On the retrieval of attenuation and site amplifications from ambient noise on linear arrays: Further numerical simulations

Research output: Contribution to journalArticlepeer-review

Abstract

While it is clear that ambient noise correlations contain information on seismic attenuation, retrieval of that information requires disentangling the influences of site amplification factors and anisotropic noise intensity. Recent work has argued that recognition that ambient noise intensity is governed by a radiative transfer equation will facilitate that retrieval. This sufficiently constrains the noise field to permit amplitudes of noise correlation waveforms to be fit to models of spatially varying attenuation, noise intensity and site amplification factors. Numerical simulations, for the case of a uniformly spaced linear array of receivers, homogeneous attenuation and wave speed and mild directionality to the noise field, were shown earlier to be consistent with the assertion. Here, the numerical simulations are extended to a broader class of systems. Accurate retrieval is demonstrated using a priori error estimates and weighted least-squares fitting. Attenuations and site factors are retrieved in systems with spatially varying attenuation, highly directional noise intensities, irregular line arrays and poor signal-to-noise ratios. Examination of a variety of systems illustrates some of the range of applicability and limits of the approach.

Original languageEnglish (US)
Pages (from-to)1644-1657
Number of pages14
JournalGeophysical Journal International
Volume193
Issue number3
DOIs
StatePublished - May 2013

Keywords

  • Seismic attenuation
  • Seismic monitoring and test-ban treaty verification
  • Seismic tomography
  • Site effects
  • Statistical seismology
  • Wave propagation

ASJC Scopus subject areas

  • Geophysics
  • Geochemistry and Petrology

Fingerprint

Dive into the research topics of 'On the retrieval of attenuation and site amplifications from ambient noise on linear arrays: Further numerical simulations'. Together they form a unique fingerprint.

Cite this