We perform the finite‐extent fault inversion of the three main events of the 2016 Central Italy seismic sequence using near‐source strong motion records. We demonstrate that both earthquake nucleation and rupture propagation were controlled by segmentation of the (N)NW‐(S)SE trending Quaternary normal faults. The first shock of the sequence (24 August, Mw 6.0) ruptured at the relay zone between the Laga Mts (LF) and the Cordone del Vettore (CVF) normal faults. The second shock (26 October, Mw 5.9) nucleated at a minor relay zone within the Mt. Vettore‐Mt. Bove fault (VBF), while the third and largest one (30 October, Mw 6.5) initiated at the relay zone between the VBF and CVF, triggering the multiple rupture of the VBF, CVF, and probably LF. We show that this latter relay zone corresponds to the deeper, high‐angle, fault zone of the Sibillini Mts cross structure, a thrust‐ramp inherited from the Miocene‐Pliocene contractional phase of the Apennines. This structure acted as a barrier to rupture propagation of the first two events thus defining an area of large stress concentration until it acted as the initiator of the rupture originating the largest Mw 6.5 event that crossed the barrier itself. We suggest that the “young” CVF have started to cut through the barrier acting as a soft‐linkage between the two long‐lived LF and VBF. The evidence that coseismic cumulative slip shows a maximum at the CVF, provided by both slip inversion and original surface rupture data, suggests that the CVF is growing faster than the adjacent faults.

International audienc

In 2016 Central Italy was struck by a sequence of three normal-faulting earthquakes with moment magnitude (M w ) larger than 6. The M w 6.2 Amatrice event (24 August) was the first one, causing building collapse and about 300 casualties. The event was recorded by a uniquely dense network of seismic stations. Here we perform its dynamic source inversion to infer the fault friction parameters and stress conditions that controlled the earthquake rupture. We consider a linear slip-weakening friction law with spatially variable parameters along the fault. The inversion uses a novel Bayesian framework developed in our companion paper, which combines efficient finite-difference dynamic rupture simulations and the Parallel Tempering Monte Carlo algorithm to sample the posterior probability density function. The main advantage of such formulation is that by subsequent analysis of the posterior samples we can infer stable features of the result and their uncertainty. The inversion results in a million of visited models. The preferred model ensemble reveals intriguing dynamic features. The rupture exhibits a slow and irregular nucleation followed by bilateral rupture propagation through two asperities, accelerating toward the heavily damaged city of Amatrice. The stress drop reaches locally 10-15 MPa, with slip-weighted mean of 4-4.5 MPa. The friction drop ranges from 0.1 to 0.4. The characteristic slip-weakening distance is the most heterogeneously distributed dynamic parameter, with values of 0.2-0.8 m. The radiation efficiency was rather low, 0.2, suggesting that approximately 80% of the total available energy was spent in the fracture process, while just 20% was radiated by seismic waves.

No abstract

Earthquake slip inversions aiming to retrieve kinematic rupture characteristics typically assume 1-D velocity models and a flat Earth surface. However, heterogeneous nature of the crust and presence of rough topography lead to seismic scattering and other wave propagation phenomena, introducing complex 3-D effects on ground motions. Here we investigate how the use of imprecise Green's functions-achieved by including 3-D velocity perturbations and topography-affect slip-inversion results. We create sets of synthetic seismograms, including 3-D heterogeneous Earth structure and topography, and then invert these synthetics using Green's functions computed for a horizontally layered 1-D Earth model. We apply a linear inversion, regularized by smoothing and positivity constraint, and examine in detail how smoothing effects perturb the solution. Among others, our tests and resolution analyses demonstrate how imprecise Green's functions introduce artificial slip rate multiples especially at shallow depths and that the timing of the peak slip rate is hardly affected by the chosen smoothing. The investigation is extended to recordings of the 2009 Mw6.3 L'Aquila earthquake, considering both strong motion and high-rate GPS stations. We interpret the inversion results taking into account the lessons learned from the synthetic tests. The retrieved slip model resembles previously published solutions using geodetic data, showing a large-slip asperity southeast of the hypocenter. In agreement with other studies, we find evidence for fast but subshear rupture propagation in updip direction, followed by a delayed propagation along strike. We conjecture that rupture was partially inhibited by a deep localized velocity-strengthening patch that subsequently experienced afterslip.