This dissertation includes a series of chapters that evaluate tornado casualties in a spatial, historical, and social context. Previous studies have identified the location of tornado casualties using the best available data and assessed the link between social, economic, and demographic factors to the number of tornado casualties. This research builds and deviates from earlier research through the use of a number of geographical methods. More specifically, the research presented here aims to (1) predict the rate of tornado casualties as a function of tornado strength and population, (2) provide a dasymetric method to estimate socioeconomic and demographic variables at the tornado-level, and (3) define and identify unusually devastating tornadoes---those that cause significantly more casualties than some expected rate---which in turn recognizes vulnerable communities throughout the United States. Tornado records are fit to two different statistical models containing estimates of energy dissipation and population density: (1) an additive model and (2) an interactive model. Tornado energy dissipation is estimated through a physical model, which expresses the strength of a tornado. Results show that, for the additive model, a doubling in population increases the casualty rate by 21% [(17, 24)%, 95% credible interval] and a doubling in energy dissipation increases the casualty rate by 33% [(30, 35)%, 95% credible interval]. Results also show that, for the interactive model, the percentage increase in casualties with increasing energy dissipation increases with population density, and the percentage increase in casualties with increasing population density increases with energy dissipation. Estimates of socioeconomic and demographic variables at the tornado-level are found through dasymetric calculations that can be analyzed independently or in combination with other attributes in the historical record. These estimates are validated using known fatalities and actual tornado paths found within the National Weather Service's Damage Assessment Toolkit. Results show large correlation between estimated and observed fatalities exceeding .93 for four distinct age groups and .99 for sex. Unusually devastating tornadoes are defined and identified through a model for tornado casualties that builds on the interactive model, but includes estimates of socioeconomic and demographic variables. Results show that unusually devastating tornadoes can occur anywhere in the United States, but appear more consistently over parts of the rural South. By identifying clusters of unusually devastating tornadoes, individual communities can be further examined. For example, three examples of unusually devastating tornadoes include: (1) the 1998 Spencer, South Dakota tornado, (2) the 2000 and 2003 Camilla, Georgia tornadoes, and (3) the 2015 Garland-Rowlett, Texas tornado. Each of these cities have their own socioeconomic and demographic profiles, yet were hit by tornadoes that caused more casualties than expected given a model for tornado casualties. The results provide a substantial improvement in the quantitative and geographic understanding of tornado casualties across the United States. They also set the stage for future critiques of the systems in place that drive human vulnerability to tornadoes. In this sense, the results highlight the need for additional work to be devoted to understanding the tornado casualty problem. The methods presented here supply a foundation for future studies that evaluate changes in the tornado casualty landscape and make sense of the challenges many communities face with regard to the tornado hazard. / A Dissertation submitted to the Department of Geography in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester 2018. / November 30, 2018. / Casualties, Energy, Tornado, Vulnerability / Includes bibliographical references. / James B. Elsner, Professor Directing Dissertation; Robert Hart, University Representative; Victor Mesev, Committee Member; Tyler McCreary, Committee Member.
| Identifer | oai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_709752 |
| Contributors | Fricker, Tyler (author), Elsner, James B. (Professor Directing Dissertation), Hart, Robert E. (Robert Edward) (University Representative), Mesev, Victor (Committee Member), McCreary, Tyler (Committee Member), Florida State University (degree granting institution), College of Social Sciences and Public Policy (degree granting college), Department of Geography (degree granting departmentdgg) |
| Publisher | Florida State University |
| Source Sets | Florida State University |
| Language | English, English |
| Detected Language | English |
| Type | Text, text, doctoral thesis |
| Format | 1 online resource (127 pages), computer, application/pdf |
Page generated in 0.0016 seconds