A Novel Time-Frequency Similarity Method for P-Wave First-Motion Polarity Detection

Area of Science:

• Seismology
• Earthquake Science
• Signal Processing

Background:

• P-wave first-motion polarity is crucial for earthquake focal mechanism determination.
• Seismic noise complicates accurate detection of P-wave onsets and polarities.
• Existing methods like normalized cross-correlation (NCC) are sensitive to noise.

Purpose of the Study:

• To develop a noise-resistant method for detecting P-wave first-motion polarity.
• To improve the accuracy of earthquake location and focal mechanism inversion.

Main Methods:

• Proposed a novel time-frequency similarity coefficient (TFSC) method based on the normal time-frequency transform (NTFT).
• Calculated relative delays and similarity coefficients using the real part of NTFT coefficients.
• Validated the TFSC method using synthetic and real earthquake data.

Main Results:

• The TFSC method demonstrated superior robustness and reliability compared to conventional NCC.
• Effective detection of P-wave polarity was achieved even without filtering or preprocessing.
• The method accurately identified relative delays and similarity coefficients.

Conclusions:

• The TFSC method offers a significant advancement for accurate P-wave polarity detection in noisy seismic environments.
• This technique shows strong potential for practical applications in seismology.
• Provides a novel perspective for seismic signal analysis.