Session M30: Turbulence: Environmental Flows

Monami dynamics and instability dependence on Reynolds number and seagrass buoyancy

The interaction between flow and submerged seagrass beds can generate instabilities that evolve into vortices, induce grass blade oscillations, and affect the transport of sediment and nutrients in aquatic systems. We develop a numerical two-phase model to simulate the unstable flow through a seagrass bed. The seagrass is modeled as buoyant blades that do not resist shear and deform to adjust instantaneously to the fluid drag. The shape of the grass blades at all times can therefore be directly computed from the velocity field, and the flow within the seagrass bed responds to the drag, which depends on the shape of the blades. We study the instability onset with respect to a steady state as a function of grass buoyancy, which dictates deformability in the model, and Reynolds number. We then visualize how the vortices induce a collective waving motion in the seagrass bed, known as monami, and the vortex-driven material exchange resulting from this interaction.   

Instability-driven material exchange in flow through deformable, buoyant seagrass bed

The monami phenomenon results from the interaction of fluid flow and deformable, submerged seagrass beds. This interaction results in a shear instability that drives a collective waving amongst the grass blades, vortex generation in the free streaming fluid above the bed, and potentially material exchange between the bed and the fluid. Using a two-phase, coupled numerical model, we investigate how the instability and vortex strength are functions of the fluid Reynolds number and the grass buoyancy, which dictates grass deformability in the model. Then, using both a scalar field and Lagrangian particles, we evaluate the flux dependence on the instability strength and grass waving. As the vortices become sufficiently large, there is the potential that neighboring vortices will interact as they propagate down the channel. The conditions for vortex interactions and their impact on material transport out of the bed will also be presented. Finally, as the vortices travel down the channel, they entrain material from the bed. To better quantify the downstream transport, a coherent structure analysis is performed to evaluate the bounds of the materially coherent vortices.   

Energy and infrequent fluctuations of temperature related to atmospheric mechanisms for various climate change scenarios

Understanding, modeling, and predicting complex systems such as climate require coupling distinct phenomena, acting at different space/temporal scales:  wavelike features, and turbulent cascade, with different regimes, crucial for mixing and dissipation. In addition, turbulent statistics appear to be correlated with the long-time (large-scale) filtered field of the same quantity. For example, local and strong temperature fluctuations are most likely related to daily, seasonal, and sometimes interdecadal phenomena.  This contribution aims to provide physical arguments of this conditioning by investigating turbulent statistics at each scale and for a particular time/phase of the large-scale, long-time phenomena.  

The methodology uses transport equations for second and fourth-order moments of temperature,  filtered at different space/time scales.  Data originate from experimental measurements performed at the level of the ground in Hong Kong. The effect of daily and annual periodicity over one-and-two point statistics has been assessed by resorting to the theoretical framework based on the advection-diffusion for scalar fluctuations.  It is shown that extreme/rare temperature fluctuations are related to the enhancement of temperature cascade and the large-scale, meandering,  temperature gradient. Further extensions of this approach deal with improved modeling of extremes, such as heavy rainfall, dry spells, in the context of large-scale climate change and variability.   

Turbulence Measurements and Coherent Structures in a PIV Experiment on Grassland Fires

