Workshop on Turbulence and Hydrodynamical Instabilities

17 Nov 2008, 08:00 → 19 Nov 2008, 18:00 Europe/Berlin

IPP auditorium (MPI für Plasmaphysik, Garching)

Andreas Burkert (USM)

The Excellence Cluster Universe organizes an interdisciplinary workshop on hydrodynamics, magnetohydrodynamics and aerodynamics.

We plan to discuss selected topics from astrophysics, meteorology, aerodynamics, medicine, computational fluid dynamics, mathematics, numerics, visualization as well as applications in the industries.

More information will be given soon.

Monday 17 November

09:00 → 12:00 IPP session

09:00 → 10:00 Jean-Francois Pinton, ENS Lyon/France

09:00 Turbulence and the dynamo effect

As proposed by Larmor almost en century ago, the magnetic field of planets and stars is generated in a conversion from kinetic to magnetic energy in the motions of an electrically conducting fluid. These motions are very intense and the associated flows are quite turbulent. However, turbulence can have ambivalent effects, either promoting or hindering the dynamo process. I shall review and discuss these features, in the light of recent numerical and laboratory experiments.

Speaker: Jean-Francois Pinton (ENS Lyon, France)

Slides Pinton.pdf

10:00 → 10:30 Coffee break

10:30 → 11:00 Alexei Kritsuk, University of California, San Diego

10:30 Supersonic Turbulence and Star Formation in Molecular Clouds

Turbulence within molecular clouds generates shock waves that tear the cloud material into a hierarchy of smaller and smaller clumps. It also provides the necessary kick to overcome the outward pressure and cause the densest cloud cores to collapse leading to the birth of stars. This "turbulent fragmentation" is believed to shape the initial mass function of newly born stars. However, scaling properties of highly compressible, magnetized isotropic turbulence that constitute the basis for this new statistical theory of star formation are still poorly understood. In my talk I shall review results from large-scale numerical simulations that investigate the properties of supersonic hydrodynamic and MHD turbulence. Our nonmagnetic simulations are large enough to isolate the inertial range in density and velocity statistics. We find strong departure from the incompressible Kolmogorov velocity scaling at high turbulent Mach numbers. We propose an extension of Kolmogorov's phenomenology to compressible regimes and discuss how magnetic fields modify these results. I will also discuss the effects of large-scale driving force used in the simulations on the derived turbulent statistics, intermittency and fractal dimension of dissipative structures in supersonic turbulence.

Speaker: Alexei KRITSUK (University of California, San Diego)

Slides Kritsuk.pdf

11:00 → 11:30 Rudolf Friedrich, Universität Münster

slides Friedrich.pdf

11:00 Statistics of Eulerian and Lagrangian Velocity Increments of Turbulent Flows

We adress the statistical properties of small scale velocity increments in fully developped turbulent flows. The starting point is the Monin-Lundgren hierarchy of evolution equations for probability distributions of n-point velocity increments. Closure assumptions, based on results from direct numerical simulations of turbulence, allows one to obtain closed kinetic equations relating the statistics of velocity increments at different scales. This clarifies the phenomenological description of the turbulent cascade in terms of a Fokker-Planck equation in scale [1]. The second part of the talk adresses the Lagrangian statistics of turbulence. We present a bridging relation, which connects the Lagrangian velocity increment statistics with the statistics of the Eulerian velocity increment [2].

Speaker: Rudolf FRIEDRICH (Univ. Münster)

Slides Friedrich.pdf

11:30 → 12:00 Wolf-Christian Müller, IPP Garching

11:30 Inherent properties of turbulence

Certain statistical characteristics of turbulence are particularly interesting due the information they yield about the still not fully understood nonlinear turbulent dynamics. In addition, their possible universality might allow a better understanding of a variety of physically different systems dynamically governed by turbulence. The talk reviews some efforts in this direction carried out with the help of large-scale direct numerical simulations of statistical homogeneous turbulence. Results are presented with regard to two-point statistics and spectral cascades in magnetohydrodynamic turbulence and rotating hydrodynamic turbulence. The fruitfulness of the Lagrangian point of view on turbulent dynamics by studying passive tracer particles in the flow is also touched upon.

