Advertisement

Application of Multilayer Perceptrons to Earthquake Seismic Analysis

  • Hengchang Dai
Part of the Modern Approaches in Geophysics book series (MAGE, volume 21)

Abstract

Two multilayer perceptron (MLP) methodologies are developed to tackle the problems of earthquake seismic arrival picking and identification. The arrival picking is achieved using the vector modulus of three-component (3-C) seismic recording or the absolute value of amplitude of single-component recording, as the input to an MLP. A discriminate function, determined from the output of the trained MLP, is then employed to define the arrival onset. Arrival identification is achieved using the degree of polarization of the arrival segment of the 3-C recordings as the input to an MLP; the output of the trained MLP indicates the arrival type. The results are encouraging. This work demonstrates significant potential in the application of MLPs to automatic earthquake analysis and other geophysical problems. In principle, it should be possible to develop a complete automatic seismic analysis system based on the two methodologies described herein.

Keywords

Onset Time Input Node Noise Burst Arrival Type Vector Modulus 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Anderson, K. R, 1978, Automatic processing of local earthquake data: PhD thesis, Massachusetts Institute of Technology, Cambridge, Massachusetts, USAGoogle Scholar
  2. Basham, P. W., and Ellis, R. M., 1969, The composition of Pcode using magnetic tape seismograms: Bull., Seis. Soc. Am., 59, 473–486.Google Scholar
  3. Cichowicz, A., 1993, An automatic S-phase picker: Bull., Seis. Soc. Am., 83, 180–189.Google Scholar
  4. Dai, H. C., 1995, Analysis of local earthquake data using artificial neural networks: Ph.D. thesis, Univ. of Edinburgh, Scotland.Google Scholar
  5. Dai, H. C., and MacBeth, C., 1995, Automatic picking of seismic arrivals in local earthquake data using an artificial neural network: Geophys. J. Internat., 120, 758–774.Google Scholar
  6. Dai, H. C., and MacBeth, C., 1997a, The application of back-propagation neural network to automatic picking seismic arrivals from single-component recordings: J. Geophys. Res., 102, 15105–15113.CrossRefGoogle Scholar
  7. Dai, H. C., and MacBeth, C., 1997b, Application of back-propagation neural network to identification of seismic arrival types: Phys. Earth Plan. Int., 101, 177–188.Google Scholar
  8. Fausett, L. V., 1994, Fundamentals of neural networks: architectures, algorithms and applications: Prentice-Hall, Inc. Haykin, S., 1994, Neural networks, a comprehensive foundation: Macmillan College Publ. Co.Google Scholar
  9. Lomax, A J., and Michelin, A, 1988, The use of spherical coordinates in the interpretation of seismograms: Geophys. J. Internat, 93, 405–412.Google Scholar
  10. Lovell, J. H.,1989, Source parameters of a microearthquake swann in Turkey: M.Sc. thesis, Univ. of Edinburgh, Scotland. Rumelhart, D., Hinton, G. E., and Williams, R J., 1986, Learning representations by back-propagating errors: Nature, 323, 533–536.Google Scholar
  11. Samson, J. C., 1977, Matrix and Stokes vector representations of detectors for polarized waveforms: theory, with some applications to teleseismic waves: Geophys. J. Roy. Astr. Soc., 51, 433–452.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2003

Authors and Affiliations

  • Hengchang Dai
    • 1
  1. 1.British Geological SurveyEdinburghScotland

Personalised recommendations