Degree Type

Thesis

Date of Award

1-1-2006

Degree Name

Master of Science

Department

Geological and Atmospheric Sciences

Major

Meteorology

Abstract

Because of the lack of field measurements near the ground in tornadoes, numerical simulation may provide the best estimate of the near-ground wind profiles, assuming such simulations are realistic and agree well with the available observations at higher levels. An accurate understanding of the loads requires knowledge about near-ground tornado winds. The numerical simulations based on the ISU laboratory tornado model agree well with the Spencer, South Dakota tornado of 30 May 1998. The wind measurements taken by portable Doppler radars are restricted to levels higher than at least 20-50 m above the ground. It is necessary to understand tornado-induced wind loads on typical structures to help improve structural designs to resist tornado winds. Computational Fluid dynamic (CFD) simulations are used as a tool to validate a laboratory model's ability to simulate a real tornado vortex by studying the near ground flow field. The sensitivity of solutions to parameters such as the inflow depth, inflow radius, outflow radius, mesh size, boundary condition, surface roughness and Swirl ratio was explored by designing a numerical model based on the ISU laboratory tornado simulator. The study suggests that it is important to correctly choose the inflow radius to be far enough away from the tornado to minimize the influence of the boundary conditions on the vortex, but close enough to the tornado to reduce the influence of difficult-to-simulate surface roughness. The simulated core radius is greatly affected by both outflow radius and Swirl ratio. The fine mesh size provides a higher resolution than Doppler radar data, and stronger tangential velocities are thus simulated at the core radius. The effects of surface roughness are to reduce the tangential velocity and slightly enlarge the core radius. Three ways of increasing roughness in the numerical simulation are tested. It was found that the most efficient way to represent roughness was to increase the roughness elements' height. These conclusions could assist in the design of future numerical or laboratory experiments exploring the near ground flow more closely.

DOI

https://doi.org/10.31274/rtd-20200618-27

Copyright Owner

Le Kuai

Language

en

OCLC Number

71341305

File Format

application/pdf

File Size

88 pages

Share

COinS