Date of Award

12-14-2016

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Mathematics and Statistics

First Advisor

Jing Zhang

Second Advisor

Yi Jiang

Third Advisor

Xin Qi

Abstract

Sea-level rise is projected to have a wide range of effects on coastal environments, developments, and infrastructure. Based on the U.S. Geological Survey (USGS) Coastal Vulnerability Index (CVI) system data, we developed a two-stage model; firstly, the Bayesian Network (BN) is used to define relationship among driving forces; secondly, the logistic regression is used to evaluate direct association for direct factors related to Shoreline Erosion. Using this two-stage approach, increased sea-level (OR: 4.03[3.72,4.38]), higher Wave Height (OR: 0.56[0.54,0.61]), smaller Tidal Range (OR: 1.68[1.52,1.87]) and smaller Coastal Slope (OR: 0.45[0.44,0.49]) are directly associated with Shoreline Erosion in Atlantic Ocean; Geomorphology setting (OR: 9.35[6.33,14.18]) in high risk regions, such as beaches, is identified as direct association with Shoreline Erosion in Gulf of Mexico; Smaller tidal range (OR: 0.10[0.04,0.27] directly associated with Shoreline Erosion in Pacific Ocean. These direct factors were evaluated predictive ability with accuracy rates 0.59 and AUC 0.63.

DOI

https://doi.org/10.57709/9409224

Share

COinS