Showing papers on "Open-channel flow published in 2006"

Scaling of the velocity fluctuations in turbulent channels up to Reτ=2003

Abstract: A new numerical simulation of a turbulent channel in a large box at Reτ=2003 is described and briefly compared with simulations at lower Reynolds numbers and with experiments. Some of the fluctuation intensities, especially the streamwise velocity, do not scale well in wall units, both near and away from the wall. Spectral analysis traces the near-wall scaling failure to the interaction of the logarithmic layer with the wall. The present statistics can be downloaded from http://torroja.dmt.upm.es/ftp/channels. Further ones will be added to the site as they become available.

Proposal for Numerical Benchmarking of Fluid-Structure Interaction between an Elastic Object and Laminar Incompressible Flow

Abstract: We describe new benchmark settings for the rigorous evaluation of different methods for fluid-structure interaction problems. The configurations consist of laminar incompressible channel flow around an elastic object which results in self-induced oscillations of the structure. Moreover, characteristic flow quantities and corresponding plots are provided for a quantitative comparison.

Channel flow, ductile extrusion and exhumation in continental collision zones: an introduction

Abstract: The channel flow model aims to explain features common to metamorphic hinterlands of some collisional orogens, notably along the Himalaya–Tibet system. Channel flow describes a protracted flow of a weak, viscous crustal layer between relatively rigid yet deformable bounding crustal slabs. Once a critical low viscosity is attained (due to partial melting), the weak layer flows laterally due to a horizontal gradient in lithostatic pressure. In the Himalaya–Tibet system, this lithostatic pressure gradient is created by the high crustal thicknesses beneath the Tibetan Plateau and ‘normal’ crustal thickness in the foreland. Focused denudation can result in exhumation of the channel material within a narrow, nearly symmetric zone. If channel flow is operating at the same time as focused denudation, this can result in extrusion of the mid-crust between an upper normal-sense boundary and a lower thrust-sense boundary. The bounding shear zones of the extruding channel may have opposite shear sense; the sole shear zone is always a thrust, while the roof shear zone may display normal or thrust sense, depending on the relative velocity between the upper crust and the underlying extruding material. This introductory chapter addresses the historical, theoretical, geological and modelling aspects of channel flow, emphasizing its applicability to the Himalaya–Tibet orogen. Critical tests for channel flow in the Himalaya, and possible applications to other orogenic belts, are also presented. The hinterlands of collisional orogens are often characterized by highly strained, high-grade metamorphic rocks that commonly display features consistent with lateral crustal flow and extrusion of material from mid-crustal depths towards the orogenic foreland. A recent model for lateral flow of such weak mid-crustal layers has become widely known as the ‘channel flow’ model. The channel flow model has matured through efforts by several research groups and has also been applied to a variety of geodynamic settings. Thermal-mechanical modelling of collision zones, including the Himalayan–Tibetan system, has brought the concept of channel flow to the forefront of orogenic studies. Original contributors to the concept of channel flow initiated an important paradigm shift (Kuhn 1979), from geodynamic models of continental crust with finite rheological layering to the more encompassing channel flow model. This time-dependent midto lower crustal flow process, which will be reviewed in this chapter, may progress into foreland fold-and-thrust tectonics in the upper crust, thereby providing a spatial and temporal link between the early development of a metamorphic core in the hinterland and the foreland fold-and-thrust belt at shallower structural levels. Outcomes and implications of such a viscous flowing middle to lower crust include a dynamic coupling between mid-crustal and surface processes, and limitations to accurate retro-deformation of orogens (non-restorable orogens, e.g. Jamieson et al. 2006). This Special Publication contains a selection of papers that were presented at the conference ‘Channel flow, extrusion, and exhumation of lower to mid-crust in continental collision zones’ hosted by the Geological Society of London at Burlington House, in December 2004. Because most of the ongoing debate on crustal flow focuses on the Cenozoic age Himalaya–Tibet collisional system, some of the key questions that are addressed in this volume include the following. . Does the model for channel flow in the Himalaya–Tibet system concur with all available geological and geochronological data? From: LAW, R. D., SEARLE, M. P. & GODIN, L. (eds) Channel Flow, Ductile Extrusion and Exhumation in Continental Collision Zones. Geological Society, London, Special Publications, 268, 1–23. 0305-8719/06/$15.00 # The Geological Society of London 2006. . How do the pressure–temperature-time (P-T-t) data across the crystalline core of the Himalaya fit with the proposed channel flow? . Are the microstructural fabric data (pure shear and simple shear components) compatible with crustal extrusion (thickening or thinning of the slab)? . If the channel flow model is viable for the Himalaya–Tibet system, what may have initiated channel flow and ductile extrusion? . Why did the extrusion phase of the Himalayan metamorphic core apparently cease during the late Miocene–Pliocene? . Are some of the bounding faults of the potential channel still active, or were they recently active? . Is the Himalayan channel flow model exportable to other mountain ranges? This introductory paper addresses the historical, theoretical, geological and modelling aspects of crustal flow in the Himalaya–Tibet orogen. Critical tests for crustal flow in the Himalaya, and possible applications to other orogenic belts, are presented and difficulties associated with applying these tests are discussed. Personal communication citations (pers. comm. 2004) identify comments expressed during the conference. The Himalaya–Tibetan plateau system The Himalaya–Tibet system initiated in Early Eocene times, following collision of the Indian and Eurasian plates (see Hodges (2000) and Yin & Harrison (2000) for reviews). The collision resulted in closure of the Tethyan Ocean, southward imbrication of the Indian crust, and northward continental subduction of Indian lower crust and mantle beneath Asia. The collision thickened the southern edge of the Asian crust to 70 km, and created the Tibetan Plateau, the largest uplifted part of the Earth’s surface with an average elevation of 5000 m (Fielding et al. 1994). The Himalayan orogen coincides with the 2500km-long topographic front at the southern limit of the Tibetan Plateau. It consists of five broadly parallel lithotectonic belts, separated by mostly north-dipping faults (Fig. 1). The Himalayan metamorphic core, termed the Greater Himalayan sequence (GHS), is bounded by two parallel and opposite-sense shear zones that were both broadly active during the Miocene (Hubbard & Harrison 1989; Searle & Rex 1989; Hodges et al. 1992, 1996). The Main Central thrust (MCT) zone marks the lower boundary of the GHS, juxtaposing the metamorphic core above the underlying Lesser Himalayan sequence. The South Tibetan detachment (STD) system defines the upper boundary roof fault of the GHS, marking the contact with the overlying unmetamorphosed Tethyan sedimentary sequence. The apparent coeval movement of the MCT and STD, combined with the presence of highly sheared rocks and high grade to migmatitic rocks within the GHS, has led many workers to view the GHS as a north-dipping, southward-extruding slab of mid-crustal material flowing away from the thick southern edge of the Tibetan Plateau, towards the thinner foreland fold-thrust belt. Dynamics of channel flow The concepts of crustal extrusion and channel flow originated in the continental tectonics literature in the early 1990s. Unfortunately, these two processes are often referred to interchangeably without justification. One of the main points that emerged from the Burlington House conference was that a distinction between channel flow and crustal extrusion must be made. Parallel versus tapering bounding walls on channel flow and/or extrusion processes, and how these processes may replenish over time, are two resolvable parameters that are critical for distinguishing channel flow from extrusion. Brief definitions and overviews of the two processes are presented below. A more detailed overview of the mechanics of the related processes is provided by Grujic (2006).

