Empirical correlations between an FAS non-ergodic ground motion model and a GIT derived model for Central Italy
- 11 November 2022
- journal article
- research article
- Published by Oxford University Press (OUP) in Geophysical Journal International
- Vol. 233 (1) , 51-68
Abstract
In this study we investigate the correlation between the residuals of a neGMM (non-ergodic Ground Motion Model) and the physics-based parameters obtained using a non-parametric GIT approach (Generalized Inversion Technique) to lay the groundwork for the implementation of an ad-hoc FAS (Fourier Amplitude Spectra) neGMM for the Central Italy region. This region is particularly suitable for data-driven methodologies as those applied in this work because of the large amount of available data due to the recent multiple mainshock-aftershock sequences occurred in this area. Both neGMM and GIT models are developed for Fourier spectra in the frequency range between 0.5 and 25 Hz and using the same reference sites. The comparison of the non-ergodic terms with the source, path and site spectral parameters provides interesting results. First, we find a strong correlation between the source parameters, stress drop Δσ and decay ksource, and the source neGMM corrective terms (the combination of the between-event δBe and the location-to-location terms δL2L). This correlation is frequency dependent and, at high frequency, is remarkably positive for Δσ and negative for ksource. Concerning the attenuation terms, the path-to-path residuals (δP2P) are clearly associated with the deviations from the regional Q estimates obtained from GIT analysis. This indicates that the neGMM properly captures the properties of the anelastic attenuation and that the corrective terms δP2P can be used to account for differences in travel paths across different crustal domains. Finally, adopting the same reference sites for neGMM and non-parametric GIT, we observe that the systematic site terms (δS2Ss) and the GIT-derived amplification functions are in good agreement. The next step for an appropriate modeling is to identify the physical parameters (for example VS,30 and k0) describing the empirical amplification curves to be introduced as explanatory variables in the ground motion model.Keywords
Funding Information
- Istituto Nazionale di Geofisica e Vulcanologia
- Catholic University of Pusan (D53J19000170001)
- Ministero dell’Istruzione, dell’Università e della Ricerca
- Dipartimento della Protezione Civile, Presidenza del Consiglio dei Ministri
This publication has 50 references indexed in Scilit:
- A Methodology to Improve Ground-Motion Prediction Equations by Including Path CorrectionsBulletin of the Seismological Society of America, 2011
- Spectral Analysis of K-NET and KiK-net Data in Japan, Part II: On Attenuation Characteristics, Source Spectra, and Site Response of Borehole and Surface StationsBulletin of the Seismological Society of America, 2011
- The Variability of Ground-Motion Prediction Models and Its ComponentsSeismological Research Letters, 2010
- The 2009 L'Aquila (central Italy) MW6.3 earthquake: Main shock and aftershocksGeophysical Research Letters, 2009
- Strong evidence for non‐similar earthquake source scaling in central ItalyGeophysical Research Letters, 2008
- High‐stress strike‐slip faults in the Apennines: An example from the 2002 San Giuliano earthquakes (southern Italy)Geophysical Research Letters, 2008
- Static and Dynamic Scaling Relations for Earthquakes and Their Implications for Rupture Speed and Stress DropBulletin of the Seismological Society of America, 2004
- Estimating regression models with unknown break‐pointsStatistics in Medicine, 2003
- Probabilistic Seismic Hazard Analysis without the Ergodic AssumptionSeismological Research Letters, 1999
- Seismic sources, fault plane studies and tectonicsEos, 1971