Imprint of Rupture Directivity From Ground Motions of the 24 August 2016 M_w6.2 Central Italy Earthquake

Abstract

An Mw6.2 earthquake occurred in Central Italy on 24 August 2016. The objective of this study was to reveal the imprint of rupture directivity using the strong motion recordings. The strong motion stations were separated into two groups: southeast (SE) and northwest (NW). The effects of rupture directivity on the peak ground acceleration (PGA), peak ground velocity (PGV), and pseudo spectral acceleration (PSA) were investigated. The observed values of these parameters were compared with predicted values derived from ground motion prediction equations. The results showed that the residuals between the observed and predicted PGAs, PGVs, and PSAs at periods of T < 1 s were correlated significantly with azimuth angle and generally larger in the NW sector, reflecting that the observed PGAs, PGVs, and short‐period PSAs in the NW sector were generally larger than observed in the SE sector. These phenomena are accordant with the theoretical law that the rupture directivity causes higher amplitudes in the forward direction compared with the backward direction. Finally, selected source rupture parameters were inverted using PGAs and PGVs. This revealed that the rupture was predominantly unilateral rupture, the major rupture was likely at an azimuth of ~360°, and the length of the major rupture was proportional to 70%–100% of the total ruptured fault, confirming that rupture directivity caused the differences in the ground motions observed in the SE and NW sectors.

Faults Geometry and the Role of Fluids in the 2016–2017 Central Italy Seismic Sequence

“…The two largest mainshocks, the 24 August Mw = 6.1 Amatrice and the 30 October Mw = 6.5 Norcia, ruptured adjacent normal faults with ancillary structures (Cheloni et al, , Livio et al, ) and with overlapping slip on shallow faults surfacing on the western slope of the Vettore mountain range (Galli et al, ; Pucci et al, ). While some authors have modeled the 24 August shock, as continuous slip on a single fault (Huang et al, ), others proposed bilateral rupture with hypocenter located in between two separate faults (Liu et al, ; Ren et al, ; Tinti et al, ). The relatively young normal faults accommodate the 2–3 mm/year of extension across the mountain range.…”

Breakdown of Earthquake Self‐Similar Scaling and Source Rupture Directivity in the 2016–2017 Central Italy Seismic Sequence

“…C d γ indicate greater significance of the azimuthal variation of the residuals. As shown in Figure a, the v r / β values generally show a pronounced trend of decrease with increasing γ , indicating the vital impact of γ on the retrieved v r / β parameters, which has been widely verified (Pacor, Gallovič, et al, ; Ren, Wang, & Wen, ). In addition, larger v r / β values appear to result in the larger values of Max.…”

“…The kinematic source models for the M L 6.0 mainshock and two subsequent major earthquakes ( M L 5.9 and 6.1) were inverted using near‐source strong‐motion data, teleseismic waveforms, and geodetic data to reveal the slip heterogeneity, rupture directivity, and seismogenic mechanism (Chiaraluce et al, ; Liu et al, ; Pizzi et al, ; Scognamiglio et al, ; Tinti et al, ). According to analyses of the spatial distributions of various ground‐motion intensity measures, Ren et al () both confirmed the asymmetrical bilateral rupture propagation of the M L 6.0 mainshock and explained the directivity effects, especially on short‐period ground motions. Similarly, Convertito et al () inferred the horizontal source rupture directivity of the M L 6.0 mainshock by analyzing the peak ground velocity, and they suggested a contribution from the dynamic strain, enhanced by rupture directivity, to the occurrence of the ongoing aftershocks.…”

“…However, ambiguity remains regarding the determination of the optimal value of γ to reflect source rupture directivity (Ruiz et al, ). In practical applications, γ = 1 (Bernard et al, ; Boatwright, ; Wen et al, ) and γ = 0.5 (Gallovič, ; Ren, Wang, & Wen, ) have been validated in explanations of the directivity effects on either observed ground motions or apparent source spectra produced by asymmetrical source rupture. Gallovič () showed the best fit with observed strong‐motion data near a fault for frequencies >1 Hz is provided by a model with γ = 0.5.…”

“…Gallovič () showed the best fit with observed strong‐motion data near a fault for frequencies >1 Hz is provided by a model with γ = 0.5. Meanwhile, Ren, Wang, and Wen () discussed the strong influence of uncertainty regarding γ on the estimation of v r / β and k , and they proposed that a model with γ = 0.5 produces reliable estimations of v r / β .…”