Modeling and Inversion of Shallow Seismic Data Including Nongeometrical Waves

Abstract

Solving a geophysical inverse problem requires making inferences about the earth from data. Since one always has only a finite number of (uncertain) data and since the models used to describe the earth are infinite dimensional (i.e., functions of space), it follows that if there are any models at all that fit the data, there will likely be many of them. Thus, finding a single model that fits the data is of limited value without a quantitative assessment of its uncertainty. During the course of this project we have developed novel theoretical and computational strategies for making statistically rigorous inferences about the earth's near-surface from full-waveform reflection and borehole seismic data. Our approach allows us to assimilate information at vastly different length scales and to take advantage of all the information in the seismic waveforms, as well as quantifying uncertainties in the data due to noise and theoretical errors. We have demonstrated the efficiency and utility of this approach on field data and have produced computer codes (using freely available compilers and message passing libraries) which perform in a scalable, distributed-parallel fashion on heterogeneous networks of workstations or shared-memory multiprocessors.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 29, 1997
Accession Number
ADA344956

Entities

People

  • Alberto Villarreal
  • John A. Scales
  • W. C. Navidi

Organizations

  • Colorado School of Mines

Tags

Communities of Interest

  • Advanced Electronics
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Abstracts
  • Case Studies
  • Computers
  • Data Sets
  • Earth Models
  • Frequency
  • Heterogeneous Networks
  • Inverse Problems
  • Models
  • Networks
  • Physics
  • Probability
  • Reflection
  • Uncertainty
  • Wave Phenomena
  • Waveforms
  • Waves

Readers

  • Computational Modeling and Simulation
  • Parallel and Distributed Computing.
  • Seismology

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms
  • Space