Constraining families of dynamic models using geological, geodetic and strong ground motion data: the Mw 6.5, October 30th, 2016, Norcia earthquake, Italy

Abstract

The 2016 Central Italy earthquake sequence is characterized by remarkable rupture complexity, including highly heterogeneous slip across multiple faults in an extensional tectonic regime. The dense coverage and high quality of geodetic and seismic data allow to image intriguing details of the rupture kinematics of the largest earthquake of the sequence, the Mw 6.5 October 30th, 2016 Norcia earthquake, such as an energetically weak nucleation phase. Several kinematic models suggest multiple fault planes rupturing simultaneously, however, the mechanical viability of such models is not guaranteed.Using 3D dynamic rupture and seismic wave propagation simulations accounting for two fault planes, we constrain 'families' of spontaneous dynamic models informed by a high-resolution kinematic rupture model of the earthquake. These families differ in their parameterization of initial heterogeneous shear stress and strength in the framework of linear slip weakening friction.First, we dynamically validate the kinematically inferred two-fault geometry and rake inferences with models based on only depth-dependent stress and constant friction coefficients. Then, more complex models with spatially heterogeneous dynamic parameters allow us to retrieve slip distributions similar to the target kinematic model and yield good agreement with seismic and geodetic observations. We discuss the consistency of the assumed constant or heterogeneous static and dynamic friction coefficients with mechanical properties of rocks at 3-10 km depth characterizing the Italian Central Apennines and their local geological and lithological implications. We suggest that suites of well-fitting dynamic rupture models belonging to the same family generally exist and can be derived by exploiting the trade-offs between dynamic parameters.Our approach will be applicable to validate the viability of kinematic models and classify spontaneous dynamic rupture scenarios that match seismic and geodetic observations at the same time as geological constraints.

Community‐Driven Code Comparisons for Three‐Dimensional Dynamic Modeling of Sequences of Earthquakes and Aseismic Slip

“…In earthquake science, models of earthquake source processes are aimed at capturing dynamic earthquake ruptures from seconds to minutes and slow slip processes subject to short‐term anthropogenic or environmental forcing, or tectonic loading over timescales of years and longer. For individual earthquakes, dynamic rupture simulations have emerged as powerful tools to reveal the influence of fault structure, geometry, constitutive laws, and prestress on earthquake rupture propagation and associated ground motion (e.g., Andrews, 1976a, 1976b; Ben‐Zion, 2001; Bhat et al., 2007; Bizzarri & Cocco, 2003, 2006; Day, 1982; Das & Aki, 1977; Duan & Day, 2008; Dunham et al., 2011b, 2011a; Gabriel et al., 2012; Harris et al., 1991, 2021; Kozdon & Dunham, 2013; Lozos et al., 2011; Ma & Beroza, 2008; Madariaga et al., 1998; Mikumo & Miyatake, 1978, 1993; Nielsen et al., 2000; Olsen et al., 1997; Ripperger et al., 2007; Z. Shi & Day, 2013; Tinti et al., 2021; Wollherr et al., 2019; Xu et al., 2015). These simulations are limited to single‐event scenarios and subject to imposed artificial prestress conditions and ad hoc nucleation procedures.…”

The Dynamics of Unlikely Slip: 3D Modeling of Low‐Angle Normal Fault Rupture at the Mai'iu Fault, Papua New Guinea

“…Only a small portion of the fault needs to reach failure to nucleate a rupture, and slip can propagate into regions of velocity‐ or slip‐strengthening frictional behavior (e.g., Kaneko et al., 2010; Thomas et al., 2014) depending on the patterns of static and dynamic stress transfer arising from the initial fault prestress and frictional properties (e.g., Ariyoshi et al., 2009; Cochard & Madariaga, 1996; Rundle et al., 1984). However, few dynamic rupture models exist for normal fault earthquakes and these are restricted to planar faults (Oglesby et al., 1998, 2000, 2008; Aochi, 2018; Aochi & Twardzik, 2020; Gallovič et al., 2019; Tinti et al., 2021); to the best of our knowledge dynamic rupture models have not been used to explore conditions allowing LANF rupture.…”

Conversations between earthquakes: Dynamics and delays of the 2019 Ridgecrest rupture sequence

“…We use a systematic approach that allows us to constrain the orientation of all principal stresses, the magnitudes of deviatoric stresses (26), and that extends the Mohr-Coulomb theory of frictional failure with dynamic parameters while reducing the large parameter space common in dynamic rupture modeling (53). We assume that the ambient prestress is 3D heterogeneous and always Andersonian (54), i.e.…”