Master of Science (MS)


Environmental Sciences

Document Type



Quantifying resilience is difficulties due to the different definitions of resilience, the interchangeable uses with two other terms “vulnerability” and “adaptability”, as well as the lack of consensus on what indicators should be selected to quantifying resilience. This thesis research studied the community resilience in Louisiana by applying the Resilience Inference Measurement (RIM) model at two geographic levels: county level and zip code level. The RIM model accesses resilience by using three dimensions (exposure, damage, and recovery) and two abilities (vulnerability and adaptability). The types of coastal hazards included in this study were: coastal, flooding, hurricane/tropical storm, tornado, and severe storm/thunder storm. The study time period was 2000 to 2010. K-means clustering analysis was used to derive the resilience groups. Discriminant analysis was applied to validate the resilience rankings by using a set of indicator variables. At the county level, discriminant analysis yielded a remarkably high 93.8% classification accuracy when population growth rate in 2000-2010 was used as a recovering indicator and 28 adaptability variables were used to characterize the counties. The accuracy at the zip-code level decreased to 80.2% when population growth rate was used as a recovering indicator. In general, the findings at two different scales are consistent; counties and zip codes with higher socioeconomic status and more resources were found to be more resilient. Interestingly, the three most potent indicators revealed at both scales were the same, which are median rent, median value of owner-occupied housing units and housing density. These findings support the use of the RIM model to further explore adaptability indicators and the underlying process leading to resilience.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Lam, Nina