Speaker: Wolf-Christian MÜLLER (IPP Garching)

Slides Mueller_Wolf.pdf

12:00 → 13:00 Lunch break

13:00 → 16:00 MPA session

13:00 → 14:00 Wolfram Schmidt, Universität Würzburg

13:00 Turbulence in Astrophysics

Starting with examples such as supernovae and molecular clouds, I will explain the features of astrophysical turbulence in contrast to terrestrial turbulent flow. Then I will touch on methods for the numerical simulation of turbulence in astrophysics. Apart from large eddy simulations, I will discuss attempts to treat turbulent flow with adaptive mesh refinement. In conclusion, I will outline a new approach that combines adaptive mesh refinement and subgrid scale models and show first applications.

Speaker: Wolfram SCHMIDT (Univ. Würzburg)

Slides Schmidt.pdf

14:00 → 14:30 Wolfgang Hayek, MPA Garching

14:00 Radiation Hydrodynamics in Cool Stellar Atmospheres

The envelopes of cool stars are convective, with the flowing plasma transferring energy from the hot stellar interior to the atmosphere. The gas rises, expands and eventually releases most of its heat into radiation in a very thin boundary layer at the surface. There it rapidly becomes transparent for the photons, allowing them to escape into space. The cooled gas then descends back into the star in turbulent downflows. Studying radiation hydrodynamics at stellar surfaces opens up a rich field of astrophysical research. Continuous developments of numerical methods as well as increasingly powerful computers allow us to better understand turbulent plasma flow and stellar granulation using high-resolution time- dependent 3D models. Detailed simulations which include magnetic fields in the plasma with their large variety of phenomena, such as Sun spots and flux tubes, have come within reach. Further applications exploit the radiation field itself for understanding spectral line formation in stellar atmospheres and measuring the chemical composition of the gas. This talk will briefly outline the physical environment found at the surface of stars, describe some important aspects of the numerical methods and give examples for applications of the models.

Speaker: Wolfgang Hayek (MPA Garching)

Slides hayek.pdf

14:30 → 15:00 Coffee break

15:00 → 15:30 Martin Obergaulinger, MPA Garching

15:00 Simulations of the magneto-rotational instability in core-collapse supernovae

The magneto-rotational instability (MRI) may amplify the weak magnetic field of a supernova progenitor in the differentially rotating post-collapse core, leading to MHD turbulence and enhanced transport of angular momentum. Confirmation of the potential importance for the explosion of rapidly rotating cores is hampered by numerical difficulties, in particular a very stringent requirement on the grid resolution which is difficult to meet in global simulations. We have performed a set of axisymmetric and three-dimensional local simulations of simplified models of the MRI in stellar cores. I highlight important differences of this case from the more standard case of accretion discs, and discuss the growth and saturation of the instability in our simulations.

Speaker: Martin OBERGAULINGER (MPA Garching)

Slides Obergaulinger.pdf

15:30 → 16:00 Thierry Foglizzo, CEA/Saclay, France

15:30 From the whistle of a kettle to the asymmetric explosion of supernovae

The advective-acoustic instability has been introduced in astrophysics in order to xplain the instability of the accretion flow onto a black-hole moving supersonically. Like the whistle of a kettle, its mechanism is based on the interplay of acoustic waves with the vorticity advected by the flow. Its effects seem to be most promising in the problem of core collapse supernovae, where it contributes to revive the stalled shock during the first second of the explosion, leading to an asymmetric explosion and a pulsar kick. I will explain our understanding of this instability, and comment on its treatment in numerical simulations.

