Three-dimensional crustal image of Arraiolos aftershock sequence, earthquake of M=4.9, in Alentejo region, Portugal.

Abstract

This work presents the results obtained after the Arraiolos aftershock sequence tomographic inversion in terms of velocity th distribution. The study deals with the 4.9 ML magnitude earthquake (Mw4.3, Vales et al., 2018) which occurred on January 15 , 2018 at 11:51 UTC in Aldeia da Serra in the Northeast of Arraiolos (Alentejo, Portugal). The hypocentral location, determined by the Instituto Português do Mar e da Atmosfera (IPMA), has coordinates 38.79 N, 7.93 W at 11 km depth. After the main shock occurred, 437 events were recorded and inverted in order to obtain a three-dimensional velocity image of the region. These records were performed by 34 temporary stations during a period of 6 months including IDL and Evora stations. On a first stage of study, preliminary results were presented by Hamak et al. (2020) coming from the tomographic inversion of 317 aftershocks recorded by 21 short-period stations (CDJ, 2.0 Hz), of the Instituto Dom Luiz (IDL, Lisbon), along a month of records. After analysing the obtained three-dimensional velocity model, a poor ray coverage has been noticed leading to a poor model resolution especially on the edges of the studied area. Thus, additional stations and local data have been integrated to this second stage of study in order to increase the ray coverage which leaded to a more accurate three-dimensional velocity model. Local Tomography Software (LOTOS program, Ivan Koulakov (2009)) was used to perform all the inversions, in order to relocate accurately the aftershock sequence and obtain velocity contrast over the entire region of interest. This program gives the ability to perform a simultaneous inversion of sources location and velocity model. Comparison between preliminary and new aftershock sequence inversion results are presented in this study in order to show the evolution and improvement of the 3D velocity model quality. Nevertheless, despite the increasing model resolution, the studied area is still too small for a good understanding of the complex tectonic of the area. Therefore, in order to extend the area of study, local, regional and teleseismic events must be integrated. As tomographic resolution is related to waves propagation, as this second stage of study demonstrated, by increasing the ray density of the region a more accurate and reliable 3D velocity model will be obtained. Thus, the tectonic features responsible for the seismic activity in the region will be better constrain.