Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 425 - Nonparametric Methods for Dependent Data
Type: Contributed
Date/Time: Wednesday, August 10, 2022 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics and the Environment
Abstract #323019
Title: Prediction and Uncertainty Quantification of Non-Gaussian Spatial Processes with Applications to Large-Scale Flooding in Urban Areas.
Author(s): Sweta Rai*
Companies: Colorado School of Mines

This is about developing a framework for river flood modeling that leverages large hydraulic models and facilitates rapid large-scale flood simulations at a high resolution. Due to global warming, large floods have become a characteristic feature of the climate around the world. This demands attention to more robust, efficient, and real-time flood modeling. For most river engineering problems, there is need for accurate quantification of water depth and elevation. There are numerous hydraulic modeling tools capable of representing this but they are computationally expensive and time consuming. In particular, we aim to establish methodology for building accurate and fast statistical emulators as alternatives to running complex physical models directly using machine learning algorithms with spatial extreme distributions to represent flooding in a complex domain. We use the application of generalized Pareto distribution (GPD) to the statistical analysis of extremes.

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

Back to the full JSM 2022 program