Estimation of Explosion-Induced Spall Surface Using a Parametric Inversion of Infrasound Data

Session: Explosion Seismology Advances [Poster]
Type: Poster
Date: 4/29/2020
Time: 08:00 AM
Room: Ballroom
Description: 

Buried explosions can dynamically deform the ground surface. The creation of this deformation surface (spall) is known to be a highly effective method to generate infrasound signals. The physical dimensions of the spall depend strongly on the depth and yield of the buried explosion. Furthermore, the physical dimensions of the spall surface have a significant effect on the amplitude, frequency and radiation pattern of the infrasound. Thus, being able to estimate the physical form of the spall surface using observed infrasound may help to estimate the depth and yield of a buried explosion. In this work, we investigate an inverse method to estimate the physical form of the spall surface based on observed infrasound signals. Our inverse scheme is based on the Rayleigh integral, which is a forward model that predicts the infrasonic signals given the acceleration history of an arbitrarily shaped spall surface. However, we simplify the problem by approximating the spall surface with analytical functions of low dimensionality. For example, by approximating the spall surface as a two-dimensional Gaussian, we only have to invert for three parameters: the width of the spall surface in the x- and y-directions and the maximum acceleration of the surface. This approach, termed a parametric inversion, significantly reduces the non-uniqueness of the inversion. We explore the feasibility, uniqueness and stability of a parametric inversion using both synthetic data and infrasound data collected as part of the Source Physics Experiment. We employ a parallelized grid search algorithm to fit data (both synthetic and observed) to analytical surfaces that may represent actual spall surfaces and test the relative sensitivity of the inversion parameters to the data fit. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Presenting Author: Christian Poppeliers

Authors