← Back to Sessions

On the Variability Discrepancy Between PGA and Spectral Stress Drop: Insight From Double-Corner-Frequency Spectra

Description: 

The stress drop estimated from the corner frequency (spectral stress drop) is considered an important parameter in ground motion prediction. Cotton et al. (2013) found that the observed variability of the spectral stress drop is significantly larger than expected from the observed PGA variability. In this study, we conducted spectral ratio analysis for 34 crustal earthquakes with M_w 5.0–7.1 in Japan to resolve this variability paradox. First, we applied the standard spectral ratio method, which assumes the single-corner-frequency (SCF) spectrum model, to estimate the spectral stress drop Δσ_fc. Second, we applied the two-stage spectral ratio method developed in this study. This method enables us to obtain the double-corner-frequency (DCF) spectra at broadband frequencies. We determined the corner frequency f_ch by fitting the SCF model to the DCF model at low and high frequencies. The spectral stress drop calculated using f_ch is called the stress parameter Δσ_fch, which characterizes the intensity of the high-frequency radiation.

The standard deviation of lnΔσ_fc is 1.0, as high as the ones found in previous corner frequency studies. In contrast, the standard deviation of lnΔσ_fch is 0.55, which predicts the standard deviation of lnPGA of 0.44, comparable to the between-event variability of PGA observed in Japan. The corner frequency obtained from the standard spectral ratio method crudely corresponds to the lower corner frequency of the DCF model, f_ca. Thus, the variability discrepancy between Δσ_fc and PGA could mainly arise from the difference in the variability of f_ca and f_ch of the DCF spectrum. The DCF model successfully explains the spectra for all events and has the potential to explain both the observed Δσ_fc and PGA variabilities. Finally, we found that the deviation of the observed spectra from the SCF model becomes significant on average as the magnitude increases. Implementing the DCF model can be important for predicting ground motions.

Session: Understanding and Quantifying the Variability in Earthquake Source Parameter Measurements [Poster Session]

Type: Poster

Date: 5/3/2024

Presentation Time: 08:00 AM (local time)

Presenting Author: Shota

Student Presenter: No

Invited Presentation: 

Authors