B0217
Title: Adapting conditional simulation using circulant embedding for irregularly spaced data
Authors: Soutir Bandyopadhyay - Colorado School of Mines (United States) [presenting]
Maggie Bailey - Colorado School of Mines (United States)
Douglas Nychka - Colorado School of Mines (United States)
Abstract: Computing an ensemble of random fields using conditional simulation is an ideal method for retrieving accurate estimates of a field conditioned on available data and for quantifying the uncertainty of these realizations. However, methods for generating random realizations are computationally demanding, especially when the estimates are conditioned on numerous observed data and for large domains. A new and approximate conditional simulation approach is applied that uses circulant embedding, a fast method for simulating Gaussian processes. However, standard CE is restricted to simulating stationary Gaussian processes (possibly anisotropic) on regularly spaced grids. We explore two new algorithms that adapt CE for irregularly spaced data points with applications to the U.S. Geological Survey's software ShakeMap, which provides near-real-time maps of shaking intensity after a significant earthquake occurs. It is found that one method provides better accuracy and efficiency.
