E. Buforn, A. Udías and J. Martín Dávila

Dpt. Geofisica, Facultad CC. Fisicas, Universidad Complutense, Madrid (Spain)

Introduction

On February 12, 2007 an earthquake occurred at 10h 35m 25.36s, 180 km SW of San Vicente Cape. The earthquake was felt without appreciable damage over a large area, in south and central Spain and Portugal. In Madrid the shock was felt in parts of the city and some tall buildings were evacuated.

The earthquake occurred in the western part of the Ibero-Maghrebian region, a complex region which includes the western part of the plate boundary between Eurasia and Africa. It extends from Iberia to the Maghreb (northern Africa) including the Gulf of Cádiz and the Alboran Sea to both sides of the Strait of Gibraltar (figure 1). On the Gulf of Cádiz, the epicenters are distributed on an E-W narrow band that marks the plate boundary (Buforn et al. 1988), with the occurrence of large shallow earthquakes as the February 28, 1969 (Ms=8.1) or the March 15, 1964 (Ms=6.4) (figure 1). The area where the 2007 event has occurred is one of the hypothetical locations suggested for the 1755 Lisbon earthquake (Machado, 1966; Buforn et al., 1988).

Figure 1.- Focal mechanisms for earthquakes with magnitude greater than 4.0 for the Gulf of Cádiz region. In red reverse solutions, blue strike-slip and green normal solutions, symbols are proportional to the magnitude. The star shows the location of the 12-02-07 earthquake. Solutions are given in Buforn et al. (2004). Dashed line represents the plate boundary.

The preliminary hypocentral location given by the Instituto Geográfico Nacional (IGN) is 35.956º N, 10.409º W, depth=64 km, maximum intensity IV (EMS-98) in Sevilla, Cádiz, Huelva (SW Spain). This hypocenter is very close to the 1969 event (36.10ºN, 10.60ºW). The focus of the 1969 earthquake was estimated at 29km by Grimisson and Chen (1988) from waveform modelling. The focus of the 1964 event, located to the East of the 1969 and 2007 shocks, the focus was estimated by waveform modelling by Grimisson and Chen (1988) as a complex rupture, with two sources at 14 and 20 km respectively and at 12 km by Buforn et al (1988) using a single rupture. However, both these studies were carried out using WWSSN analogue data, and in consequence, no details about the rupture process were obtained. The 2007 earthquake is the first earthquake to occur in this area with magnitude larger than 5.5 for the last 38 years.

Focal mechanism

A preliminary solution for the focal mechanism of the 2007 event has been estimated from body wave inversion at teleseismic distances using the Kikuchi and Kanamori (1991) method. The Green functions have been generated using a crustal 3-layer 1-D model derived from local studies in the region composed by three flat layers (Gonzalez et al., 2001). A total of 12 P waves and 6 SH waves recorded at stations located at epicentral distances from 30º to 90º have been used. Data were converted to ground motion by removing the instrument response and were then filtered using a pass band 0.01 to 1Hz. The time window used on the inversion is 35 s. and the rupture velocity 3.0 km/s.

As a preliminary solution in the inversion process the fault plane obtained from first motion of 58 observations located at teleseismic and regional distances has been used. We tried depths ranging from 10 to 90 km for the focus and the fault plane has been divided into a 2.5 x 2.5 km grid. Best results (minimum rms) have been obtained for 30 km depth, with good agreement between observed and synthetic waveforms. The obtained focal mechanism corresponds to reverse faulting with planes dipping about 45º and oriented on E-W direction. Scalar seismic moment is 8.5 x 1017 Nm, which corresponds to a Mw=5.9. The source time function is a single impulse of 4s. From this solution we have carried out the slip distribution (figure 2).

Figure 2.- Slip distribution for the San Vicente 2007 earthquake. On top scalar seismic moment (Nm), magnitude Mw, source time function, focal mechanism and slip over the fault plane. On bottom observed and synthetic seismograms for P and SH waves, numbers corresponds to azimuth and amplitude of each station.

The rupture starts at 30 km depth, propagating upward and showing a possible asperity at 32 km depth and 0.63 m of maximum dislocation. Faulting takes place on a plane (azimuth 264º) of 14 km length and 12.5 km width, dipping to the north with the African plate underthrusting the Eurasian plate.

The obtained solution is very similar to the moment tensor solutions obtained by other agencies (figure 3): Harvard (CMT), INGV (Italy) and ETH (Suitzerland). Depths obtained by Harvard, INGV and ETH are deeper (44 and 43 km respectively). The IGN solution shows a vertical plane trending NE-SW, depth 65 km and Mw=6. The USGS solution is similar to the IGN but with shallow depth (14km). These two solutions do not agree with those found for earthquakes in this area, specially that of 1969 earthquake. We prefer our solution due to the large number of stations used (18) and the good azimuthal coverage. For example, the ETH solution has been obtained using 4 stations with azimuths 24º to 56º or the IGNV has been obtained using only MEDNET station with a poor azimuthal coverage. The IGN solution has been obtained using only 3 stations at regional distances with azimuthal coverage between 43º to 59º.

Figure 3.- Depth versus Mw magnitude for the focal mechanism of the 2007 earthquake estimated by different agencies.

The focal mechanism estimated for the 2007 earthquake is very similar to the solutions obtained for the 1964 and 1969 shocks, the largest in this area (figure 1). This solution is in agreement with the stress pattern in this area, corresponding to horizontal NNW-SSE compression, a consequence of the convergence between the Eurasian and African plates.

Acknowledgements

This work has been supported in part by the Ministerio de Ciencia y Tecnología (Spain), project CGL2006-10311-C03-01 and by the Universidad Complutense de Madrid, project AE10/07. Part of data was provided by IRIS and ORFEUS.

References

Buforn, E., Udías, A. and Colombás, M.A. (1988). Seismicity, source mechanisms and seismotectonics of the Azores-Gibraltar plate boundary. Tectonophysics, 152, 89-118.

Buforn, E., Bezzeghoud, M., Udías, A. and Pro, C. (2004). Seismic sources on the Iberia-African plate boundary and their tectonic implications. Pure Appl. Geophys. 161, 623-646.

Gonzalez-Fernández, A., Córdoba, D., Matías, L. and Torné, M. (2001). Seismic structure in the Gula of Cádiz (SW Iberian Peninsula). Mar. Geophys. Res. 22, 207-223.

Grimison, N. and Chen, W. (1988). The Azores-Gibraltar plate boundary: focal mechanisms, depths of earthquakes and their tectonic implications. J. Geophys. Res. 91, 2029-2047.

Kikuchi, A. and Kanamori, H. (1991). Inversion of complex waves III. Bull. Seism. Soc. Am. 81, 2335-2350.

Machado, F. (1966). Contribuçao para o studio do terremoto de 1 de Nov. de 1755. Rev. Fac. Cienc., Univ. Lisboa, Ser. C., 14; 19-41