Crustal flow modes in large hot orogens

Abstract: Crustal-scale channel flow numerical models support recent interpretations of Himalayan—Tibetan tectonics proposing that gravitationally driven channel flows of low-viscosity, melt-weakened, middle crust can explain both outward growth of the Tibetan Plateau and ductile extrusion of the Greater Himalayan Sequence. We broaden the numerical model investigation to explore three flow modes: homogeneous channel flow (involving laterally homogeneous crust); heterogeneous channel flow (involving laterally heterogeneous lower crust that is expelled and incorporated into the mid-crustal channel flow); and the hot fold nappes style of flow (in which mid-/lower crust is forcibly expelled outward over a lower crustal indentor to create fold nappes that are inserted into the mid-crust). The three flow modes are members of a continuum in which the homogeneous mode is driven by gravitational forces but requires very weak channel material. The hot fold nappe mode is driven tectonically by, for example, collision with a strong crustal indentor and can occur in crust that is subcritical for homogeneous flows. The heterogeneous mode combines tectonic and gravitationally driven flows. Preliminary results also demonstrate the existence and behaviour of mid-crustal channels during advancing and retreating dynamical mantle lithosphere subduction. An orogen temperature—magnitude (T-M) diagram is proposed and the positions of orogens in T-M space that may exhibit the flow modes are described, together with the characteristic positions of a range of other orogen types.

Stabilization of Flow Boiling in Microchannels Using Pressure Drop Elements and Fabricated Nucleation Sites

Abstract: The flow boiling process suffers from severe instabilities induced due to nucleation of vapor bubbles in a superheated liquid environment in a minichannel or a microchannel. In an effort to improve the flow boiling stability, several modifications are introduced and experiments are performed on 1054197 m parallel rectangular microchannels (hydraulic diameter of 332 m) with water as the working fluid. The cavity sizes and local liquid and wall conditions required at the onset of nucleation are analyzed. The effects of an inlet pressure restrictor and fabricated nucleation sites are evaluated as a means of stabilizing the flow boiling process and avoiding the backflow phenomenon. The results are compared with the unrestricted flow configurations in smooth channels. DOI: 10.1115/1.2165208

Population trends of spanwise vortices in wall turbulence

Abstract: The present effort documents the population trends of prograde and retrograde spanwise vortex cores in wall turbulence outside the buffer layer. Large ensembles of instantaneous velocity fields are acquired by particle-image velocimetry in the streamwise–wall-normal plane of both turbulent channel flow at , the retrograde populations differ considerably beyond this point, highlighting the influence of the opposing wall in channel flow.

A theoretical prediction of friction drag reduction in turbulent flow by superhydrophobic surfaces

Abstract: We present a theoretical prediction for the drag reduction rate achieved by superhydrophobic surfaces in a turbulent channel flow. The predicted drag reduction rate is in good agreement with results obtained from direct numerical simulations at Reτ≃180 and 400. The present theory suggests that large drag reduction is possible also at Reynolds numbers of practical interest (Reτ∼105–106) by employing a hydrophobic surface, which induces a slip length on the order of ten wall units or more.

Wall Pressure and Shear Stress Spectra from Direct Simulations of Channel Flow

Abstract: Wall pressure and shear stress spectra from direct numerical simulations of turbulent plane channel flow are presented in this paper. Simulations have been carried out at a series of Reynolds numbers up to Re? = 1440, which corresponds to Re = 6:92 x 10(4) based on channel width and centerline velocity. Single-point and two-point statistics for velocity, pressure, and their derivatives have been collected, including velocity moments up to fourth order.§ The results have been used to study the Reynolds number dependence of wall pressure and shear stress spectra. It is found that the point spectrum of wall pressure collapses for Re? ? 360 under a mixed scaling for frequencies lower than the peak frequency of the frequency-weighted spectrum, and under viscous scaling for frequencies higher than the peak. Point spectra of wall shear stress components are found to collapse for Re? ? 360 under viscous scaling. The normalized mean square wall pressure increases linearly with the logarithm of Reynolds number. The rms wall shear stresses also increase with Reynolds number over the present range, but suggest some leveling off at high Reynolds number.

