New insights into fault activation and stress transfer between en echelon thrusts: The 2012 Emilia, Northern Italy, earthquake sequence

Abstract

Here we present the results of the inversion of a new geodetic data set covering the 2012 Emilia seismic sequence and the following 1 year of postseismic deformation. Modeling of the geodetic data together with the use of a catalog of 3‐D relocated aftershocks allows us to constrain the rupture geometries and the coseismic and postseismic slip distributions for the two main events (Mw 6.1 and 6.0) of the sequence and to explore how these thrust events have interacted with each other. Dislocation modeling reveals that the first event ruptured a slip patch located in the center of the Middle Ferrara thrust with up to 1 m of reverse slip. The modeling of the second event, located about 15 km to the southwest, indicates a main patch with up to 60 cm of slip initiated in the deeper and flatter portion of the Mirandola thrust and progressively propagated postseismically toward the top section of the rupture plane, where most of the aftershocks and afterslip occurred. Our results also indicate that between the two main events, a third thrust segment was activated releasing a pulse of aseismic slip equivalent to a Mw 5.8 event. Coulomb stress changes suggest that the aseismic event was likely triggered by the preceding main shock and that the aseismic slip event probably brought the second fault closer to failure. Our findings show significant correlations between static stress changes and seismicity and suggest that stress interaction between earthquakes plays a significant role among continental en echelon thrusts.

Poroelasticity and Fluid Flow Modeling for the 2012 Emilia-Romagna Earthquakes: Hints from GPS and InSAR Data

“…The temporal function of the first independent component (V1, Figure 2(a)) shows a decay with time after the May 20 mainshock. The corresponding spatial distribution (Figure 2(c)) agrees with what we would expect from afterslip occurring on the two faults involved in the seismic sequence, as described by Cheloni et al [6]. Fitting the temporal dependence of IC1 with an exponential curve, we estimate a decay time of τ d = 103 days (Figure 2(b)).…”

“…We analyzed the GPS position time series after the May 20 mainshock with the vbICA method, finding three independent signals (Figure 2): the largest one, IC1, is related to afterslip and lasted for about 100-160-day time scale, close to the 180-day estimate of Cheloni et al [6]; the second one, IC2, interests a wider region and apparently it is not related to the earthquake sequence, while the IC3 is strongly localized in near field and it is characterized by a 10-day rising time.…”

“…Figure 8(a) shows the co-and postseismic displacement evolution modeled along the line of sight (LOS) at points 1 and 2 (see Figure 1). In particular, observed LOS displacements (orange and green dots) are InSAR observations taken, respectively, from Albano et al, [18] and Cheloni et al, [6]. The major modeled poroelastic response leads to a LOS increase of about 1 cm at point 1 (inset of Figure 8) and 0.5 cm at point 2, and it occurs within the first 10 days after the earthquake at both points following the relatively fast poroelastic relaxation of the shallow layers located at less than 3.3 km depth (green and purple lines in Figure 8(a)).…”

“…Both the aftershock distribution [1] and the InSAR deformation measurements [3,6,7] clearly suggest that the overall seismic sequence involved two different, partially overlapping, fault planes: the May 20 mainshock occurred on the Ferrara fault (the easternmost one), while the May 29 mainshock occurred on the Mirandola fault (the westernmost one). Both faults are mainly east-west oriented, dipping towards south.…”

“…Both faults are mainly east-west oriented, dipping towards south. Several fault geometries, based on different geodetic data inversions, are proposed in the literature (e.g., [3,6,8]). According to the model proposed by Pezzo et al [3] and by Nespoli et al [7], most of the slip of the May 20 event occurred along the Ferrara fault plane, even though some minor, aseismic slip on the Mirandola fault before the May 27 event is required to explain a tail-shaped deformation detected by the SAR satellites in the western part of the interferogram (Figure 2b in [7]) within the time window May12-27, 2012.…”