Online Program Home
My Program

Abstract Details

Activity Number: 573 - Simulation and Stochastic Bayesian Modeling
Type: Contributed
Date/Time: Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #305127
Title: Computer Model Emulation with High-Dimensional Zero-Inflated Spatial Data: An Application to Storm Surge
Author(s): Pulong Ma*
Companies: SAMSI/Duke University
Keywords: Uncertainty quantification; Probabilistic risk assessment; Storm surge; Zero-inflated emulator; Gaussian process; Computer experiments

Complex computer models of real-world processes (or simulators) are an essential ingredient to carry out uncertainty quantification in virtually every field of science and engineer. In coastal emergency risks assessment, storm surge is one of the most severe natural disasters, and it can lead to significant flooding in coastal areas and severe damages to the life and property from a hurricane. Quantifying the risk to storm surge hazard requires large-scale numerical simulations of hurricanes from storm surge modeling systems. The Advanced Circulation model is one of the primary storm surge models used to predict storm surge and control the impact of storm damage in the United States. A crucial need is the development of an emulator to facilitate risk quantification. However, the model outputs are high-dimensional zero-inflated spatial data, and Gaussian processes are unable to handle such data directly. To tackle both modeling and computational challenges, we take a hierarchical modeling approach to develop a zero-inflated emulator. This methodology is applied to storm surge over coastal areas of the United States and coastal emergency risks assessment.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program