A dynamic subgrid-scale eddy viscosity model with a global model coefficient

Abstract: In the present study, a dynamic subgrid-scale eddy viscosity model is proposed for large eddy simulation of turbulent flows in complex geometry. A subgrid-scale eddy viscosity model recently proposed by Vreman [Phys. Fluids 16, 3670 (2004)] which guarantees theoretically zero subgrid-scale dissipation for various laminar shear flows, is considered as a base model. A priori tests with the original Vreman model show that it predicts the correct profile of subgrid-scale dissipation in turbulent channel flow but the optimal model coefficient is far from universal. A dynamic procedure of determining the model coefficient is proposed based on the “global equilibrium” between the subgrid-scale dissipation and the viscous dissipation. An important feature of the proposed procedure is that the model coefficient determined is globally constant in space but varies only in time. A posteriori tests of the proposed dynamic model are conducted through large eddy simulations of forced isotropic turbulence at Reλ=103, tur...

Subgrid modeling in particle-laden channel flow

Abstract: Direct numerical simulation (DNS) and large-eddy simulation (LES) of particle-laden turbulent channel flow, in which the particles experience a drag force, are investigated for two subgrid models and several Reynolds and Stokes numbers. In this flow, turbophoresis leads to an accumulation of particles near the walls. The objectives of the work are to investigate the accuracy of the subgrid models studied with respect to particle behavior and to explain the observed particle behavior predicted by the different models. The focus is on particle dispersion and mean particle motion in the direction normal to the walls of the channel. For a low Reynolds number, it is shown that the turbophoresis and particle velocity fluctuations are reduced compared to DNS, if the filtered fluid velocity calculated in the LES is used in the particle equation of motion. This is a combined effect of the disregard of the subgrid scales in the fluid velocity and the inadequacy of the subgrid model. Better agreement with DNS is obt...

Normal stress-driven migration and axial development in pressure-driven flow of concentrated suspensions

Abstract: Pressure-driven flow of a noncolloidal suspension is studied in two-dimensional channel and axisymmetric circular pipe geometries at bulk solids fractions of 0.2 ≤ ϕ ≤ 0.5 . Flows are modeled by the “suspension balance” approach, consisting of mass and momentum balances for the bulk suspension and particle phase. For particles in Newtonian fluid, cross-stream motion is driven by spatial variation of particle phase normal stresses. The particle phase stress model is based strictly upon the computed rate of strain, with a nonlocal contribution to the normal stress. Two solution procedures for the suspension flow equations are described. The first is a “solve–evolve” scheme based upon a full two-dimensional solution of the unsteady, axially varying behavior using a conservative finite volume method to solve the bulk mass and momentum conservation equations. The flow solution is coupled to an explicit update (evolve) step of the particle conservation equation. The second is a nonconservative but efficient marching solution for the asymptotically steady, but axially varying, problem. Predicted axial variation of the particle fraction, velocity and pressure gradient, as well as the fully developed profiles in channel and pipe flows are presented. The rate of axial development is strongly dependent upon the ratio of particle size to channel half-width (or pipe radius), ɛ ≡ a / B (or a / R ). The agreement of marching method and full model solutions is very close for the cases studied; both agree quantitatively well with available experimental results, including axial development in the pipe flow, where the model predicts the second normal stress difference to influence the migration. Migration in a 2:1 contraction flow provides an illustration of a flow where the full solution is required.

Sustained sub-laminar drag in a fully developed channel flow

Abstract: It is shown, by direct numerical simulations, that the skin-friction drag in a fully developed channel can be sustained below that corresponding to the laminar profile when the flow is subjected to surface blowing and suction in the form of an upstream travelling wave. A key mechanism that induces the sub-laminar drag is the creation of positive (negative) Reynolds shear stress in the wall region, where normally negative (positive) Reynolds shear stress is expected given the mean shear. This mechanism is contained in the linearized Navier–Stokes equations, thus allowing linear analysis of the observed phenomena. When applied to a fully developed turbulent channel flow, skin-friction drag is also significantly reduced by an upstream travelling wave, demonstrating that the surface blowing and suction in the form of such a wave is also effective in fully developed turbulent flows. Consideration of the energy budget shows a possibility of net drag reduction in turbulent channel flows with the present open-loop control.