High frequency (30 Hz) two-dimensional particle image velocimetry (PIV) data, exploring fire spread in ignited hand-spread pine needles under calm ambient wind conditions, are analyzed in this study. As the flame spreads away from the ignition point in the absence of ambient wind forcing, it entrains cooler ambient air into the warmer fire core and experiences the dynamic pressure offered by the entrained air. Coherent structures describe the initial shape of the fire-front and its response to local wind shifts while also revealing possible fire-spread mechanisms. Vortex tubes originating outside the fire spiral inward and get stretched thinner at the fire-front leading to higher vorticity there. These tubes comprise circulation structures that induce a radially outward velocity close to the fuel bed, which pushes hot gases outward, thereby causing the fire to spread. Such structures also confirm the presence of counter-rotating vortex pairs, known to be a key fire-spread mechanism. Precessing of the axis of the vortex tubes causes them to be kinked. The strong updraft observed at the fire-front could potentially advect and tilt the kinked vortex tubes vertically upward leading to fire-whirl formation. As the fire evolves, its perimeter disintegrates in response to flow instabilities to form smaller fire "pockets". These pockets are confined to certain points in the flow field that remain relatively fixed for a while and resemble the behavior of a chaotic system in the vicinity of an attractor. Increased magnitudes of the turbulent fluxes of horizontal momentum, computed at certain such fixed points along the fire-front, are symptomatic of irregular fire bursts and help contextualize the fire spread. Most importantly, the time-varying transport terms of the turbulent kinetic energy (TKE) budget equation computed at adjacent fixed points imply that horizontal turbulent transport is the primary mechanism for interaction between local fires along the fire-front.   

Lagrangian and Eulerian Perspectives of Turbulent Transport Mechanisms in a Lateral Cavity

The dynamics of the turbulent flows past lateral cavities are relevant for multiple environmental and industrial applications. In rivers and coastal environments, lateral recirculating regions constitute surface storage zones in channels or the shore, where large-scale coherent structures dominate mass and momentum transport, playing a fundamental role in the dynamics of sediments, biogeochemical processes, and nutrient cycles. In this work we carry out LES of the flow in a straight rectangular channel with a lateral square cavity at Re=1.2×10⁴. The model is coupled with an advection-diffusion equation and a Lagrangian particle model to investigate the transport mechanisms in the cavity and across the interface, and to provide quantitative comparisons of these processes from both perspectives.  We also characterize statistically the variables that describe mixing and transport in the flow, analyzing particle trajectories and time scales that describe the mass and momentum exchanges, including residence time distributions and contaminant dispersion. Understanding the small-scale dynamics provides information to improve parameterizations in large-scale models, using statistical analyses that yield new insights of transport mechanisms in environmental flows.   

Behavior of oil droplets under the influence of counter-rotating vortices in a Langmuir cell tank

Langmuir circulation (LC), a form of wind shear-driven turbulence in the upper ocean layer, can enhance vertical mixing of positively or negatively buoyant particles like pollutants and organisms in the water column. During oil spills, LC can facilitate underwater entrapment and dispersion of oil droplets generated from crude oil slicks and facilitate oil/sediment interaction, especially in shallow waters. However, experimental knowledge on the influence of LC on the behavior of submerged oil droplets is limited. Using a 1×0.2×0.5 m³ Langmuir tank facility that can repeatably inject an oil jet in a counter-rotating vortex pair, we study the flow and resulting oil droplet behavior. The flow field in the Langmuir tank is characterized using 2D particle image velocimetry. The central downwelling region has maximum velocity of 14 to 26 mm/s and turbulence kinetic energy of 6.52E-5 to 1.16E-4 (m/s)². The spatiotemporal oil droplet size distribution is characterized using multiple high-speed cameras viewing different regions of the tank and a droplet detection algorithm. Tests using dyed vegetable oil showed that oil droplets with ~500 µm diameter can be retained underwater for more than 10 min after injection and that retained oil droplet size positively correlates with LC strength.   

