Large Earthquakes at the IberoMaghrebian Region: Basis for an EEWS Elisa Buforn, Agustín Udías & Carmen Pro Pure and Applied Geophysics pageoph ISSN 0033-4553 Pure Appl. Geophys. DOI 10.1007/s00024-014-0954-0 1 23 Your article is protected by copyright and all rights are held exclusively by Springer Basel. This e-offprint is for personal use only and shall not be self-archived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”. 1 23 Author's personal copy Pure Appl. Geophys. " 2014 Springer Basel DOI 10.1007/s00024-014-0954-0 Pure and Applied Geophysics Large Earthquakes at the Ibero-Maghrebian Region: Basis for an EEWS ELISA BUFORN,1 AGUSTÍN UDÍAS,1 and CARMEN PRO2 Abstract—Large earthquakes (Mw [ 6, Imax [ VIII) occur at the Ibero-Maghrebian region, extending from a point (128W) southwest of Cape St. Vincent to Tunisia, with different characteristics depending on their location, which cause considerable damage and casualties. Seismic activity at this region is associated with the boundary between the lithospheric plates of Eurasia and Africa, which extends from the Azores Islands to Tunisia. The boundary at Cape St. Vincent, which has a clear oceanic nature in the westernmost part, experiences a transition from an oceanic to a continental boundary, with the interaction of the southern border of the Iberian Peninsula, the northern border of Africa, and the Alboran basin between them, corresponding to a wide area of deformation. Further to the east, the plate boundary recovers its oceanic nature following the northern coast of Algeria and Tunisia. The region has been divided into four zones with different seismic characteristics. From west to east, large earthquake occurrence, focal depth, total seismic moment tensor, and average seismic slip velocities for each zone along the region show the differences in seismic release of deformation. This must be taken into account in developing an EEWS for the region. Key words: Ibero-Maghrebian region, focal mechanism, total seismic moment tensor, earthquake early warning system. 1. Introduction The large earthquakes (Mw [ 6, Imax [ VIII) that occur in the Ibero-Maghrebian region have characteristics that differ according to their location, and have caused considerable damage and many casualties. The region’s seismic activity is a reflection of its tectonic complexity, and is associated with the boundary between the Eurasian and African 1 Dpto. de Geofı́sica y Meteorologı́a, Facultad de C.C. Fı́sicas, Universidad Complutense de Madrid, Ciudad Universitaria, 28040 Madrid, Spain. E-mail: ebufornp@fis.ucm.es; audiasva@fis.ucm.es 2 Dpto. de Fı́sica, Centro Universitario de Mérida, Universidad de Extremadura, c/Sta. Teresa de Jornet, 38, 06800 Mérida, Spain. E-mail: cpro@unex.es lithospheric plates. This boundary stretches from the Azores Triple Junction (of the North American, Eurasian, and African plates) on the Mid-Atlantic Ridge eastwards to Tunisia to join with the Sicilian-Calabrian arc. The boundary has a clear oceanic nature in its westernmost part, from the Azores to a point (12!W) southwest of Cape St. Vincent (CSV, SW Iberia). At this point and continuing to the east (to about 1!E), there begins a transition from an oceanic to a continental plate boundary, with the interaction of the southern border of the Iberian Peninsula, the northern border of Africa, and the Alboran basin between them (Fig. 1). One manifestation of this transition is a change in the characteristics of the occurrence of earthquakes, with seismicity spread over a wide region corresponding to an extensive zone of deformation. Earthquakes in this zone have different focal depths, focal mechanisms, and average seismic slip velocities. Further east, the plate boundary recovers its oceanic nature following the northern coastline of Algeria and Tunisia. In