Sensitivity analysis of moment magnitude (Mw) spectral fitting parameters based on orthogonal design: using aftershocks of the 2019 Mw 6.5 Ambon, Indonesia earthquake.

Autor: Kusumawati, Dian, Sahara, David P., Nugraha, Andri D., Puspito, Nanang T., Widiyantoro, Sri, Zulfakriza, Zulfakriza, Rosalia, Shindy, Shiddiqi, Hasbi Ash, Ardianto, Ardianto, Baskara, Aria Widhi, Rusdin, Andi Azhar, Wibowo, Adi, Heliani, Leni S.
Předmět:
Zdroj: Bulletin of Earthquake Engineering; Mar2023, Vol. 21 Issue 4, p1793-1819, 27p
Abstrakt: Spectral analysis method is commonly used in estimating moment magnitude for small to medium size earthquake. Most moment magnitude studies have been focused on the consistency of the average values with other methods. However, little attention is put into the study of moment magnitude sensitivity to its input parameter. In this study we attempted to analyse the sensitivity of moment magnitude to spectral fitting parameters for P and S-waves (Mw,PS) and P-waves (Mw,P) by using Taguchi Orthogonal Array (OA) design. Utilizing the 2019 Ambon aftershocks catalogue, we analysed the impact of uncertainty in the radiation pattern, time window length, hypocenter location, 1D velocity, and 1D density model on the magnitude. We considered 10% uncertainty for all the parameters; except for hypocenter location parameter where we considered ± 4 km deviation. We observed that the 1D velocity model had the greatest percentage contribution on magnitudes variation (Mw,PS and Mw,P), up to 65%. Compared to the hypocenter location parameter, the magnitudes are less sensitive to changes in the 1D density model and radiation pattern as these two parameters gave contribution to magnitude variation consistently at ≈ 7%. Through spatial analysis of earthquake clusters, we observed that event's depth coupled with station location have significant effect to magnitude changes. Care has to be taken when events are located in the 1D velocity boundary or near station, as these factors could increase the magnitude variation. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index