Controllable gas-liquid phase flow patterns and monodisperse microbubbles in a microfluidic T-junction device

Abstract: This letter describes the gas-liquid phase flow patterns and the mechanism of generation of monodisperse microbubbles in a T-junction microfluidic device using the crossflowing shear-rupturing technique. The bubble size is ranged from 100 to 500μm. The air phase states as isolate air slugs, “pearl necklaces,” periodic isolate bubbles, zig-zag bubble patterns, and multiple-bubble layer can be observed in the wider measured channel. The bubble size relates with the continuous phase flow velocity and viscosity as Vb∝1∕(μcuc), while being almost independent of surface tension γ and air phase flow rate Qg, for the conditions used in this work. The bubble formation mechanism by using the crossflowing shear-rupturing technique is different from the hydrodynamic flow focusing and both geometry-dominated breakup techniques. Our system provides independent control of both the size and volume fraction of dispersed bubbles.

Mechanism of flow instability and transition to turbulence

Abstract: In this paper, a new mechanism of flow instability and turbulence transition is proposed for wall bounded shear flows. It is stated that the total energy gradient in the transverse direction and that in the streamwise direction of the main flow dominate the disturbance amplification or decay. Thus, they determine the critical condition of instability initiation and flow transition under given initial disturbance. A new dimensionless parameter K for characterizing flow instability is proposed which is expressed as the ratio of the energy gradients in the two directions for the flow without energy input or output. It is suggested that flow instability should first occur at the position of K max which may be the most dangerous position. This speculation is confirmed by Nishioka et al.'s experimental data. Comparison with experimental data for plane Poiseuille flow and pipe Poiseuille flow indicates that the proposed idea is really valid. It is found that the turbulence transition takes place at a critical value of K max of about 385 for both plane Poiseuille flow and pipe Poiseuille flow, below which no turbulence will occur regardless the disturbance. More studies show that the theory is also valid for plane Couette flows which holds a critical value of K max of about 370.