order to mitigate the risk of damaging earthquakes, the implementation of Earthquake Early Warning System (EEWS) technologies in this zone is of considerable interest. An EEWS is a real-time system able to detect an earthquake in progress, and to provide fast notification of its potential to cause damage in a target area before the destructive waves arrive. In this study, we examine how the characteristics of the large earthquakes that have occurred in this region differ along the plate boundary. We consider large earthquakes with shallow focus (depth\40 km) that occurred in the historical period (prior to 1900) and in the instrumental period (after 1900). For the purposes of the study, we divide the region into four zones: Gulf of Cádiz (GC) from 12 to 7!W, Morocco (MR) from 7 to 3!W, Southern Iberian Peninsula Author's personal copy E. Buforn et al. Pure Appl. Geophys. Figure 1 Seismotectonic scheme for the Azores-Tunisia region. Arrows show the horizontal stress pattern in the region, together with the main plate boundary and main faults. ATJ Azores triple junction, MAR middle Atlantic ridge, CSV cape of saint Vicente, GC Gulf of Cadiz, SIP South Iberian Peninsula, AB Alboran Basin, MR Morocco, AR Algeria, T Tunisia (based on BUFORN et al. 1988) (SIP) from 9!W to 0!, and Algeria (AR) from 0! to 5!E. This division is based on characteristics of the seismicity, stress pattern, and tectonics and was partially proposed by BUFORN et al. (2004). 2. Seismicity of the Gulf of Cádiz Zone The GC zone, extending in the west to offshore CSV and the zone of the Gorringe Ridge (6.5–12!W and 35–37!N), is the site of large earthquakes which may produce tsunamis. The most important was the 1755 Lisbon quake. The large earthquakes in this zone are presented in Table 1 and Fig. 2 [Instituto Geográfico Nacional (IGN) Data Bank, http://www. ign.es/ign/layout/sismologiaObtencionDatosSismiscos. do, last accessed May 2014; EL MRABET 2005; BUFORN et al. 1988; PRO et al. 2013. In the historical period (before 1900), there were five earthquakes. The earliest for which there is evidence occurred in approximately 246 BC or 218 BC, and is associated with stories about islands sinking near Cádiz and coastal flooding in this zone. This evidence may point to the occurrence of a large offshore earthquake followed by a tsunami, but the year is uncertain. The 881 earthquake is referenced in Arabic sources and is the first whose exact date is known— Table 1 Saint vicente cape-gulf of cadiz Date 246 BC/218 BC? 881/05/26 2 1356/08/24 1531/01/26 1755/11/01 1960/12/05 1964/03/15 1969/02/28 2007/02/12 Time Lat (N) Lon (W) Intensity/ Magnitude Ref 1 36.00 8.00 36.30 39.00 10:16 36.5 21:21:47 35.6 22:34:13.8 36.2 02:40:32.5 36.1 10:35:24 35.9 10.00 8.92 10.0 6.5 7.6 10.6 10.5 VIII 1, VIII IX X (8.7Mw) 6.2 (Ms) 6.1(Ms) 8.0 (Ms) 5.9 (Mw) 1 1 1 3 3 3 4 1 Martı́nez Solares and Mezcua (2002); 2 EL MRABET (2005); 3 BUFORN et al. (1988); 4 PRO et al. (Pro 2013) Lat latitude, Lon longitude, Ref reference Hegira, 22 Chawwâl 267—corresponding to 26 May 881 (POIRIER and TAHER 1980; MARTÍNEZ SOLARES and MEZCUA 2002; EL MRABET 2005). EL MRABET (2005) estimates a maximum intensity of VIII, and POIRIER and TAHER (1980) X. The shock was felt over a wide region of northern Morocco and Algeria, and in southern Iberia it was felt especially strongly in Cordova, with some destruction and landslides. All evidence points to a large offshore earthquake near CSV followed by a tsunami. Author's personal copy Large Earthquakes at the Ibero-Maghrebian Figure 2 Distribution of epicenters for large historical earthquakes (Imax C IX, triangles) and instrumental earthquakes (M C 6.0, circles) taken from the Instituto Geográfico Nacional Data File. Symbols are proportional to the size of earthquake. CSV Saint Vicente Cape, SG Strait