5:15 PM - 7:15 PM
[SSS11-P24] On the correlation between rupture area residuals and average slip residuals of the source inversion models
Keywords:Source inversion, Scaling law, Strong motion simulation
Total rupture area (S) and average slip (Dave) in the earthquake source are initial parameters for the prediction of strong ground motion. Several scaling relationships are proposed to relate these parameters with seismic moment (M0), which is another important initial parameter. In this study we used database of rupture models from the source inversion results for crustal earthquakes, compiled by Nagashima et al. (2021), and analyzed residuals between data and average 3-stage model. Data and the 3-stage model for S(M0) and Dave(M0) are shown in Fig. 1 and 2. Fig. 3 shows plot of residuals of the S-value vs residuals of the Dave-value.
There is a negative correlation between residuals. This negative correlation is a natural outcome of the trade-off coming from M0 definition: M0 = m*S*Dave, where m is rigidity. To keep M0-value, Dave values of data points with the same M0-values should be larger for data points with smaller S, and oppositely Dave values should be smaller for data points with larger S. Similar correlation was found by analysis of the validated dynamic ruptures from the earthquake cycle simulations (Galvez et al., 2021). Average stress drop of dynamic ruptures is larger for ruptures with smaller S and larger Dave, and vice versa.
There are several important issues that need to be studied in the future work. Does the S vs Dave residuals correlation of observed earthquakes also correlate with stress drop or not. Preliminary analysis of stress drop using crack model suppose existence of such correlation. Does the ground motions residuals (between observed data and estimated data based on scaling laws) also correlate with S and Dave residuals or not. Existence of such correlation could support the stress drop mechanism of correlation. Finally, are there any specific tectonic features of faults that could be associated with correlations above or not. Identification of such features and their extrapolation into faults that are not ruptured yet, could help us reducing uncertainties of the strong ground motion prediction.
References.
Galvez et al. (2021), Bull.Seismol.Soc.Am., doi: 10.1785/0120210104.
Nagashima et al., 2021, JAEE, 21(5) pp. 140-160.
There is a negative correlation between residuals. This negative correlation is a natural outcome of the trade-off coming from M0 definition: M0 = m*S*Dave, where m is rigidity. To keep M0-value, Dave values of data points with the same M0-values should be larger for data points with smaller S, and oppositely Dave values should be smaller for data points with larger S. Similar correlation was found by analysis of the validated dynamic ruptures from the earthquake cycle simulations (Galvez et al., 2021). Average stress drop of dynamic ruptures is larger for ruptures with smaller S and larger Dave, and vice versa.
There are several important issues that need to be studied in the future work. Does the S vs Dave residuals correlation of observed earthquakes also correlate with stress drop or not. Preliminary analysis of stress drop using crack model suppose existence of such correlation. Does the ground motions residuals (between observed data and estimated data based on scaling laws) also correlate with S and Dave residuals or not. Existence of such correlation could support the stress drop mechanism of correlation. Finally, are there any specific tectonic features of faults that could be associated with correlations above or not. Identification of such features and their extrapolation into faults that are not ruptured yet, could help us reducing uncertainties of the strong ground motion prediction.
References.
Galvez et al. (2021), Bull.Seismol.Soc.Am., doi: 10.1785/0120210104.
Nagashima et al., 2021, JAEE, 21(5) pp. 140-160.