Speaker: Thierry FOGLIZZO (CEA/Saclay France)

Slides foglizzo.pdf

16:00 → 16:30 Coffee break

16:30 → 17:30 Dmitrii Mironov, Deutscher Wetterdienst

16:30 Modelling Pressure Scrambling Terms in the Second-Moment Equations

One of the key issues in the second-order turbulence closure approach is the modelling of the pressure scrambling (redistribution) terms in the transport equations for the second-order moments. Various formulations for these terms are discussed, emphasising their physical realism, their advantages and shortcomings, and their utility for modelling turbulence in geophysical and engineering flows. A common approach nowadays to modelling the pressure scrambling terms in the Reynolds-stress and the scalar-flux equations is to decompose them into the contributions due to the non-linear interactions, referred to as the slow part of the pressure scrambling term, and due to the mean-gradient, buoyancy and the Coriolis effects, referred to as the rapid parts of the pressure scrambling term. These contributions are then modelled separately. The relaxation\return-to-isotropy" approximation (Rotta 1951) is applied to the slow part. For the rapid parts, the linear models are most often used. The respective rapid parts are simply set proportional to the mean-gradient, buoyancy and Coriolis terms in the equations for the Reynolds stress and for the scalar fluxes. The proportionality coefficients are adjusted (tuned) so that to provide a good fit of the model results to observational and numerical data. Although linear models of the pressure scrambling terms are in common use in geophysics and engineering, they entirely fail in many situations of interest. An illustrative example is turbulent convection driven by the surface buoyancy flux and affected by rotation. In the seemingly simple case where the rotation axis is aligned with the vector of gravity, linear models are unable to properly account for the effect of rotation, leading to erroneous prediction of the Reynolds stress and of the scalar fluxes. Then, a more sophisticated non- linear formulation is required. Of particular value are the so-called realisable models (Schumann 1977, Lumley 1978). Prominent among them is a two- component limit (TCL) model (Craft et al. 1996). Examples of its application to convective flows are presented. A TCL formulation for the pressure gradient- potential temperature covariance that properly accounts for the effect of rotation (Mironov 2001) is discussed in some detail. Results of its testing against large-eddy simulation data are presented. It is emphasised that the use of numerical data (from large-eddy and direct numerical simulations) is the only viable alternative in situations where observational data are difficult or impossible to take. This is the case for the fluctuating pressure in turbulent flows.

Speaker: Dmitrii MIRONOV (Dt. Wetterdienst)

Slides Mironov.pdf

17:30 → 18:30 Reception

Tuesday 18 November

09:00 → 10:00 Alexei Ivlev, MPE Garching

09:00 Complex plasmas: Hydrodynamics at the discreteness limit

Laboratory complex (dusty) plasmas are weakly ionized gases containing microparticles (dust grains) which are charged due to absorption of ambient electrons and ions. Although complex plasmas are intrinsically multi-species systems, the rate of momentum exchange through binary collisions between the microparticles can exceed that of other interactions (e.g., neutral gas drag) significantly. Therefore, fluid complex plasmas (viz., charged microparticles) can act as an essentially single-species system. This gives us unique opportunity to investigate fluid phenomena at the kinetic (individual particle) level and go beyond hydrodynamic limits, down to the smallest length scale available – the interparticle distance. How relevant are liquid plasmas for the study of conventional liquids? The implication is clear – if they are relevant, this opens up a completely new approach to nanofluidics, the kinetic approach, which will then have the major impact on the field. Comparison in terms of similarity parameters (e.g., Reynolds and Mach numbers) shows that liquid complex plasmas are remarkably like conventional liquids, e.g., water – observed at the molecular level. This suggests that liquid plasmas can indeed serve as a powerful new tool for investigating fluid flows on (effectively) nanoscales, including the all- important transition from collective fluid behavior to individual kinetic behavior, as well as nonlinear processes on scales that have not been accessible for studies so far. Various examples of fluid behavior observed recently in experiments with complex plasmas will be presented and discussed, such as kinetics of shear flows (similar to the Couette and Poiseuille flows), non-Newtonian viscosity (with well-pronounced shear-thinning and thickening effects), electrorheological fluids, etc.

