10:30
Geophysical and astrophysical turbulence 3
Chair: Alex Liberzon
10:30
15 mins
|
Dual Cascades in Two-dimensional Compressible Turbulence
Alexei Kritsuk, Gregory Falkovich
Abstract: We report on results from a set of numerical simulations of isothermal compressible turbulence forced at an intermediate scale with grid resolutions up to 8192^2. At a moderate pumping rate, the total energy of a typical frictionless system, subject to periodic boundary conditions, exhibits linear growth characteristic of an inverse cascade followed---as in the incompressible case---by condensation. The energy capacity of the compressible condensate, however, is limited by shock dissipation. Effects of compressibility are discussed in both shaping the inverse energy cascade at moderate Mach numbers and mediating the energy condensate growth that involves the formation of strong shocks, connecting centers of large vortices and short-circuiting the energy cycle. In the saturated regime, the total energy exhibits strong irregular oscillations caused by metastability of the compressible condensate.
|
10:45
15 mins
|
Mean flow generation by Görtler Vortices in a rotating annulus with librating side walls
Abouzar Ghasemi V, Marten Klein, Uwe Harlander, Andreas Will
Abstract: Longitudinal libration of the cylinder side walls of a rotating annulus in the supercritical regime induces a centrifugally unstable Stokes boundary layer which generates Görtler vortices only in a portion of a libration cycle. We show for the first time that these vortices propagate into the fluid bulk and generate an azimuthal mean flow which is retrograde (prograde) over the outer (inner) cylinder side wall. Direct numerical simulations (DNS) are carried out and Reynolds-averaged equations and kinetic energy budget of mean and fluctuating flow are used as diagnostic equations to discuss the generation mechanism and scaling behavior of the azimuthal mean flow in the fluid bulk.
|
11:00
15 mins
|
Entrainment of a turbulent patch in a stratified fluid
Lilly Verso, Maarten van Reeuwijk, Roi Gurka, Peter Diamessis, Alex Liberzon
Abstract: Turbulent patches are localized events of turbulence, typically characterized by sharp differences between the flow characteristics across their interfaces. These localized events might add to the global mixing, heat exchange and mass transfer, playing a non-negligible role in the total energy balance in lakes or the ocean. This study takes a detailed look at the inner structure of a localized, mechanically forced patch in a linearly stratified ambient using laboratory experiments utilizing synchronized PIV and PLIF.
The results point out that the role of the turbulent/non-turbulent interface at the edge of the patch could be significant in determining the growth rate and the maximum size of the patch.
|
11:15
15 mins
|
ON THE TURBULENCE INDUCED BY NON-BREAKING SURFACE WAVES
Francisco OCAMPO-TORRES, Pedro Osuna, Aldo Hernandez Olivares
Abstract: Upper ocean turbulence is a very relevant process within the context of quantifying the exchange between the atmosphere and the ocean. We present results of a study of turbulence induced by non-breaking waves from detailed observation in a small wave tank facility in the Faculty of Marine Science of the University of Baja California. The experiments were carried out with mechanically generated monochromatic waves in a tank of approximately 12m x 0.5m x 0.3m with a varying steepness (ak) from about 0.02 to 0.18, and detailed measurements of the vertical profile of the 3d velocity field within a water column of about 0.035 m in length. Routine data quality control includes the use of only high beam correlation signal within each of the acoustic beams used by the profiler. A rotation matrix was applied to the velocity data matrix in order to secure that the x, y, and z axes were properly aligned with the wave tank. Rather low wave reflection was obtained through the implementation of a beach at the end of the wave tank. The intermittent character of the turbulence present is shown as a region following the -5/3 power law in the spectrum. Nevertheless, this spectral shape is being observed in most of our experimental results, particularly in those where the wave steepness was not too small. Root mean square values are obtained from the turbulent fluctuations time series to evaluate an integral quantity to characterize the turbulence intensity. This intensity is analyzed in terms of the wave steepness showing a linear relationship. The turbulent kinetic energy dissipation rate is also estimated and the relevance of the velocity spectra directly estimated over a limited number of wavenumber bands is addressed.
This work represents a RugDiSMar Project (CONACYT 155793) contribution and support from CONACYT project CB-2011-01- 168173 is greatly appreciated.
|
11:30
15 mins
|
PREFERRENTIAL CONCENTRATION OF PARTICLES IN PROTOPLANETARY NEBULA TURBULENCE
Thomas Hartlep, Jeffrey Cuzzi
Abstract: Preferential concentration in turbulence is a process that causes inertial particles to cluster in regions of high strain (in-between high vorticity regions), with specifics depending on their stopping time or Stokes number. This process is thought to be of importance in various problems including cloud droplet formation and aerosol transport in the atmosphere, sprays, and also in the formation of asteroids and comets in protoplanetary nebulae.
In protoplanetary nebulae, the initial accretion of primitive bodies from freely-floating particles remains a problematic subject. Traditional growth-by-sticking models encounter a formidable ``meter-size barrier'' [1] in turbulent nebulae. One scenario that can lead directly from independent nebula particulates to large objects, avoiding the problematic m-km size range, involves formation of dense clumps of aerodynamically selected, typically mm-size particles in protoplanetary turbulence. There is evidence that at least the ordinary chondrite parent bodies were initially composed entirely of a homogeneous mix of such particles generally known as ``chondrules'' [2]. Thus, while it is arcane, turbulent preferential concentration acting directly on chondrule size particles are worthy of deeper study.
