Construction of a Tropical Cyclone Size Dataset using Retroactive Analysis Data with a Damage Application
Semester of Graduation
Master of Science (MS)
Geography and Anthropology
This thesis introduces a new tropical cyclone (TC) size dataset. Using the radius of the outermost closed isobar (ROCI) as the size metric of focus, a comprehensive record of TC size at landfall was constructed for tropical storms and hurricanes that made landfall in the United States along the East Coast and in the Gulf of Mexico. The ROCI information was derived from mean sea level pressure maps generated using the NCEP/NCAR Reanalysis I dataset over a 69-year period (1948 – 2016). Construction of the dataset involved using two methodologies, one based on hourly interpolated HURDAT2 best tracks and the other based on landfall indicated by the reanalysis generated maps. Descriptive statistics were calculated for ROCI with respect to the dataset as a whole, intensity, and landfall location. Both methods were compared against each other, both with respect to ROCI as well as landfall locations. The results indicated that the two methods generated statistically identical ROCI, even though individual TCs could have differing ROCI values. The results also indicated that there was no significant trend in landfall ROCI over time. With respect to landfall locations, the results indicate that roughly two-thirds of all TCs in the dataset experienced a westward shift in TC center landfall location relative to the best track center location, with the displacement more prevalent in the Gulf of Mexico. A secondary analysis was conducted to ascertain the relationship between TC size and total economic damage, using damage data collected by Icat. The results of this analysis suggest a significant relationship between TC size and damage. This dataset serves as a prototype, with future work focusing on improving and extending the dataset.
Thompson, Derek Trent, "Construction of a Tropical Cyclone Size Dataset using Retroactive Analysis Data with a Damage Application" (2018). LSU Master's Theses. 4737.