Wind Flow Characterizations within Uniform and Non-uniform Building Configurations using Wind Tunnel Data and Large Eddy Simulations
Citation:
Pirhalla, M., J. Retter, D. Heist, C. Owen, W. Tang, L. Brouwer, AND T. Odom. Wind Flow Characterizations within Uniform and Non-uniform Building Configurations using Wind Tunnel Data and Large Eddy Simulations. Presented at 2023 EPA International Decontamination Research and Development Conference, Charleston, SC, December 05 - 07, 2023.
Impact/Purpose:
This poster presentation will be presented at EPA's International Decontamination Conference organized by ORD/CESER. This project leverages data from a series of large-eddy simulations (LES) and EPA meteorological wind tunnel experiments that use two 1:200 scale models designed to represent a residential neighborhood with avenues and rows of uniform and non-uniform buildings. The goal of this project is to compare the wind tunnel and LES datasets to inform improved urban wind speed and turbulence profiles (and potential parameterizations) in Gaussian dispersion models, which may inform emergency response scenarios following a harmful release in a populated area. This presentation may be of interest to personnel responsible for planning for emergency airborne release situations because the presence of buildings can impact downwind dispersion. This work may also interest the atmospheric science community through fine scale turbulence and dispersion modeling.
Description:
Dispersion models, which are important tools for efficient routine releases and emergency response preparation scenarios, typically incorporate logarithmic wind speed profiles when simulating flow and downwind dispersion. This may be acceptable for regions with low surface roughness and minimal obstructions. However, within urban (or suburban) canopies, vertical wind speed and turbulence profiles may be altered due to wake turbulence and channeling flows generated by the presence of buildings and other structures. The added drag tends to slow the wind speed within the canopy and develop an inflection point at or above the mean building height. These situations can affect the prediction of downwind dispersion and concentration plume behavior, which may not be accounted for in some traditional modelling exercises. This project leverages data from a series of large-eddy simulations (LES) and EPA meteorological wind tunnel experiments that use two 1:200 scale models designed to represent a residential neighborhood with avenues and rows of two-story rectangular houses. Two building configurations are compared: one with uniform rows of buildings with heights of 12 m full-scale, and a second with an alternating array of variably sized, non-uniform buildings with heights ranging from 12-20 m. The LES model results enable fine-scale comparisons with Laser Doppler Velocimetry (LDV) and Particle Image Velocimetry (PIV) data gathered in the wind tunnel. The LES also permits us to simulate a variety of additional oblique incident wind directions (from 0-50 degrees). The goal of this project is to compare the wind tunnel and LES datasets to inform improved urban wind speed and turbulence profiles (and potential parameterizations) in Gaussian dispersion models, which may inform emergency response scenarios following a harmful release in a populated area. Preliminary results show that wind tunnel datasets compare very well with the LES model simulations and accurately capture regions of high wind shear at the tops of buildings, including profiles measured within intersections and in areas bounded by buildings. The LES simulations suggest the consideration of dispersive stresses (in addition to turbulent stresses) when developing future characterizations in dispersion models, especially for non-uniform building scenarios.