Purdue University Graduate School
Geldner Dissertation Final 2.0.pdf (3.02 MB)

Advances in Coastal Flood Risk Analysis

Download (3.02 MB)
posted on 2023-05-10, 21:08 authored by Nathan B GeldnerNathan B Geldner


Much work has been done to advance the state of risk-informed decision-making to protect against coastal flooding. The state of the art as practiced in the state of Louisiana, the Coastal Louisiana Risk Assessment Model (CLARA) characterizes flood risk driven by surge and waves from tropical storms as a random process and estimates the distribution of flood depths and resulting economic damage. This dissertation identifies three key limitations of coastal flood risk assessment as applied to the state of Louisiana, proposes methods to address them, and demonstrates each method in a case study. The first limitation identified is that the CLARA model addresses surge- and wave-driven flood hazard but does not account for rainfall-driven and riverine hazard. To address this limitation, chapter 2 presents an extension of the methods used in CLARA developed as part of the Louisiana Watershed Initiative which permits characterization of compound hazard consisting of surge, rainfall, and riverine hazard from tropical cyclones. The second limitation identified is that due to its computational cost CLARA is used in Louisiana’s 2023 Coastal Master Plan to evaluate pre-specified and static flood risk mitigation projects over a small set of possible future landscapes; it would be preferable to use an optimization-driven approach to generate efficient and adaptive combinations of projects which balance performance across a diverse set of possible future landscapes. To address this limitation, chapter 3 presents an analysis applying a spatially narrower but computationally inexpensive model based on CLARA called the SWaMPS model to a case study to determine the extent to which optimization-based adaptive mitigation strategies which respond to observed climate change trajectories can outperform similar static strategies. The third and final limitation identified is that while the most recent iteration of the CLARA model can support algorithmic optimization of cost-effectiveness of building-level mitigation strategies such as retrofits to increase the first floor elevation of single-family residences, the principal metric historically used in flood risk mitigation is reduction in economic damage measured in dollars, the optimization of which on the level of individual buildings would implicitly prioritize expensive structures and therefore may neglect impoverished neighborhoods. Chapter 4 addresses this limitation with a proposed alternative efficiency metric which treats individual homes as equally valuable, the optimization of which results in greater investment in impoverished neighborhoods without explicitly targeting final expected damage distributions or individual groups. 


Degree Type

  • Doctor of Philosophy


  • Industrial Engineering

Campus location

  • West Lafayette

Advisor/Supervisor/Committee Chair

Professor David R. Johnson

Additional Committee Member 2

Professor Andrew Liu

Additional Committee Member 3

Professor Venkatesh Merwade

Additional Committee Member 4

Dr. Scott Hemmerling

Usage metrics



    Ref. manager