Speaker: Alexei IVLEV (MPE Garching)

Slides Ivlev.pdf

09:00 → 12:00 MPE session

10:00 → 10:30 Coffee break

10:30 → 11:00 Hartmut Löwen, Universität Düsseldorf

10:30 Particle-resolved hydrodynamic instabilities in colloids and complex plasmas

Colloidal particles in a fluid solvent and dust particles in a plasma can exhibit hydrodynamic instabilities when they are driven by an external field. This instability can be watched in real-space on the particle scale on convenient time scales. Recent progress obtained by computer simulation, theory and experiments is summarized. In particular, the effect of laning in driven binaru mixtures will be discussed, as well as the Rayleigh-Taylor instability of driven colloids.

Speaker: Hartmut LÖWEN

Slides Loewen.pdf

11:00 → 11:30 Boris Klumov, MPE Garching

11:00 Local properties of 3D complex plasmas

In this talk we compare recent three dimensional complex (dusty) plasmas experiments (both ground-based and performed on-board International Space Station at microgravity conditions) with the molecular dynamics simulations. We discuss in detail crystallization properties of the complex plasma in narrow channels and local properties of 3D complex plasma.

Speaker: Boris KLUMOV (MPE Garching)

Slides Klumov.pdf

11:30 → 12:00 Markus Thoma, MPE Garching

11:30 Instabilities, Convection, and Turbulence in Complex Plasmas

Complex or dusty plasmas are low-temperature discharge plasmas containing micron size particles. These microparticles are highly charged due to electron collection on their surface. The microparticle component is a strongly-coupled, non-equilibrium system exhibiting complex phenomena such as the formation of a plasma crystal. I will report on experiments with complex plasma, in which instabilities, convection, and turbulence were observed. Various mechanisms for this behavior of complex plasmas will be discussed and an outlook to future experiments will be given.

Speaker: Markus Thoma

Slides Thoma.pdf

12:00 → 13:00 Lunch break

13:00 → 16:00 TUM session

13:00 → 14:00 Xiangyu Hu, TU München

13:00 Smoothed Particle Dynamics (SPD): macroscopic method and bridge to micro-scopic world

An overview of Smoothed Particle Hydrodynamics (SPD) method and its application in simulating macro-scopic and meso-scopic flows will be presented. The SPD is a fully Lagrangian, grid free method in which a smoothing kernel is introduced to approximate functions and their spatial derivatives from the interactions between neighboring particles. It is referred to as smoothed particle hydrodynamics (SPH) when simulating macro-scopic flows, and smoothed dissipative particle dynamics (SDPD) when simulating meso-scopic flows. The main advantage of SPD is the flexibility on modeling complex flows, such as multi-phase flow, polymer flow, with a board rang of scales.

Speaker: Xiangyu HU (TU München)

Slides hu.pdf

14:00 → 14:30 Michael Meyer, TU München

14:00 Efficient Modeling of Turbulent Flow Phenomena

Most flows occurring in nature and engineering applications are turbulent. For a profound understanding of these flow phenomena numerical simulation can be used. The high Reynolds number of many problems prohibits the exact resolution of the Navier Stokes equations in terms of a Direct Numerical Simulation (DNS). In Large Eddy Simulation (LES) only the large flow structures are resolved while small, stochastic structures are modeled with the help of Subgrid Scale Models (SGS). In Implicit Large Eddy Simulation (ILES) the truncation error of the discretization of the convective terms acts as a SGS model which is therefore implicit to the discretization. One implementation of ILES is the Adaptive Local Deconvolution Method (ALDM) which has shown considerable potential for the efficient representation of physically complex flows in generic configurations, such as isotropic turbulence and turbulent channel flow. To extend ALDM to the efficient treatment of complex configurations the usage of Cartesian grids in connection with Immersed Interface Methods (IIM) is expedient. Further improvement of efficiency is achieved by modeling the turbulent boundary layer using a wall model and by locally adapting the mesh resolution with Adaptive Mesh Refinement (AMR). The talk will give an overview on modeling turbulent flows with ALDM using the different efficiency increasing aspects.

