Bulletin of the American Physical Society
62nd Annual Meeting of the APS Division of Fluid Dynamics
Volume 54, Number 19
Sunday–Tuesday, November 22–24, 2009; Minneapolis, Minnesota
Session BS: Geophysical: Atmospheric II - Simulation |
Hide Abstracts |
Chair: Fernando Porte-Agel, University of Minnesota Room: 200G |
Sunday, November 22, 2009 10:30AM - 10:43AM |
BS.00001: Direct Simulation of the Turbulent Ekman Layer: Evaluation of Closure Models Scott Waggy, Stuart Marlatt, Sedat Biringen A direct numerical simulation (DNS) at a Reynolds number of 400 was performed for the atmospheric boundary layer (ABL) using an Ekman layer approximation. Turbulence energy budgets show that these simulations are pertinent to the ABL. First-order closure models also compared well with the DNS results as the general distribution of the eddy diffusivity was captured correctly for both the neutral Ekman layer as well as the stably-capped Ekman-layer. This led to the conclusion that the vertical profile of the eddy diffusivity can be accurately represented by a cubic polynomial as is done in both first-order closure models. [Preview Abstract] |
Sunday, November 22, 2009 10:43AM - 10:56AM |
BS.00002: DNS of stably stratified open channel flow Oscar Flores, James Riley The surface layer of the atmospheric boundary layer under stable conditions (i.e., night time) is approximated here by a DNS of a stably stratified open channel, with buoyancy effects included using the Boussinesq approximation. The Reynolds number is $Re_\tau=u_*h/\nu=560$, where $u_*$ is the friction velocity and $h$ is the height of the channel. Night time cooling is simulated by a negative heat flux at the ground, which imposes a stable density profile in the flow. The simulations are initialized with velocity fields obtained from a non-stratified case, and they are run until a quasi- steady state is achieved. As a result of the ground-cooling, the Reynolds stresses and the turbulent fluctuations decay in times of the order of $L/u_*$, where $L$ is the Obukov length-scale. For relatively weak cooling, the turbulence survives, and the flow evolves towards a quasi-steady state. If the cooling is too strong, the flow becomes laminar. Our results indicate that parameter controlling that behavior is $Lu_*/\nu$, rather than $L/h$ as proposed before. Interestingly, for the turbulent cases, the flow in the near-wall region is very similar to the non-stratified case, except for the longest scales of the streamwise velocity. In the outer region the profiles of velocity fluctuations are also similar, but the turbulence structure and the energy balance are drastically changed by the stable stratification. Funded by ARO Grant No. W911NF-08-1-0155. [Preview Abstract] |
Sunday, November 22, 2009 10:56AM - 11:09AM |
BS.00003: ABSTRACT WITHDRAWN |
Sunday, November 22, 2009 11:09AM - 11:22AM |
BS.00004: Simulated statistics of polydisperse sedimenting inertial particles in a turbulent flow under experimental conditions Lian-Ping Wang, Hossein Parishani, Bogdan Rosa, Colin Bateson, Alberto Aliseda, Wojciech Grabowski In recent years, point-particle based or hybrid direct numerical simulations (DNS) have increasingly been used to study pair statistics of inertial particles relevant to turbulent collision of cloud droplets. Equivalent experiment data are rare but are slowly becoming available. In this talk, we will discuss simulated statistics of sedimenting inertial particles under conditions similar to our parallel wind-tunnel experiment (to be reported here by Bateson et al.). The key parameters to be matched are flow Reynolds number, dissipation rate, particle Stokes number, and dimensionless settling velocity. A prescribed droplet size distribution will be used in the simulation to reproduce the polydisperse condition in the experiment. High-resolution DNS will be used to maximize the computational domain size. Single-particle and particle-pair statistics (e.g., fluctuation velocities, radial distribution function, relative velocity statistics) will be compared to the experimental data. Statistics obtained from lower dimensions will be linked to statistics in three dimensions. [Preview Abstract] |
Sunday, November 22, 2009 11:22AM - 11:35AM |
BS.00005: Reduced Order Estimation of the Atmospheric Boundary Layer Using POD-LSE Jonathan Naughton, Manjinder Singh, Edward Patton, Peter Sullivan The complimentary POD-LSE (proper orthogonal decomposition -- linear stochastic estimation) technique has been used to develop a reduced order model from large eddy simulations (LES) of the atmospheric boundary layer. The technique allows modeling of the coherent turbulence that is inherent to real atmospheric flows, where older spectrum based methods fail. The usefulness of POD is the low number of modes required to capture the energy containing structure in the flow. The power of LSE lies in its ability to estimate the time-dependent flow field using sparsely distributed data points. For the daytime unstable atmospheric boundary layer studied here, approximately 90{\%} of the total energy was captured by accurate estimation of less than 2{\%} of the POD modes. Although the approach was developed using a Large Eddy Simulation as input, the approach provides the foundation for designing a field experiment. The POD modes would be determined from spatially-resolved measurements that lack time resolution, and the flow field dynamics would then be estimated using a few optimally placed sensors providing time-dependent information. [Preview Abstract] |
Sunday, November 22, 2009 11:35AM - 11:48AM |
BS.00006: LES of wind turbine wakes: Evaluation of turbine parameterizations Fernando Porte-Agel, Yu-Ting Wu, Leonardo Chamorro Large-eddy simulation (LES), coupled with a wind-turbine model, is used to investigate the characteristics of wind turbine wakes in turbulent boundary layers under different thermal stratification conditions. The subgrid-scale (SGS) stress and SGS heat flux are parameterized using scale-dependent Lagrangian dynamic models (Stoll and Porte-Agel, 2006). The turbine-induced lift and drag forces are parameterized using two models: an actuator disk model (ADM) that distributes the force loading on the rotor disk; and an actuator line model (ALM) that distributes the forces on lines that follow the position of the blades. Simulation results are compared to wind-tunnel measurements collected with hot-wire and cold-wire anemometry in the wake of a miniature 3-blade wind turbine at the St. Anthony Falls Laboratory atmospheric boundary layer wind tunnel. In general, the characteristics of the wakes simulated with the proposed LES framework are in good agreement with the measurements. The ALM is better able to capture vortical structures induced by the blades in the near-wake region. Our results also show that the scale-dependent Lagrangian dynamic SGS models are able to account, without tuning, for the effects of local shear and flow anisotropy on the distribution of the SGS model coefficients. [Preview Abstract] |
