Date of Award


Document Type

Campus Access Thesis



First Advisor

Susan L Cutter


Beginning in the latter part of the 20th century, the United States` population has migrated away from rural environments towards urban centers and coastal areas. Both areas are considered prone to natural hazards. Two contrary trends have appeared at the same time: 1) property and crop damage from natural hazards has increased; and 2) hazard mortality has decreased. Despite the overall decrease in natural hazard related fatalities, there is little understanding of spatial differences in mortality between urban and rural environments.

Hazard mortality data, found in multiple databases, are inconsistent. The collection and reporting of data lacks consistent standards and it is impossible to determine the exact number of fatalities resulting from a specific event such as 2005`s Hurricane Katrina. This thesis utilizes two premier hazard mortality databases, the National Climatic Data Center`s Storm Data and the National Center for Health Statistics - Centers for Disease Control and Prevention`s Compressed Mortality File, as well as several auxiliary databases to establish an all-encompassing hazard mortality dataset covering the period from 1 January 1996 through 31 December 2008. This analysis provides insight into how different databases, methods of collection, standardization techniques, and specific hazards affect spatial patterns of mortality. The final hazard mortality database (HazMort) is used to determine if there are differences in urban/rural mortality as a function of age, gender, and race. The results of the analysis show that un-standardized mortality values exhibit an excess number of urban fatalities. The reverse is true after standardization: rural areas experience an excess number of fatalities. Exceptions do exist. For example, after standardization, some specific hazards and demographic variables such as age (under 19), African-American, and other racial groups continue to exhibit excess urban fatalities.