Speaker: Michael Meyer (TU München)

14:30 → 15:00 Coffee break

15:00 → 15:30 Eric Lauer, TU München

15:00 Conservative multi-fluid flow simulation

The interaction of two or more different materials is of fundamental interest for many technical applications and numerous phenomena in nature, ranging from combustion to atrophysical supernovae. For the understanding of these precesses numerical simulation can be very helpful. Various numerical methods have been developed to study the behavior of multi-fluid flows. The major difference between these models is the way they deal with the material interface. In general there are two main approaches: one is the introduction of a transition region with a steep gradient; the other is the use of a sharp interface representation. The advantage of the latter is that the material- interface can be identified at any time. But tracking the interface can be rather complicated and connected with high numerical costs depending on the used tracking method. The levelset approach is a rather simple interface tracking method, but models based on levelset normally show a lack of conservative properties, leading to low accuracy near the interface. In the talk a sharp interface approach based on levelset will be presented, which updates the governing equations fully conservatively and uses a special mixing routine for small cut cells to avoid instabilities and mass losses at the interface.

Speaker: Eric LAUER (TU München)

Slides Lauer.pdf

15:30 → 16:00 Stefan Adami, TU München

15:30 Simulations of multiphase phenomena using Smoothed Particle Hydrodynamics

A two-dimensional Smoothed Particle Hydrodynamics (SPH) method to simulate macroscopic and mesoscopic flows is presented. Our focus is on modelling interfacial phenomena of multi-phase problems, especially of surface tension effects. The main advantage of this Lagrangian formulation is the adaptive interface representation. Therefore no complicated interface capturing (Front- tracking, Level-Set,...) is needed and the simulation of more complex cases including several different phases is straightforward. We use this method to simulate the dynamic surface tension of air-water interfaces (like present in the lung alveoli) containing surface active agents ("surfactant") and present a local conservative formulation of the interfacial transportation of the surfactant.

Speaker: Stefan ADAMI (TU München)

Slides Adami.pdf

16:00 → 16:30 Coffee break

16:30 → 17:30 Thorsten Naab, USM

16:30 Hydrodynamical simulations of astrophysical systems

I will present an overview of smoothed particle hydrodynamics (SPH) simulations applied to a range of astrophysical systems from protostellar disks over molecular clouds to galaxies in a full cosmological context in addition to recent progress on algorithms and the use of special hardware.

Speaker: Thorsten NAAB (USM)

18:30 → 19:30 Guided tour at BMW

Wednesday 19 November

09:00 → 12:00 BMW session

09:00 → 10:00 Norbert Grün, BMW Group

09:00 Industrial Application of a Lattice-Boltzmann Method in Automobile and Motorcycle Aerodynamics.

Due to the level of maturity that simulation methods have reached today, CFD tools are more and more employed in a productive manner in the aerodynamic development process. The objective is twofold. First one tries to reduce the number of cost and time intensive experiments in the early phase and on the other hand CFD enables a deeper understanding of the flow phenomena around and through detailed vehicle models in the later phase. This presentation demonstrates the status of CFD usage at the BMW Group during the aerodynamic development of passenger cars and motorcycles, in particular employing a commercial Lattice-Boltzmann tool (PowerFLOW). Details of this code will be presented in a following talk. First an overview is given of the development process and the questions that can be tackled by simulation in the different phases. Then the simulation process is explained from geometry preparation to results analysis. Some selected validation cases demonstrate the accuracy that can be achieved at the moment. Finally a number of practical examples is used to show the capabilities for in detail analysis of the flow field.

Speaker: Norbert Grün (BMW Group München)