Channel flow and continental collision tectonics: an overview

Abstract: The principle of channel flow as defined in fluid dynamics has been used in continental geodynamics since the 1980s. The basic equations for one-dimensional flow introduced to geologists by Turcotte and Schubert were further developed by several research groups to meet the needs of specific studies. The most substantive differences among numerical models are results of different solutions for flow in crust, developed for different boundary conditions. The concept of channel flow has met with strong opposition and criticism from geophysicists and modellers. Although it is difficult to prove unambiguously that there is an active weak channel, it is still the most successful model to explain and predict the tectonics, metamorphism and exhumation of high-grade terranes in some orogens. Moreover, the concept of channel flow has stimulated novel approaches to the study of both the tectonics and metamorphism of large, hot orogens and the interaction between tectonic and surface processes. The concept of channelized flow of a weak crustal layer has been applied to various tectonic settings (Godin et al. 2006): (a) asthenospheric counterflow; (b) lower crustal channels; (c) intra-crustal channels; (d) subduction channels; (e) salt tectonics. In this overview only the concept of intra-crustal channels in collision orogens will be discussed, while detailed discussion of the range of proposed geologically relevant solutions to Stoke’s equation will be presented elsewhere. In active collision orogens the channel flow model has been used to explain the coupling between crust and mantle, strain in the crust, metamorphism, synorogenic exhumation of high-grade terranes and landscape evolution. Some numerical modelling strongly supports existence of a weak crustal layer (e.g. Royden 1996; Clark & Royden 2000; Beaumont et al. 2001a, b; 2004, 2006; Clark et al. 2005) while others oppose it (e.g. Toussaint & Burov 2004; Toussaint et al. 2004). Similarly, geophysical data can be interpreted both in favour of an active loweror mid-crustal channel (Nelson et al. 1996; Klemperer 2006) or against it (e.g. Flesch et al. 2005; Hilley et al. 2005). In the models the details of channel flow are sensitive to the flow laws assumed to operate in the crust, the mechanical properties of bounding crustal layers and density distribution in the crust. A number of analytical solutions to Stoke’s equation have been derived and experiments were developed for fundamentally different boundary conditions: conclusions will thus be model-dependent. The Himalaya–Tibet orogen is the most studied putative example of both activeand palaeochannel flow. The Greater Himalayan Sequence (GHS), also referred to as Higher Himalayan Crystallines, is a sequence of amphiboliteto granulite-facies orthoand paragneisses, migmatites and syntectonic leucogranites. The GHS is both underlain and overlain by greenschist and lower grade metasediments and forms the metamorphic core of the Himalaya (e.g. Hodges 2000, 2006; Jamieson et al. 2004, 2006). The GHS is bounded by the Main Central thrust (MCT) at the base and by the South Tibetan detachment (STD) at the top. The two shear zones are subparallel, have opposite senses of shear and have operated coevally over an extended period of time (Hodges 2000; Godin et al. 2006). The change from burial to exhumation of the GHS is reflected in the change in shear sense along the upper bounding shear zone. The earlier phase of deformation is dominated by thrust-sense shear zones while the return channel flow and ductile extrusion are characterized by normal-sense shear along the roof of the exhumed channel. The presence of melt during both southand north-directed shearing at the base and top of the GHS respectively (Fig. 1; e.g. Grujic et al. 1996, 2002; Davidson et al. 1997; Daniel et al. 2003; Harris et al. 2004) shows that high-temperature metamorphism and anatexis were an integral part of the exhumation process of the GHS (Godin et al. 2006; Hollister & Grujic 2006). Two closely related terms, ductile extrusion and channel flow, are used in the literature in regard to tectonics of a weak crustal layer. Kinematic models for both are based on the concept of a pair of coeval subparallel dip-slip shear zones with opposite senses of shear bounding a crustal layer with significantly lower strength or viscosity than the bounding layers. The concept of channel flow is based on the physical laws of fluid dynamics with a particular set of boundary conditions. Ductile extrusion is more difficult to define From: LAW, R. D., SEARLE, M. P. & GODIN, L. (eds) Channel Flow, Ductile Extrusion and Exhumation in Continental Collision Zones. Geological Society, London, Special Publications, 268, 25–37. 0305-8719/06/$15.00 # The Geological Society of London 2006. precisely and uniquely and will probably be of different forms in different orogens.

Measurements of Unsteady Wake Interference Between Tandem Cylinders

Abstract: A multi-phase, experimental study in the Basic Aerodynamics Research Tunnel at the NASA Langley Research Center has provided new insight into the unsteady flow interaction around cylinders in tandem arrangement Phase 1 of the study characterized the mean and unsteady near-field flow around two cylinders of equal diameter using 2-D Particle Image Velocimetry (PIV) and hot-wire anemometry These measurements were performed at a Reynolds number of 166 x 10(exp 5), based on cylinder diameter, and spacing-to-diameter ratios, L/D, of 1435 and 37 The current phase, Phase 2, augments this dataset by characterizing the surface flow on the same configurations using steady and unsteady pressure measurements and surface flow visualization Transition strips were applied to the front cylinder during both phases to produce a turbulent boundary layer upstream of the flow separation For these flow conditions and L/D ratios, surface pressures on both the front and rear cylinders show the effects of L/D on flow symmetry, pressure recovery, and the location of flow separation and attachment Mean streamlines and instantaneous vorticity obtained from the PIV data are used to explain the flow structure in the gap and near-wake regions and its relationship to the unsteady surface pressures The combination of off-body and surface measurements provides a comprehensive dataset to develop and validate computational techniques for predicting the unsteady flow field at higher Reynolds numbers

