Evaluation of the atmospheric RAMS model in an obstacle resolving configuration

Abstract

The Regional Atmospheric Modelling System (RAMS) was modified and configured to simulate the flow around a rectangular building, corresponding to Case A1-1 in the CEDVAL Project. This experiment was designed to provide a comprehensive compilation of wind tunnel datasets that can be used for validation of numerical flow and dispersion models. A Cartesian grid implemented in the RAMS model permitted the explicit description of the obstacle. The model grid resolution was of the order of 1 m, high enough to resolve the most significant flow characteristics. Two turbulence closures largely used at these scales, k − ε and Renormalization Group k − ε, were implemented in RAMS, tested and evaluated. Some preliminary tests were performed to investigate the effect of the boundary conditions on the flow behaviour at the obstacle walls. A simplified approach, assuming that the wind changes logarithmically in the ambient atmosphere near the building's faces was implemented. Model results show that RAMS successfully reproduces the main flow characteristics. The performances of the different turbulence closures are discussed through a comparison with the observed data and a statistical analysis. A good agreement was obtained for the horizontal and vertical velocities, while the results for the turbulent kinetic energy are generally less satisfactory, especially above the obstacle's height.

The Quasi-Solid Box Method for Simulating Wind Around Obstacles in the System for Atmospheric Modeling

“…Also, the simulated vortex extends as far as x/H = 1.9 (47 m) in the simulation compared to x/H = 1.7 (42 m) in the observations. We note that other models also have tended to overextend the leeside vortex in this CEDVAL case (e.g., Trini Castelli & Reisin, 2010;Zhang et al, 2016).…”

“…To test the performance of the QSBM in simulating the flow around solid obstacles, we performed two LES simulations for the case of a single rectangular building and compared it to the observational dataset from the Compilation of Experimental Data for Validation of Microscale Dispersion Models (CEDVAL; https://mi-pub.cen.uni-hamburg.de/index.php?id=433). We use reference cases from A1-1, which have been previously used for evaluation of other numerical models (e.g., Trini Castelli & Reisin, 2010;Zhang et al, 2016). Specifically, we will use A1-1 to evaluate the velocity field and A1-5 to evaluate the dispersion of a gas tracer.…”

An intercomparison study between RAMS and CRES-Flow-NS models and evaluation with wind tunnel experimental data: Toward improving atmospheric modeling for wind resource assessment

“…RAMS uses the "staggered" Arakawa C-grid discretization scheme. The RAMS model is capable of resolving a wide range of atmospheric turbulent flows according to their scale and amongst them microscale turbulent flows (Ying et al, 1994;Canuto, 1995, 1997;Trini Castelli et al, 2005, Trini Castelli andReisin, 2010;De Wekker et al, 2005). It has several other capabilities such as the two-way interactive nesting with any number of grids and the use of a large variety of lateral Boundary Conditions (BC).…”

Review and Validation of MicroSpray, a Lagrangian Particle Model of Turbulent Dispersion

“…In studying the flow around buildings, two-equation models, like k-e or k-ω closures, being k the turbulent kinetic energy and ω the vorticity, are generally applied. A standard version of the k-e turbulence closure and the renormalization group (RNG) k-e turbulence model were implemented and tested in RAMS, giving origin to the RAMS6.0-mod [Trini Castelli et al, 2001Ferrero et al, 2003;Reisin et al, 2007;Trini Castelli and Reisin, 2010]. Both closures solve the dynamical equations for the turbulent kinetic energy, k, and its dissipation term, e. With respect to the standard k-e model, the RNG-k-e turbulence scheme [Yakhot et al, 1992] includes also an additional sink term in the turbulence dissipation equation to account for nonequilibrium strain rates and employs different values for the closure coefficients [Trini Castelli and Reisin, 2010].…”