Seismic imaging of a fluid storage in the actively extending Apennine mountain belt, southern Italy

Abstract

A picture of the upper crustal structure of the Irpinia active faults system in southern Italy was obtained by combining new geological evidences, lithological properties, and microseismicity distribution. P and S wave velocity models indicate high V P /V S and low V P × V S values, suggesting fluid accumulation within ã 15 km wide rock volume where intense microseismicity is located. The 1980 Irpinia, M s 6.9, earthquake nucleated within the same fault-bounded volume. We suggest that concentration of background seismicity is mainly controlled by high pore fluid pressure. Its increase in fluid-filled cracks around major faults leads to earthquakes' nucleation. Seismic pumping along major faults carries fluids through the conduit system represented by the intensely fractured damage zone. Conversely, the cross-fault barrier behavior of the low-permeability fault core leads to pore fluid pressures building up within the fault-bounded block, thus producing a positive feedback triggering earthquake nucleation within the volume, which behaves as an "earthquake reservoir."

From velocity and attenuation tomography to rock physical modeling: Inferences on fluid‐driven earthquake processes at the Irpinia fault system in southern Italy

“…The strong lateral SW-NE variation of Q s coincides with the NW-SE trending lateral velocity discontinuity previously revealed by Amato et al [1992] and recently confirmed by Amoroso et al [2014] and Improta et al [2014], affecting the shallow upper crustal volume across the Irpinia earthquake fault zone. This discontinuity can be associated with the contrast between the thick cover of low-velocity, high-attenuation Miocene basin sediments to the NE and the high-velocity, low-attenuation carbonate outcrops of the Apennine platform to the SW [Ascione et al, 2013].…”

“…We selected the best located events in the 3-D velocity model of Amoroso et al [2014] among those that occurred from August 2005 to April 2011 and were recorded by the stations operated by the University of Naples ISNet (Irpinia Seismic Network, AMRA) and INGV (Istituto Nazionale di Geofisica e Vulcanologia) ( Figure 1). We thereby analyzed a data set of 4801 t * P and 1833 t * S relative to 670 earthquakes with local magnitude 0.1 ≤ M L ≤ 3.2.…”

“…In order to perform the 3-D attenuation tomography, we used a modified version of the code originally used by Amoroso et al [2014] to perform the delay time inversion. We inverted the t * data for Q following a multiscale strategy [Chiao and Kuo, 2001;Zollo et al, 2002] whereby several optimization runs are performed by progressively increasing the density of grid points describing the attenuation model (more details in the supporting information) [Latorre et al, 2004].…”

“…Another core taken at 6 km depth in alternating dolostones and anhydrites provided a porosity range of 0.70-1.37%. The total porosity at the scale considered for our modeling can be significantly higher due to the likely presence of dense networks of fractures [Amoroso et al, 2014]. The modal value of average porosity at a depth of 5.25-5.75 km in the huge data set of Ehrenberg and Nadeau [2005] is 6.2%.…”

“…These quantitative images complement the already existing velocity images inferred from local earthquake tomography [Amoroso et al, 2014]. We performed an attenuation tomographic inversion, which provides 3-D images of the anelastic attenuation properties in terms of body wave quality factors (Q P and Q S ).…”