ORCID

https://orcid.org/0000-0001-7802-2662

Keywords

storm surges, temporal clustering, frequency, independent events

Abstract

Coastal regions face compounding risks when extreme events strike in rapid succession, yet most statistical frameworks treat such events as independent, a simplification that can critically underestimate hazard. This thesis addresses that gap through three interconnected contributions focused on the temporal clustering of storm surges at the global scale. A new automated method is developed to extract independent storm surge events from continuous tide gauge records. By leveraging event correlations and a soft-margin approach that accounts for local variability, the method identifies characteristic event durations across 1,485 global tide gauge stations, revealing the influence of both local site conditions and seasonality. The identified events are then used to validate a state-of-the-art hindcast model. Temporal clustering of storm surges is then analyzed globally. Results show that 92% of coastal locations exhibit significant clustering for 1-year return level events, with two distinct regimes identified: short-timescale (intra-annual) and long-timescale (inter-annual) clustering. Critically, over 80% of stations violate the Poisson independence assumption commonly used in coastal hazard assessments. Lastly, other overdispersed count models are evaluated as alternatives to the Poisson framework for characterizing seasonal storm surge frequencies. A Hurdle Negative Binomial formulation emerges as a robust and accessible solution that outperforms or matches more complex models.

Together, these methodological advances provide risk managers, researchers, and policymakers with improved tools for quantifying coastal hazards and lay the groundwork for extending temporal clustering analyses to other hazards, compound events, and future climate scenarios.

Completion Date

2026

Semester

Spring

Committee Chair

Thomas Wahl

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Department of Civil, Environmental, and Construction Engineering

Format

PDF

Document Type

Dissertation

Identifier

DP0053116

Download

Share

COinS
 

Accessibility Statement

This item was created or digitized prior to April 24, 2027, or is a reproduction of legacy media created before that date. It is preserved in its original, unmodified state specifically for research, reference, or historical recordkeeping. In accordance with the ADA Title II Final Rule, the University Libraries provides accessible versions of archival materials upon request. To request an accommodation for this item, please submit an accessibility request form.