Here, we present the statistical determination of particle multiplier distributions from numerical simulations of particle-laden isotopic turbulence, and a cascade model for modeling turbulent concentration at lengthscales and Reynolds numbers not accessible by numerical simulations. We find that the multiplier distributions are scale dependent at the very largest scales but have scale-invariant properties under a particular variable normalization at smaller scales.
|
11:45
15 mins
|
Prandtl number effects on the decaying and the forced turbulence in stratified fluids
Okino Shinya, Hanazaki Hideshi
Abstract: Effects of high-Prandtl number density-stratifying scalar, i.e., active scalar, on decaying and forced turbulence in stratified fluids are investigated by numerical simulations. In decaying turbulence, potential energy spectrum of the high-Prandtl number active scalar (Pr=6) agrees with the kinetic energy spectrum even at small scales. In forced steady turbulence, these two spectra again approach each other at small scales. These phenomena, which are in disagreement with the Batchelor scaling for a high-Schmidt number passive scalar, occur at scales even smaller than the Ozmidov scale, suggesting that these effects would not be negligible in general.
|
12:00
15 mins
|
Turbulent flows driven by the libration of an ellipsoidal container
Benjamin Favier, Alexander Grannan, Michael Le Bars, Jonathan Aurnou
Abstract: We present a combination of laboratory experiments and numerical simulations modelling a geophysically relevant mechanical forcing: libration. Longitudinal libration corresponds to the periodic oscillation of a body's rotation and is, along with precessional and tidal forcings, a possible source of turbulence in the fluid interior of satellites and planets. In this study, we investigate the fluid motions inside a librating tri-axial ellipsoidal container filled with an incompressible fluid. The turbulent flow is driven by the elliptic instability which is a triadic resonance between two inertial modes and the base flow with elliptical streamlines. This is called the libration driven elliptical instability (LDEI). We characterize the transition to turbulence as triadic resonances develop while also investigating the properties of the turbulent flow that displays both intermittent or sustained regimes. The existence of such intense flows may play an important in understanding the thermal and magnetic evolution of bodies subject to mechanical forcing, which is not considered in standard models of convectively-driven magnetic field generation.
|
12:15
15 mins
|
SOLAR WIND SPECTRAL ANALYSIS IN HELIOSHEATH FROM VOYAGER 2 DATA
Federico Fraternale, Luca Gallana, Sophie Fosson, Enrico Magli, Merav Opher, John Richardson, Michele Iovieno, Daniela Tordella
Abstract: The solar wind is a supersonic flow of magnetized plasma. It is time-dependent on all scales and expands with distance. The flow has fluctuations on a broad range of scales and frequencies. This fluctuations are not just convected outward but show energy cascades among the different scales. The solar wind turbulence peculiar phenomenology has been comprehensively reviewed by Tu and Marsch (1995) and Bruno and Carbone (2013). As the distance from the sun increases, the available data on plasma and magnetic field become increasingly scarce. At distance of the order of 1 astronomic unit (AU), several measurement have been performed by various crafts, but, nowadays, only the Voyager spacecrafts can measure data in the heliosheath, the outermost layer in heliosphere where the solar wind is slowed by the pressure of the interstellar gas, and only the Voyager 2 craft can measure both plasma and magnetic fields (Voyager 1 can measure only the magnetic field, and Pioneer 10 and 11 has ceased communications). Taken together, the Voyager 1 and 2 probes have collected over 11 year of data in the heliosheath. The Voyager plasma experiment observes plasma currents in the
energy/charge range 10 – 5950 eV /q using four modulated-grid Faraday cup detectors [Bridge 1977]. The observed currents are fit to convected isotropic proton Maxwellian distributions to derive the parameters (velocity, density, and temperature) used in this work. Magnetic field and plasma data are taken the COHO web site and MIT Space Plasma Group repository. Several studies have been done in order to extend the existing models to make them consistent with the energetic particles and magnetic fields measured in the heliosheath, but so far an exhaustive explanation has not yet been obtained. In particular, the differences between the energetic particle intensity variations seen by the two crafts are unexplained. The electron intensity measured by Voyager 2 varies steeper by a factor of 10 in a single year, while the same quantity from Voyager 1 changes gradually over time [Hill et al 2014]. A possible explanation can be the presence of bubble of turbulence that travels in the heliosheat. Therefore, a characterization of turbulence and its intermittency is necessary to explain this phenomenology. The aim of this work is to provide the first spectral analysis of heliosheath solar wind, trying to characterize the plasma turbulence in that region by estimating the spectral slopes. A first result is that the low frequency spectral slope is lower when the electron intensity is low. In order to compute spectra, signal reconstruction techniques are mandatory: at distance over 80 AU, available data are very spotty. For the plasma velocity, there are 97% of missings due to unsteadiness in the signals the most important of which are: tracking gaps due to the V2 location and due to limited deep space network availability; interference from other instruments; possible errors in the measurement chain (from the Faraday cups up to the data acquisition system and the signal shipping to Earth); the temporal sequence of the nuclear propulsion used to control the Voyager trajectory and to assist in several critical repositionings of the craft. For data recovery, we mainly use two different methods. The first method used is based on the correlation computation [Matthaeus & Goldstein 1982] that allows to reconstruct correlations and use it to compute spectra. Better results can be achieved implementing the maximum likelihood reconstruction by Rybicki and Press (1992) based on a minimum-variance recovery with a stochastic component. The second methods comes from the Compress Sensing, a recent technique widely used in telecommunications, that provides the reconstruction of the signal from a sparse dataset (Donoho 1997), by using sparse Fourier matrices (Rauhut 2010). The methods used have been previously tested on 1979 data and on synthetic fluid turbulent fields. Results were in good agreement with the literature, and allows to compute largest spectra of solar wind at 5 AU, with frequencies ranging from 10−7 to 10−2 Hz [Fraternale et al (submitted), Gallana et al 2014].
|
|