Developing More Accurate Models of Tornados

  • Author(s) / Creator(s)
  • Abstract—Tornados are a major hazard and ever-present threat in many regions, with the potential to cause wide-scale loss of life and damage to infrastructure. Many researchers have attempted to develop tornado simulation techniques that amount to vortex generation with the goal of understanding the characteristics of tornado maintenance and intensity. Traditionally, these models comprise an analysis of wind-field data using Doppler radar collection, analytical tornado systems, laboratory experimental modelling, or the more recent numerical simulation techniques. These models, in application to wind engineering, focus on the tornado vortex without its parent storm and, as such, rely on artificial boundary conditions resulting in uniform, axisymmetric rotation about a vertical axis. The present work focused on developing superior tornado analysis techniques but first required generation of a method of tracking a tornado vortex centreline throughout its lifecycle, which corrects for the deficiencies in other vortex tracking methods when applied to tornados. The method identifies a clearly defined line for the vortex centre without need for extensive user-input. This method has been validated on the dataset of a tornado spawned from a supercell, in a meteorological numerical cloud model simulation at full-scale while being able to freely form and dissipate in a large, yet well-resolved, domain. Thus, quantitative assessment of a tornado wind-field may be anchored to an origin at the tornado centre that permits analysis of the discrepancy between the velocities of wind-fields generated by the previous models with those of a more physically realistic tornado. In future, analysis will be performed to understand how the in-flow at the simulation
    domain boundaries can be used to generate more accurate tornado models. This will provide the optimum ratio of tornado radius to the size of its domain and a set of detailed boundary conditions to apply to tornado simulations without the need to perform computationally expensive simulations. The work described above has provided findings that outline the extent to which current models of tornados underestimate their destructive potential and how they may be improved to provide superior engineering analysis.

    Part of Proceedings of the Canadian Society for Mechanical Engineering International Congress 2022. (after the abstract)

  • Date created
  • Subjects / Keywords
  • Type of Item
    Article (Published)
  • DOI
  • License
    Attribution-NonCommercial 4.0 International