Invited speakers

Abstract:

An outstanding feature of atmospheric flows is the range of spatial and temporal scales involved. Development of rain through gravitational collisions of small cloud droplets concerns processes at sub-centimeter scales. Size distribution of cloud droplets within turbulent cumulus and stratocumulus clouds in the tropics and subtropics, where the solar insolation is at its peak, has a critical impact on the amount of radiation reflected back to space, and thus on the planetary albedo. Tropical deep convective clouds, often organized into mesoscale convective systems with horizontal scales of tens to hundreds of kilometers, drive planetary-scale Hadley circulation, which plays an essential role in the Earth energy and water budgets. For all these scales, numerical modeling---either for scientific research or for practical purposes, like the numerical weather prediction---plays an important role. The multiscale nature of these flows, often involving variable physics (e.g., hydrostatic large-scale flow and nonhydrostatic convective dynamics for the climate problem) needs to be carefully addressed. In this lecture, I will present examples of multiscale modeling approaches to selected problems in atmospheric fluid dynamics. These will include collision/coalescence of cloud droplets, turbulent entrainment into convective clouds, and representation of cloud processes in climate models.

Abstract:

The three-vortex problem has a long and colorful history. As an integrable three-body problem it is of intrinsic interest. Its most recent generalization to three vortices in a periodic domain has opened up entirely new vistas and application areas. The lecture will survey the three-vortex problem, starting with Gröbli's thesis of 1877 and ending with recent work and applications.

Abstract:

Computional Fluid Dynamics algorithms have reached a high degree of sophistication and have becoming mainstream tools in particular area's like aeronautical analysis and design. In this presentation we highlight with a personal view some of the underlying algorithms that made these achievements possible. However, in many area's especially when multidisciplinary physics is involved, Computational Fluid Dynamics is far from mature, requiring strong interaction between specialists in physics and numerical teams. This imposes a new model of CFD development where the attention goes as much to the interdisciplinary coupling algorithms as to the way different methods and discretization tehniques can interact in a collaborative and extendible way in a component based software environment. The main principles of such a collaborative environment will be presented together with some case studies showing its use in multi-partner projects in aerospace involving both human and multidisciplinary interaction.

Abstract:

Abstract: 

The European Transonic Windtunnel (ETW) in Cologne, Germany, is the most advanced aerodynamic test facility in the world. By performing low temperature operation, this modern wind tunnel is capable of simulating actual flight conditions of modern transport aircraft, defined by the Mach number and the Reynolds number. What this tunnel stands out for is its ability to match the respective high Reynolds number which cannot be done in conventional wind tunnels at ambient temperature.

Operating in a pressurised cryogenic nitrogen atmosphere over a speed range of 0.15 < M < 1.3 puts enormous challenges on any instrumentation to achieve reliable results. While the aerodynamic performance of half- and full-models is assessed by sophisticated balances the steady and unsteady aeroelastic model behaviour can be monitored online by an ultra–modern photogrammetric system. Regarding flow visualisation, classical techniques may be combined with the application of liquid crystals. For transition measurements of the boundary layer a new bi-luminophore Temperature Sensitive Paint is available. Promising progress has also been achieved on using Pressure Sensitive Paint at low temperatures.

On the way to get non intrusive techniques available for the analysis of the flow field around a model Doppler Global Velocimetry could successfully be applied. Presently, activities are ongoing in the small cryogenic pilot wind tunnel PETW to modify the Particle Image Velocimetry for operation at a tunnel environment required for a simulation of high Reynolds number conditions.

Abstract:

Many turbulent flows in the environment and in technology may be significantly altered when a large ensemble of embedded particles is included. This pertains to situations as diverse as the formation of rain, the dispersion of hazardous pollutants in the atmosphere, transport of sediment in rivers and estuaries and in various applications in chemical process-engineering. A fundamental understanding of this `modulation' of turbulence is required in order to allow a proper management and prediction of processes at various scales. Numerical simulation can provide important access to the dynamical processes behind the occurrence of such `non-ideal' turbulence. To illustrate this, the gas-solid flow in homogeneous, isotropic turbulence, and in a vertical channel are investigated at significant volume fractions and high mass-loading. Direct- and large-eddy simulation are adopted. Different sub-filter models for representing the dynamical consequences of small-scale turbulent flow features are compared to direct numerical simulation. Attention wil be given to three topics: a. First, the accuracy of large-eddy simulation is discussed using an error-landscape analysis. We compare dynamic eddy-viscosity models with recently developed regularization models, such as the Leray model and the Lagrangian averaged Navier-Stokes-alpha model. b. Subsequently, the consequences of adding an ensemble of discrete particles is discussed. A point-particle model is coupled to a continuous turbulent flow formulation. We consider the effect of particle-fluid and particle-particle interactions on the degree of preferential clustering in homogeneous, isotropic turbulence. Moreover, in case of  channel flow we show that particle-fluid interactions yield a thinner boundary layer, a flatter velocity profile and increased streamwise velocity fluctuations. In addition, the inclusion of particle-particle interactions allows to quantify the competition between turbulent mixing and inelastic particle collisions. As a result, large-scale particle clusters form and disintegrate within the flow, which may have a considerable influence on chemical processing and dispersion characteristics.   c. Finally, we assume the particles to be small water droplets and include heat- and mass-transfer between the gas-flow and these droplets. This allows the size of each droplet to alter along its flight-path, consistent with interactions with local turbulence. Evaporation and condensation against a background humidity are included. This yields a distribution of particle sizes and a corresponding mixture of dynamical responses to local turbulence conditions. The consequences for the spatial distribution and partial de-mixing of the droplet ensemble will be discussed.