Full Envelope Templates for Low Magnitude Discrimination and Yield Estimation at Local and Regional Distances

Abstract

Monitoring seismologists have successfully used seismic coda for event discrimination and yield estimation for over a decade. In practice seismologists typically analyze long-duration, S-coda signals with high signal-to-noise ratios (SNR) at regional and teleseismic distances, since the single back-scattering model reasonably predicts decay of the late coda. However, seismic monitoring requirements are shifting towards smaller, locally recorded events that exhibit low SNR and short signal lengths. To be successful at characterizing events recorded at local distances, we must utilize the direct-phase arrivals, as well as the earlier part of the coda, which is dominated by multiple forward scattering. To remedy this problem, we have developed a new hybrid method known as full-waveform envelope template matching to improve predicted envelope fits over the entire waveform and account for direct-wave and early coda complexity. We accomplish this by including a multiple forward-scattering approximation in the envelope modeling of the early coda. The new hybrid envelope templates are designed to fit local and regional full waveforms and produce low-variance amplitude estimates, which will improve yield estimation and discrimination between earthquakes and explosions. To demonstrate the new technique, we applied our full-waveform envelope template-matching method to the six known North Korean (DPRK) underground nuclear tests. We successfully discriminated the event types and estimated the yield for all six nuclear tests.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 19, 2020
Accession Number
AD1111180

Entities

People

  • Seung-Hoon Yoo

Organizations

  • Applied Research Associates (United States)

Tags

Communities of Interest

  • Energy and Power Technologies
  • Sensors
  • Space

DTIC Thesaurus Topics

  • Algorithms
  • Computational Science
  • Databases
  • Detection
  • Explosions
  • Forward Scattering
  • Frequency Bands
  • Information Science
  • Machine Learning
  • Markov Chains
  • Monte Carlo Method
  • Radiative Transfer
  • Sampling
  • Scattering
  • Supervised Machine Learning
  • Three Dimensional
  • Two Dimensional

Readers

  • Seismology
  • Systems Analysis and Design