of Gibraltar Table 2 South Iberian Peninsula Date Time Lat (N) Lon (W) Intensity/Magnitude Ref 1504/04/05 1518/11/09 1522/09/22 1531/09/30 1680/10/09 1748/03/23 1804/08/24 1829/03/21 1858/11/11 1884/12/25 1909/04/23 09:00 23:30 10:00 04:00 07:00 06:30 08:25 18:38 07:15 21:08 17:39 37.38 37.23 36.97 37.53 36.66 39.03 36.77 38.08 38.30 37.00 38.9 5.46 1.86 2.67 2.73 4.77 0.63 2.83 0.60 8.92 3.98 8.9 VIII-IX VIII-IX VIII-IX VIII-IX VIII-IX IX VIII-IX IX-X IX IX-X IX (6.0 Mw) 1 1 1 1 5 1 1 1 1 1 6 1 MARTÍNEZ SOLARES and MEZCUA (2002); 5 GODED et al. (2008); 6 TEVES COSTA et al. (1999) Lat latitude, Lon longitude, Ref reference The 1356 earthquake’s location was either in Lisbon itself or in the same offshore zone as the Lisbon earthquake (MARTÍNEZ SOLARES and MEZCUA 2002). The shock was felt in Lisbon and Seville with intensity VIII and in Cádiz and at locations as far away as Murcia (500 km east of Seville) with intensity VI. There is no report of any tsunami generated by this earthquake. The 1531 earthquake had maximum intensity IX in the Lisbon area and generated a tsunami (MIRANDA et al. 2012). Using information about the damage in Portugal, the location has been proposed as in the Tagus valley near Lisbon (JUSTO and SALWA 1998; MARTÍNEZ SOLARES and MEZCUA 2002). However, since the earthquake was also felt in Northern Africa, an off-shore location SW of CSV, near that of the Lisbon earthquake, has also been put forward (UDÍAS et al. 1976; MOREIRA 1985; EL MRABET 2005). We agree with this latter alternative. The lack of information about damage in Cádiz or Seville may reflect strong directivity effects (PRO et al. 2013). The 1 November 1755 Lisbon earthquake with maximum intensity X is the largest known to have occurred in this zone. Its location has been estimated as SW of CSV near the Gorringe Bank, breaking through several faults in a SW–NE direction towards Lisbon (PRO et al. 2013). It generated a large tsunami that caused major damage and many casualties along the Atlantic coasts of Portugal, Spain, and Morocco (MENDES-VICTOR et al. 2009; MARTÍNEZ SOLARES 2001; BAPTISTA et al. 2003; EL MRABET 2005). JOHNSTON (1996) estimates its magnitude as Mw = 8.7. MOREIRA DE MENDONÇA (1758) considers the shocks of the 1356, 1531, and 1755 earthquakes to have been very similar, with that of 1531 being the largest. In the instrumental period, there have been three earthquakes in this zone with magnitude 6.0 or greater, namely, 15 March 1964 (Mw = 6.1), 28 February 1969 (Mw = 7.8), and 12 February 2007 (Mw = 6.0). The epicenters of the 1969 (which generated a small tsunami) and 2007 shocks are very close together, located SW of CSV near the epicenter proposed for the 1755 Author's personal copy E. Buforn et al. earthquake. Focal depths of about 30 km were estimated for these latest two quakes (FUKAO 1973; GRIMISON and CHENG 1986; STICH et al. 2007; PRO et al. 2013). This allows one to extrapolate this same value to the focal depth of the 1755 Lisbon earthquake. The 1964 earthquake is located to the east, near the coast, with a shallow (12 km) depth (BUFORN et al. 1988). The focal depth of these earthquakes thus decreases from west to east, from 30–40 km for the 1969 and 2007 events, to 12 km for that of 1964. 3. Seismicity of the Southern Iberian Peninsula Zone In the SIP zone (9!W to 0! and 36–39!N) from 1500 to 1909, there were 11 earthquakes with maximum intensities of VIII–IX or greater (Table 2 and Fig. 2). Three were located along the coast: from west to east, Málaga (1680), Almeria (1522), and Valencia (1748). For the 1680 Málaga earthquake, several authors have proposed an intermediate depth of focus of 40–50 km (MUÑOZ and UDÍAS 1988; GODED et al. 2008). In 1504, a shock occurred in Carmona, near Seville, causing major damage. The 1829 Torrevieja (Alicante) and 1884 