Master of Science (MS)


Environmental Sciences

Document Type



Natural disasters devastate the United States through both economic loss and loss of life. The world wide economic damage that results from natural disasters has more than tripled in the last thirty years. Of these natural disasters, floods are the most chronic and costly disasters, comprising an average $5 billion dollars of damage each year. FEMA has released a new software program called HAZUS-MH, which attempts to capture economic losses caused by flooding before losses occur and predict losses from real-time events. This estimate is accomplished through the coupling of flood hazard modeling with local data. FEMA’s goal is that the information constructed within the program will help planners to mitigate and capture flood related losses. This study provides a methodology for assessing the accuracy of HAZUS level one flood loss estimates by examining the extent to which HAZUS default building stock inventory data represents the built local environment. The study area is concentrated in the northwest corner of Livingston Parish, Louisiana. The area is comprised of 200 census blocks that were chosen due to their proximity to the Amite River. Thus it is an area prone to floods. Livingston Parish is located in the Mississippi River and Lake Maurepas Basin, which collectively cover approximately 236,000 acres. 70% of the Parish’s land is located within FEMA’s 100-year flood plain. Building count for structures was obtained using remote sensing technology, processed and used to populate HAZUS ® MH default databases. Flood loss estimations were run for all of the data sets and results were compared for a significant difference. Differences in flood loss between the two analyses were found in isolated areas. This demonstrated the need to incorporate growth and development information into flood loss estimation methodologies.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

John Pine