Slides Gruen.pdf

10:00 → 10:30 Coffee break

10:30 → 11:00 Hudong Chen, Exa Burlington, USA

10:30 Kinetic theory representation for turbulence modeling and computation

One of the most common approximations in turbulence for the averaged effect of small scales is by the so called eddy viscosity modeling. That is, one approximates the Reynolds stress as a linear function of the local rate of strain of the averaged flow field. The scalar proportionality factor is referred to as an eddy viscosity. This concept was first proposed over a century ago. It stems from an analogy for small eddy interactions with collisions of molecules resulting in Newtonian fluid constitutive relations. This approximation has made enormous impact particularly in computational fluid dynamics for turbulent flows. Many theoretical works were also developed since then, with various successes, in order to analytically derive such a functional relationship. However, unlike molecular interactions in a fluid, one of the apparent criticisms or difficulties in this analogy is the lack of scale separations between averaged fluid motions and fluctuating eddies. In this presentation, the speaker will give a somewhat provocative argument in favor of such analogy, provided that this concept be expanded in a generalized kinetic theory framework. In such an expanded framework, the analogy between eddy interactions and molecular collisions has a broader physical validity, while the eddy viscosity approximation is its consequence in the very long wave length limit. The kinetic theory representation itself needs not depend on scale separations. Using such an expanded analogy, one can also draw similarities between turbulent flow phenomena to that of non- Newtonian fluid flows in micro/nano scales. Nevertheless, as far as the speaker is aware, there have been limited theoretical progress in formulating such a kinetic theory description for turbulent flows via first principle. The speaker will also discuss its implications in performing large eddy simulations.

Speaker: Hudong CHEN (Exa Burlington USA)

Slides Chen.pdf

11:00 → 11:30 Martin Schulz, science+computing, Tübingen

11:00 The Interactive Experience of Locally Refined Cartesian Grids - Visualization with PowerVIZ

Locally refined cartesian grids as used in Lattice-Boltzmann simulations exhibit a large potential for efficient storage and interactive visualization, interestingly these grids are not in the focus of commonly known visualization systems. The visualization tool PowerVIZ takes advantage of these opportunities in adapted algorithms like cutting plane, streamline and isosurface calculation to achieve interactive response times even for large data sets. The lean storage structure accounts for the small memory footprint of the application. Even more advanced visualization techniques like direct volume rendering are integrated. Raycasting algorithms are implemented as GPU-shader programs. Vortices can be displayed as core lines or using Lambda2-criterion. One step beyond plain visualization is the investigation in dirt simulation. Mass-carrying particles are simulated in the fluid and their hit points are visualized on surfaces. The daily work of engineers is assisted by the recording interface and key frame animation.

Speaker: Martin SCHULZ (science+computing Tübingen)

11:30 → 12:00 Guillermo MARCUS, Univ. Mannheim

11:30 Hardware Acceleration of SPH simulations with special Hardware

The recent use of programmable graphic cards (GPUs) as scientific coprocessors brings very powerful parallel processors to the masses. This talk will be centered in SPH simulations with GPUs, their capabilities and challenges, with a focus in astrophysical simulations.

Speaker: Guillermo MARCUS (Univ. Mannheim)

Slides Marcus.pdf

12:00 → 13:00 Lunch break

13:00 → 14:00 Fabian Heitsch, Univ. Michigan

13:00 The Making of Molecular Clouds: Turbulence, Cooling and Gravity

Molecular clouds -- the birth places of stars -- are observed to be highly structured and very dynamic. Doppler-broadened molecular lines indicate turbulent Mach numbers of 2-10, and column density maps of molecular tracers suggest volume filling factors of dense cores to range around 10%. It is in those densest regions that stars are born. Almost all molecular clouds are observed to host star formation. Thus, stars must begin to form immediately after (or even during) the formation of the parental cloud. Hence the properties that allow molecular clouds to form stars must arise during their own formation process. I will explore a scenario of molecular cloud formation in large-scale flows of diffuse interstellar gas. I will discuss the dominant magneto-hydrodynamical and thermal instabilities responsible for the necessary fragmentation of the assembling flows, eventually allowing the observed rapid onset of star formation, and I will present numerical simulations of flow-driven cloud formation, including a short summary of the gas-kinetic scheme used for the modeling.