Sunday, November 22, 2009 11:48AM - 12:01PM |
BS.00007: Large Eddy Simulations of boundary layer flow over fractal trees Jason Graham, Charles Meneveau A large-eddy simulation (LES) of flow over a canopy of fractal trees in the atmospheric boundary layer (ABL) is performed. The fractal trees provide complex boundary- turbulence interactions while maintaining tractable characteristics that can be systematically studied. LES are performed using the immersed boundary method following the implementation of Chester et al. (2007, J. Comp. Phys.). Simulations are performed for each fractal generation and the Reynods stresses and drag forces are computed. The LES results are used to analyze the impact of multiscale geometry on the Reynolds stress distribution and drag forces as a function of the generation number. This effort is also a feasiblity study for Renormalized Numerical Simulations (RNS) which is a methodology that allows for drag forces of unresovled generations to be computed by combining renormalizing techniques with the information from the resolved generations. Results are used to gain insight on scaling relationships between the drag forces and the generation number, and ultimately lead to better renormalization techniques for RNS. [Preview Abstract] |
Sunday, November 22, 2009 12:01PM - 12:14PM |
BS.00008: Large Eddy Simulation of Persistent Contrails in Wind Shear and Atmospheric Turbulence Alexander Naiman, Frank Ham, Sanjiva Lele, Jordan Wilkerson, Mark Jacobson A study of contrail evolution was conducted using a three-dimensional Large Eddy Simulation (LES). The LES solves the incompressible Navier-Stokes equations with a Boussinesq approximation for buoyancy forces on an unstructured periodic grid. The numerical scheme uses a second-order finite volume spatial discretization and an implicit fractional-step method for time advancement. Lagrangian contrail particles grow according to a microphysical model of ice deposition and sublimation. The simulation is initialized with the wake of a commercial jet superimposed on a decaying turbulence field. The ambient atmosphere is stable and has a supersaturated relative humidity with respect to ice. Grid resolution is adjusted during the simulation, allowing higher resolution of flow structures than previous studies. We present results of a parametric study in which ambient turbulence levels, vertical wind shear, and aircraft type were varied. We find that higher levels of turbulence and shear promote mixing of aircraft exhaust with supersaturated ambient air, resulting in faster growth of ice and wider dispersion of the exhaust plume. These results provide sensitivity data that improves understanding of the development of persistent contrails into contrail cirrus, a poorly characterized aspect of the climate impact of aviation. [Preview Abstract] |
Sunday, November 22, 2009 12:14PM - 12:27PM |
BS.00009: Large-eddy Simulation of Urban Boundary Layer Flow over Complex Topologies Byung-Gu Kim, Changhoon Lee Boundary layer flows over arrays of regularly distributed obstacles and a scaled real urban area in which various wind directions are considered were investigated by wind-tunnel experiment and large eddy simulation. Virtual boundary method is employed to represent the immersed complex geometries. Inflow conditions are generated such that given profiles of mean wind and turbulence properties such as integral length or time scales are matched. Constant Smagorinsky subgrid-scale model is used. Surface flow parameters such as friction velocity ($u_*$), roughness length ($z_0$) and displacement thickness ($d$) were evaluated by changing wind direction. It was found that the parameters for the arrays composed of slender rectangular cylinders whose characteristics are similar to a real urban area than cube arrays are highly sensitive to the wind direction. Many previous works that have been focused on cube arrays would differ from the real urban area in the characteristics of flow field. Large-eddy simulations were extended to the region of Daejeon city, Korea to study the province's characteristics. Velocity profile along the street canyon with which a wind direction aligned were closed to a linear profile rather than a constant or exponential distribution. Detailed wind field characteristics above and below the canopy will be presented in the meeting. [Preview Abstract] |
Sunday, November 22, 2009 12:27PM - 12:40PM |
BS.00010: A dynamic subgrid-scale parameterization of the effective wall stress in atmospheric boundary layer flows over multiscale, fractal-like surfaces William Anderson, Charles Meneveau A dynamic subgrid-scale parameterization for hydrodynamic surface roughness is developed for large-eddy simulation (LES) of atmospheric boundary layer (ABL) flow over multiscale, fractal-like surfaces. The model is tested on surfaces generated through superposition of random-phase Fourier modes with prescribed, power-law surface-height spectra. Surfaces are generated at fine-grained resolution and subsequently spatially filtered to various affordable LES resolutions. The lower boundary condition is prescribed based on the logarithmic law of the wall, where the unresolved roughness from the fractal surface is modeled as the product of local root- mean-square (RMS) of the unresolved surface height and an unknown dimensionless model coefficient. This coefficient is evaluated dynamically by comparing the plane- average wall-stress at two resolutions (grid- and test-filter scale, Germano et al., 1991). The results show that the method yields convergent results and correct trends. Limitations and further challenges are highlighted. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700