Spectral Analysis of the Energy Budget of Canopy Flows

To understand the turbulent kinetic energy production, transport, and dissipation in canopy flows, we perform a spectral analysis on the energy budget of canopy flows with a range of filament flexibilities.  Our simulation resolves the filaments in the canopy and solves the interaction between the flow and flexible filaments with an immersed boundary method, such that the dynamic deformation of the filaments can be captured and no prescribed drag coefficient is necessary.  We distinguish two scales that have significances in the energy budget: the monami scale related to the progressive waving of the canopy (also known as "monami" or "honami") and the wake scale related to the wake behind each filament.  We examine the waving term associated with the correlation between the fluctuations of canopy drag and velocity.  In a flexible canopy, this term transports turbulent kinetic energy from lower canopy to upper canopy at the monami scale, and serves as a main source of dispersive kinetic energy at the wake scale.  Moreover, this term transports energy to the wake scale from the larger scales, a phenomenon termed the energy shortcut mechanism in the literature (e.g. Kaimal and Finnigan, 1994; Finnigan, 2000).  This phenomenon does not occur in the inter-scale transport term associated with the nonlinear interaction of different flow scales.  Therefore, we conclude that the energy shortcut is mainly due to the waving term, not the inter-scale transport term.   

3D Measurements of Transient Dispersion in an Urban Canopy Model: A Green's Function Approach

Predicting turbulent dispersion from street-level sources within the urban canopy is important for determining emergency response, assessing air quality, and providing input to neighborhood scale pollution models. Model improvements are challenged by the available validation data: either due to flow condition uncertainties for field measurements, or because conventional laboratory techniques are limited to point-wise and planar data. This work uses magnetic resonance imaging to obtain three-dimensional velocity and concentration fields for a transient release in a scale model of Oklahoma City circa 2003. A passive scalar is injected at ground level and data are obtained on a volumetric Cartesian grid at 12 temporal phases encompassing several blocks of the downtown business district. The source's Green's function is extracted at each voxel using a regularized optimization procedure. The Green's function generalizes the results to different release profiles, and statistics such as the plume residence time are determined objectively. Plume residence time is substantially longer than characteristic advection time scales in street canyons and corners between buildings that fill rapidly with contaminant but wash out slowly. Comparisons to the Joint Urban 2003 fields tests are given.   

Surface layer turbulence response to spanwise heterogeneous vegetative canopies

Using combined wind-tunnel measurements and large-eddy simulations (LES), the influence of spanwise heterogeneity on secondary circulations and on evaporative fluxes is investigated. The experiments were performed at the Technion environmental wind-tunnel. A 5m long canopy, modeled after available leaf-area index data for corn, consisted of triangular perforated sheet elements (h=20cm, in-line setup). Measurements were performed by hot-wire anemometry and stereo-PIV enabling to spatially and temporally resolve the turbulent flow. Measurements were performed at a bulk flow velocity of 1.5m/s (Re_h=19200). Several stereo-PIV data sets were acquired in wall-parallel planes positioned between 0.7h to 1.3h in steps of 0.05h. Evaporation rates at various locations within the canopy were measured using in-house made sensors. Measurements were performed for a homogeneous canopy and one that exhibited spanwise heterogeneity under the same bulk flow conditions. Turbulent flow characteristics, including turbulence production terms, will be presented and compared to the LES. The effect of heterogeneity on coherent structures near the canopy top was analyzed using quadrant and spectral analyses.   

Turbulent transport dynamics in open-channel floodplains for different submergence conditions

Floodplains are side regions of river channels characterized with low velocities that have a high depositional potential for contaminants transported by the flow, which can be resuspended during floods and mobilized downstream. The resuspension process is determined by the bed stresses, and transport is driven by the shear layer at the interface between the floodplain and the main channel. The temporal and spatial structure of the shear-stress distribution and the coherent-structure dynamics of the shear-layer strongly depends on the submergence ratio between the water depth of the main channel H, and the floodplain h. To provide a further understanding of the underlying physics of the exchange processes for different submergence conditions, we perform high-resolution numerical simulations using the detached-eddy simulation (DES) approach. We test four different submergence ratios, from shallow to deep-water regimes (h/H = 0.13, 0.25, 0.5, 0.62), and simulation results are validated with experimental data. We study the role of large-scale vortical structures and velocity fluctuations at the interface for different regimes by providing a comprehensive description of the effects of flow submergence on the spatial and temporal dynamics of the coherent structures of the shear-layer.