Analysis of background microseismicity for crustal velocity model, fault delineation and regional stress direction in Southern Apennines, Italy

The detailed analysis of the background microseismicity (M< 3) of the Southern Apennines (Italy) is used to identify the present active faults and stress field acting in a structurally complex area characterized by high seismic potential. The used refined techniques for the analysis and the high data quality allow to obtain a sharper picture of the spatial distribution of microseismicity and to determine the fine-scale geometry and extent of existing faults. We analysed a microearthquake data set consisting of 980 events with moment magnitude ranging between 0.9 and 3.1 occurred from August 2005 to April 2010 by integrating the data recorded by ISNet (AMRA Scarl) and the National Seismic Network managed by INGV. The first P- and S-wave arrival times have been manually picked on the raw waveforms for a total of 8663 P and 4358 S arrival times. The 3D crustal velocity model is retrieved using a delay travel times linearized, iterative tomographic algorithm. In the tomographic inversion we used as starting velocity model the minimum 1D model obtained with the same dataset using VELEST code. In this model all the events have been located using the probabilistic, non-linear, global-search earthquake location method (NonLinLoc code). The refined seismicity location well delineates a sequence of sub-parallel, NW-SE striking normal faults along the Apenninic chain and an approximately E-W oriented, strike-slip fault, transversely cutting the chain. The found fault trending and extent as suggested by the imaged earthquake locations indicate that low magnitude seismicity is likely generated along the major fault segments activated during the most recent earthquakes occurred in the Irpinia region on 23 November 1980 and between 1990 and 1991 in the Potenza region. This evidence suggests that major fault segments are still active today thirty years after the mainshock occurrences. In order to study the relation of this complex fault system with the stress field acting in the study area we performed a stress inversion from microearthquakes. We used the algorithm developed by Rivera and Cisternas (1990) that allows for the estimation of the orientation and shape of the stress tensor directly using the polarities of the P arrivals and the take-off angles. Moreover, the errors on the principal stress axes direction are estimated by computing the 95% confidence regions with a bootstrap procedure. Results show a dominant extensional regional stress field characterized by a nearly horizontal NE–SW minimum compressive stress axis (σ3), and a nearly vertical maximum compressive stress axis (σ1). These findings are consistent with the results obtained from the analysis of other surface geological, breakout and seismic data. Our study suggests that the existence of a unique, dominant anti-Apenninic extensional regional stress can explain the microearthquake generation along both the NW-SE striking normal faults and the EW oriented fault, with a dominant dextral strike-slip motion.