Speaker: Fabian Heitsch (Univ. of Michigan, USA)

Slides Heitsch.pdf

13:00 → 16:00 USM session

14:00 → 14:30 Matthias Gritschneder, Univ. Sternwarte München

14:00 Ionization and Triggered Star Formation in Turbulent Molecular Clouds

We present high resolution simulations of the impact the ionizing radiation of massive O-stars has on the surrounding turbulent interstellar medium. The simulations are performed with the newly developed software iVINE which combines ionization with smoothed particle hydrodynamics(SPH) and gravitational forces. We show that radiation from hot stars, efficiently heating cold low density gas, penetrates the ISM and amplifies overdensities seeded by the initial turbulence. The formation of observed pillar like structures (e,g, in M16) in star forming regions can be explained by this scenario. Eventually, at the tip of the pillars induced gravitational collapse leads to formation of stars. In-depth analysis of the evolution of the turbulent spectra shows that UV-radiation indeed provides an excellent possibility to sustain and even drive turbulence on intermediate scales in the ISM.

Speaker: Matthias GRITSCHNEDER (Univ.Sternwarte München)

Slides Gritschneder.pdf

14:30 → 15:00 Coffee break

15:00 → 15:30 Veronika Junk, Univ. Sternwarte München

15:00 Kelvin-Helmholtz Instability and its Significance in Astrophysical Algorithms

Smoothed Particle Hydrodynamics (SPH) simulations are a powerful tool to investigate hydrodynamical processes in astrophysics such as the formation of galactic disks. Dense gas clouds raining on the forming disk move through the halo and are possibly disrupted in the process by Kelvin-Helmholtz- Instabilities (KHI). To understand the evolution of the halo clouds, we have to ascertain the capability of SPH to treat the KHI correctly, since SPH-methods tend to suffer from an innate surface tension and viscosity effects, both of which could dampen the KHI. We analytically derive a growth rate of the KHI including surface tension and viscosity in the linear regime, and compare this growth rate to results of numerical simulations by an SPH method and a grid- based method. We find that SPH in some cases suppresses the KHI.

Speaker: Veronika JUNK (Univ.Sternwarte München)

Slides Junk.pdf

15:30 → 16:00 Hans Böhringer, MPE Garching

15:30 X-ray Observations of Turbulence in Galaxy Clusters

Turbulent motion of the intracluster plasma is expected due to cluster mergers on global scale and due to energy output from AGN in the central regions. Observations of pressure variations of the intracluster medium of the Coma cluster (a post-meger system) suggest turbulent motion with a power law power spectrum. Transport of metals in the centers of galaxy clusters can be explained by turbulent transport. Furture X-ray observatories will allow us to better assess turbulence in galaxy clusters by means of observations of line broadening of spectral emission lines.

Speaker: Hans BÖHRINGER (MPE Garching)

16:00 → 16:30 Coffee break

16:30 → 17:30 Michael Böttinger, Dt. Klimarechenzentrum

16:30 Interactive Data Visualization with Focus on Earth System Research

Data visualisation is one of the key technologies needed for the visual analysis of extensive numerical simulation data. Furthermore, visualizations are often mandatory for an effective communication of the simulation results, especially with respect to general audiences. With the the concept of remote 3D visualization in a high performance computing environment, recently introduced at DKRZ (German Climate Computing Centre), the challenges arising from continous growth of the simulation data can potentially be faced. By using examples from the practical work with results from climate models, the talk will give an brief overview about different commercial and public- domain-solutions for data visualization. At DKRZ, "Avizo" (by Mercury Comptersystems) is now mainly used for the interactive 3D-visualization of time dependent and multivariate simulation data. More examples from climate modeling will be shown to demonstrate the different 3D visualization techniques available with Avizo.

Speaker: Michael BÖTTINGER (Dt. Klimarechenzentrum)

Slides Boettinger.pdf

17:30 → 18:00 Visualization Demonstration

The Computing Centre RZG of the Max-Planck-Society and IPP provides support for visualization of data from High-Performance-Computing simulations (see http://www.rzg.mpg.de/visualisation). We outline the challenges arising due to the broad range of disciplines in the Max-Planck-Society, the corresponding variety of simulation codes and data formats, as well as due to the sheer amount of data. In a live demonstration we show results from of a number of particularly challenging visualization projects and give brief notes on the scientific background and on the specific visualization techniques employed. Experiences with interactive visualization of massive datasets on commodity hardware are discussed.

Speakers: Markus Rampp, Ralph Bruckschen

Slides bruckschen_rampp_vizdemo.pdf