Remote Sensing Application to Assess Resilience in Coastal Communities

Natural disasters such as hurricanes have been affecting thousands of people annually with devastating consequences such as loss of life, vegetation and infrastructure. Vegetation losses such as coastal erosion, downed trees and infrastructure disruptions (e.g., toppled power lines) often lead to roadway closures. These disruptions are also life threatening for the victims in the affected region. Emergency officials, therefore, have been trying to find ways to alleviate such problems by identifying those locations that pose high risk in the aftermath of hurricanes. In order to help emergency agencies achieve this goal, this dissertation explores the design and implementation of a remote sensing technology-based methodology for coastal community resiliency assessment. Specifically, it seeks to: (1) develop and adopt a modern remote sensing technology for rapid assessment of damage caused by hurricanes; (2) identify reliable indicators for detecting the impacted region and damage severity after hurricanes; (3) develop a comprehensive debris damage assessment given different storm intensities/strengths; and (4) compare the proposed storm debris prediction model findings to those obtained from existing estimation methods. Chapter 2 of this dissertation presents a Google Earth Engine and MATLAB-based remote sensing application to coastal barrier islands. This application especially focuses on the well-known critical width concept as a measure of resiliency of barrier islands to coastal overwash. Our results reveal the potential use of the developed application of remote sensing for detailed barrier island analysis and for future improvements of barrier island resilience parameterizations. Chapter 3 presents another Google Earth Engine and Geographical Information Systems-based remote sensing application as a two-stage model to analyze the impact of Hurricane Michael on the City of Tallahassee, Florida. The proposed innovative application was used to develop city-wide hurricane risk maps for the entire city at the level of census blocks. Our findings showed that the Northeast side of the city had high risk levels, which is a region where more 65+ populations live, and the disruptions on that side of the city can lead to dramatic consequences due to the fragility of these seniors in the aftermath of hurricanes. Chapter 4 developed a third Google Earth Engine and Geographical Information Systems-based remote sensing application designed to perform an assessment of the damage caused by different hurricanes and storms on example coastal communities. The method used Normalized Difference Vegetation Index (NDVI) in order to estimate the greenness change before and after the hurricanes hit. The findings on the estimated debris were also compared with the debris calculated by the U.S. Army Corps of Engineers (USACE) model. The proposed application was utilized to conduct a comparative analysis of the impacts of different storms and hurricanes on the Bay County, Florida with a focus on the demographic factors such as the population and roadway network. Results of the analysis show that sub-urban areas, urban areas, moderate and high roadway density areas generated more debris and NDVI changes than rural and low roadway density areas. The results of NDVI change approach indicate that USACE debris model is overestimating the debris volume for the hurricanes that are category 3 or greater. Particularly, higher intense hurricane would enlarge this error. The applications developed in this dissertation focused only on coastal regions; however, the proposed methodology can be successfully extended to any other location given the data availability. The proposed risk index and damage assessment can also be enhanced using relevant data on power outages and socioeconomics.