. We identify steady states, determine their stability and quantify the degree of flow enhancement associated with localised flow. We find that, even for relatively modest variations of the fissure width (${\lt } 10$ %), the non-planar geometry supports strongly localised steady states, in which the wider parts of the fissure host faster, hotter flow, and the narrower parts of the fissure host slower, cooler flow. This geometrically driven localisation differs from the spontaneous thermoviscous fingering observed in planar geometries and can strongly impact the localisation process. We delineate the regions of parameter space under which geometrically driven localisation is significant, showing that it is a viable mechanism for the observed localisation under conditions typical of basaltic eruptions, and that it has the potential to dominate the effects of spontaneous thermoviscous fingering in these cases.

The present work experimentally investigates the interaction of a buoyant (rigid) spherical particle with a single translating (water) vortex ring, focusing on the effects of particle-to-vortex core size ratio ($D_p/D_{c,o}$) on both the particle dynamics and ring dynamics ($D_p$ = particle diameter, $D_{c,o}$ = vortex core diameter). These interactions are studied for $D_p/D_{c,o}$ = 0.6–1.7, over ring Reynolds numbers ($Re={\varGamma }/{\nu }$; $\varGamma$ = ring circulation) of 6000–67 300. As the buoyant particle comes close to the ring, it gets captured into the low-pressure vortex core, and the interaction begins. The particle within the core undergoes radial oscillation, spins and translates along the ring’s azimuthal axis. As $D_p/D_{c,o}$ increases, the particle undergoes higher-amplitude radial oscillation and a relatively shorter azimuthal translation. The differences in the particle size and its motion within the ring lead to large differences in the ring’s dynamics. A larger particle is seen to lead to a higher ring disruption, substantially reducing the ring’s convection speed and azimuthal enstrophy, which are seen to scale as $(D_p/D_{c,o})^{2.3}Re^{-0.37}$ and $(D_p/D_{c,o})^{1.3}Re^{-0.25}$, respectively. The ring disruption is significant above $D_p/D_{c,o}\approx$ 1.0, beyond which the ring fragments, with up to 60 % drop in convection speed and 90 % drop in enstrophy, at low $Re$, as compared with the base ring. These results for the rigid particle size effects on the vortex ring dynamics are more dramatic than for a deforming bubble. Our results could help to better understand and model buoyant particle (and bubble) interactions with coherent structures in turbulence.

Experimental investigation of the Rayleigh–Taylor instability (RTI) and its dependence on initial conditions has been challenging, primarily due to the difficulty of creating a well-defined gaseous interface. To address this, a novel soap film technique was developed to create a discontinuous two-dimensional SF$_6$air interface with precisely controlled initial conditions. High-order modes were superimposed on a long-wavelength perturbation to study the influence of initial conditions on RTI evolution. Experiments conducted at Atwood numbers ranging from 0.26 to 0.66 revealed that bubble growth shows a weak dependence on both initial conditions and Atwood numbers, whereas spike growth is more influenced by these factors. Spike growth accelerates as the wavenumber of the imposed high-order modes decreases and/or the Atwood number increases. To quantify these effects, a variation on the previously developed potential flow model was applied, capturing the suppression of high-order modes and Atwood number dependence on RTI growth. In turbulent flow, the self-similar factors of bubbles and spikes exhibit minimal sensitivity to initial conditions. However, in relation to the Atwood number, the self-similar factors of bubbles (or spikes) demonstrate negligible (or significant) dependence. Comparisons with literature revealed that two-dimensional flows yield lower self-similar factors than three-dimensional flows. Furthermore, the discontinuity of the initial interface in this study, achieved through the soap film technique, results in faster spike growth compared with previous studies involving a diffusive initial interface. These findings provide critical insights into the nonlinear dynamics of RTI and underscore the importance of well-characterised initial conditions in experimental studies.

Non-spherical bubble collapses near solid boundaries, generating water hammer pressures and shock waves, were recognized as key mechanisms for cavitation erosion. However, there is no agreement on local erosion patterns, and cavitation erosion damage lacks quantitative analysis. In our experiments, five distinct local erosion patterns were identified on aluminium sample surfaces, resulting from the collapse of laser-induced cavitation bubbles at moderate stand-off distances of $0.4\leqslant \gamma \leqslant 2.2$, namely bipolar, monopolar, annular, solar-halo and central. Among them, the bipolar and monopolar patterns exhibit the most severe cavitation erosion when the toroidal bubbles undergo asymmetrical collapse along the circumferential direction during the second cycle. Shadowgraphy visualization revealed that asymmetrical collapse caused shockwave focusing through head-on collision and oblique superposition of wavefronts. This led to the variations in toroidal bubble radii and the positions of maximum erosion depth not matching at certain stand-off distances. Both initial plasma asymmetry and bubble–wall stand-off distance were critical in determining circumferential asymmetrical collapse behaviours. At large initial aspect ratios, the elliptical jet tips form during the contraction process, resulting in the toroidal bubble collapsing from regions with smaller curvature radii, ultimately converging to the colliding point along the circumferential direction. Our three-dimensional simulations using OpenFOAM successfully reproduce the key features of circumferentially asymmetrical bubble collapse. This study provides new insights into the non-spherical near-wall bubble collapse dynamics and provides a foundation for developing predictive models for cavitation erosion.

We investigate enstrophy variations by collapse of point vortices in an inviscid flow and, in particular, focus on the enstrophy dissipation that is a significant property characterising two-dimensional (2-D) turbulent flows. To reveal the vortex dynamics causing the enstrophy dissipation, we consider the dynamics of point vortices, whose vorticity is concentrated on points and dynamics on the inviscid flow, governed by the point-vortex system. The point-vortex system has self-similar collapsing solutions, which are expected to be a key to understand the enstrophy dissipation, but the collapsing process cannot be described by solutions to the 2-D Euler equations. We thus consider the 2-D filtered-Euler equations, which are a regularised model of the 2-D Euler equations, and their point-vortex solutions. The preceding studies (Gotoda and Sakajo, J. Nonlinear Sci. 2016, vol. 26, pp. 1525–1570, Gotoda and Sakajo, SIAM J. Appl. Math. 2018, vol. 78, 2105–2128) have proven that there exist three point-vortex solutions to the filtered model such that they converge to self-similar collapsing orbits in the three point-vortex system and dissipate the enstrophy at the event of collapse in the zero limit of the filter parameter. In this study, we numerically show that the enstrophy dissipation by the collapse of point vortices could occur for the four and five vortex problems in a filtered model. Moreover, we show the detailed convergence process of the point vortices for gradually decreasing filter parameters, which provides a new insight for the three vortex problem. In addition, numerical computations suggest that the enstrophy dissipation is caused by collapse of separated point vortices with the negative interactive energy.