Improvement of delayed detached-eddy simulation for LES with wall modelling

Abstract: Adjustments are proposed of the Delayed Detached Eddy Simulation (DDES) approach to turbulence. They preserve the DDES capabilities particularly for natural DES uses, and resolve the mismatch of the logarithmic layers discovered earlier for the basic DES technique when used for Wall-Modelled Large-Eddy Simulation (WMLES) of attached flows. The adjustments are defined both for the Spalart-Allmaras and the Menter SST models. The first one concerns the definition of the LES length scale in general for anisotropic grids near a wall, and makes use of the wall distance along with the grid spacing; it clearly benefits even the Smagorinsky model. The second one manages the blending of RANS and LES behaviour within a WMLES to advantage, greatly increasing the resolved turbulence activity near the wall, and finely adjusting the resolved logarithmic layer. This is seen in channel flow over a wide Reynolds-number range, and through some grid variations. Tests show that the new method, although somewhat more complex, returns the desired behaviour not only in channel-flow LES, but also in channel-flow RANS, in a backward-facing-step case with side-by-side LES and RANS regions, and over an airfoil in deep stall.

A self‐regulating model of bedrock river channel geometry

Abstract: [1] The evolution of many mountain landscapes is controlled by the incision of bedrock river channels. While the rate of incision is set by channel shape through its mediation of flow, the channel shape is itself set by the history of bedrock erosion. This feedback between channel geometry and incision determines the speed of landscape response to tectonic or climatic forcing. Here, a model for the dynamics of bedrock channel shape is derived from geometric arguments, a normal flow approximation for channel flow, and a threshold bed shear stress assumption for bedrock abrasion. The model dynamics describe the competing effects of channel widening, tilting, bending, and variable flow depth. Transient solutions suggest that channels may take ∼1–10 ky to adapt to changes in discharge, implying that channel disequilibrium is commonplace. If so, landscape evolution models will need to include bedrock channel dynamics if they are to probe the effects of climate change.

The changing morphology of an open lava channel on Mt. Etna

Abstract: An open channel lava flow on Mt. Etna (Sicily) was observed during May 30–31, 2001. Data collected using a forward looking infrared (FLIR) thermal camera and a Minolta-Land Cyclops 300 thermal infrared thermometer showed that the bulk volume flux of lava flowing in the channel varied greatly over time. Cyclic changes in the channel's volumetric flow rate occurred over several hours, with cycle durations of 113–190 min, and discharges peaking at 0.7 m3 s−1 and waning to 0.1 m3 s−1. Each cycle was characterized by a relatively short, high-volume flux phase during which a pulse of lava, with a well-defined flow front, would propagate down-channel, followed by a period of waning flow during which volume flux lowered. Pulses involved lava moving at relatively high velocities (up to 0.29 m s−1) and were related to some change in the flow conditions occurring up-channel, possibly at the vent. They implied either a change in the dense rock effusion rate at the source vent and/or cyclic-variation in the vesicle content of the lava changing its bulk volume flux. Pulses would generally overspill the channel to emplace pāhoehoe overflows. During periods of waning flow, velocities fell to 0.05 m s–1. Blockages forming during such phases caused lava to back up. Occasionally backup resulted in overflows of slow moving ‘a‘ā that would advance a few tens of meters down the levee flank. Compound levees were thus a symptom of unsteady flow, where overflow levees were emplaced as relatively fast moving pāhoehoe sheets during pulses, and as slow-moving ‘a‘ā units during backup. Small, localized fluctuations in channel volume flux also occurred on timescales of minutes. Volumes of lava backed up behind blockages that formed at constrictions in the channel. Blockage collapse and/or enhanced flow under/around the blockage would then feed short-lived, wave-like, down-channel surges. Real fluctuations in channel volume flux, due to pulses and surges, can lead to significant errors in effusion rate calculations.