Non-thunderstorm-related effects on the near-surface wind field in strong tornadoes

On an average annual basis, tornadoes can produce billions of dollars in damage and upwards of 50 fatalities as their extreme wind speeds and other associated hazards (e.g., debris) wreak havoc on human lives and property. Because of the extreme wind loading that can be associated with the passage of a tornado, understanding their impacts on structures remains of high interest to engineers. As the American Society of Civil Engineers prepares to add provisions for tornado loading to the next iteration of its design standard (i.e., ASCE 7-22), further understanding of a tornado’s near-surface wind characteristics is vital. To further our understanding, direct observations of tornadoes are desired such that the evolution and structure of their near-surface profiles might be better understood. This, of course, comes with many difficulties as tornadoes are inherently hard to sample directly due to their transient nature, relative rarity, and other logistical difficulties associated with their observation. For these reasons, numerical techniques and post-event reconnaissance are often used to study tornado impacts. In this study, large-eddy simulations (LES) incorporating idealized terrain and structures are used in tandem with tornado damage surveys in an effort to characterize the effects of surface features (e.g., terrain and the built environment) on tornado maintenance, and more importantly, the near surface wind velocity, which is directly relevant to tornado loading of structures. This thesis is broken into two main sections. The first section (Chapter 3) focuses on the impacts of terrain on a tornado vortex. While the effects of terrain on ABL winds have long been accounted for in ASCE 7, few studies have examined the potential effects of complex terrain on tornado vortices, which is of interest for future iterations of the tornado-based design code. In this section, we first investigate the direct effects of complex terrain on a tornado vortex using numerical modeling techniques. The results of the simulations are then compared to two cases of tornado-induced treefall that occurred in the southeastern US. The two cases, the 2018 Jacksonville, AL, tornado and the 2020 Nashville-Cookeville, TN, tornado both occurred in regions of complex terrain. A combination of ground damage survey and aerial survey data are used to map the directions of fallen trees for these cases. Observed treefall is examined with respect to the slope and orientation of terrain features and compared to both the expected fall patterns based on theoretical treefall models, and simulated fall patterns produced by the LES. In short, both the modeled and observed results show the terrain’s tendency to modify both the direction and intensity of the critical wind field (i.e., the wind field capable of downing trees) as the tornado traversed the terrain features. Fall patterns deviated from their expectation in the presence of complex with fall directions tending to paralleling terrain contours, and greater tree damage width between terrain features or in valleys. These results suggest that the wind field intensifies in these regions. The second section (Chapter 4) aims to better understand the impacts of the built environment on tornadoes, and how structures themselves might alter the wind field. First, single, static structures acting essentially as impermeable "blocks" are used to simulate large, engineered structures. The results show evidence that the structures can significantly alter the near-surface flow, including some scenarios where the maximum wind speeds are greater than those produced in the control simulation. The simulation complexity is then increased to include neighborhood-style structure orientations (e.g., cul-de-sac, rectangular city block, etc.). In the neighborhood cases, we aim to understand how groups of structures in typical neighborhood configurations interact with one another under extreme wind speeds, and how their collective aerodynamic characteristics may act to protect (or harm) those structures immediately surrounding them. While these simulations remain idealized to this point, a novel method has been developed to simulate the progressive structural damage in the model representing different degrees of structural damage. As structures are damaged, their aerodynamic impacts on the tornado’s wind field changes, which may lead to sudden changes in wind speed and direction. The resulting effects on the house-layer wind field (i.e., from the ground to roof peak) is evaluated as a function of the street configurations, and the degree to which the surrounding structures are damaged or undamaged. These results are compared to observations from ground and aerial damage surveys from several significant tornadoes including the 2021 Naperville, IL, tornado, the 2020 Nashville-Cookeville, TN, tornado, the 2013 Moore, OK, tornado, and the 2011 Joplin, MO, tornado. To conclude, the results produced herein are discussed as they pertain to our understanding of tornado-induced wind loading, future applications to ASCE 7, and the future of tornado-based structural design. Modifications to the current damage survey procedure are discussed based on the current needs of tornado damage research.U of I OnlyAuthor requested U of Illinois access only (OA after 2yrs) in Vireo ETD syste