Arenas del Rey (Granada) earthquakes, with maximum intensities of X and estimated magnitudes of 6.9 and 6.7, respectively (MUÑOZ and UDÍAS 1991), are the latest earthquakes with these magnitudes in this zone. In 1909, the Benavente earthquake near Lisbon had maximum intensity IX and an estimated Mw = 6.0 magnitude (TEVES COSTA et al. 1999). It is important to note that, from this date to the present, no large (Mw C 6.0) earthquake has occurred in the Iberian Peninsula, so that the twentieth century was an anomalously quiet seismic period. The largest earthquakes in the SIP zone in this period, such as those of 1951 Jaen, 1956 Atarfe-Albolote (Granada), and 2011 Lorca (Murcia), had magnitudes of about 5, although in some cases considerable damage was caused due to their shallow depths of focus and epicenters located very close to towns. 4. Seismicity of the Morocco Zone In the MR zone (9–3!W and 30–36!N), large earthquakes are located along the Atlantic and Pure Appl. Geophys. Mediterranean coasts (Fig. 2 and Table 3). In 1624 an IX–X intensity earthquake occurred inland in Morocco near Fes. All the houses in that city were damaged, as were those in other towns such as Meknes, Beni Weriaghel, and Beni Zerwal (EL MRABET 2005). The northern coast of Morocco had a low level of seismicity until 26 May 1994, when a magnitude Mw = 5.8 earthquake occurred near Al Hoceima, causing major damage and many casualties. Ten years later, on 24 February 2004, a magnitude Mw = 6.4 earthquake occurred very close to the 1994 shock. These two earthquakes had shallow foci, 8 and 6 km, respectively (BEZZEGHOUD and BUFORN 1999; BIGGS et al. 2006). Two earthquakes occurred near Agadir in 1731 and 1960 along the Atlantic coast, with maximum intensities IX and X, respectively. There is no information about the day or month of the 1731 event, only that ‘‘Santa Cruz (Agadir’s name at this time) was destroyed’’ (EL MRABET 2005). For that year (1731), MARTÍNEZ SOLARES and MEZCUA (2002) include an earthquake with epicenter in the Gulf of Cádiz which may correspond to that described as in Agadir by EL MRABET (2005). The 1960 earthquake (maximum intensity 9 and magnitude 5.8 Ms) caused major damage and many casualties due to its shallow depth (3–4 km) and its epicenter located beneath the city of Agadir (CHERKAOUI et al. 1991). These two earthquakes are located at the southern end of a seismic zone extending SW from the northern coast along the Atlas Range. Table 3 Morocco Date 1624/05/ 11 1731 1960/02/ 29 1994/05/ 26 2004/02/ 24 Time Lat (N) Lon (W) Intensity/ Magnitude Ref 34.26 4.57 IX–X 2 30.26 30.52 9.36 9.52 IX X (5.8 Ms) 2 2 08:26:52 35.30 4.03 VIII (5.8 Mw) 7 02:31:19 35.15 3.93 6.4 Mw 8 23:41 2 EL MRABET (2005); 7 BEZZEGHOUD and BUFORN (1999); 8 BIGGS et al. (2006) Lat latitude, Lon longitude, Ref reference Author's personal copy Large Earthquakes at the Ibero-Maghrebian 5. Seismicity of the Algeria Zone Table 4 Algeria The AR zone (0–6!E and 35.5–37.5!N) is seismically very active, with earthquakes located along the coast (Fig. 2 and Table 4). In the western part, in Algeria itself, the largest shock was the 1790 Oran earthquake (Imax = X). This left the city destroyed, and it remained abandoned for some time (LÓPEZ MARINAS and SALORD 1990). Eastwards, the El Asnam area has been subject to frequent large earthquakes, such as those in 1891 (Imax = X), 1934 (Imax = IX), 1954 (Imax = X–XI) and 1980 (Mw = 6.8); somewhat further east, large shocks occurred in 1716, 1825, and 1858; further east still, there occurred offshore the 2003 Boumerdes earthquake (Mw = 6.8) which generated a small tsunami. From that point (roughly 3.5!W) onwards, the seismicity along the coast decreases, with only a single earthquake in 1856 (Imax = X) that was located offshore. Large earthquakes (magnitudes of about 6.5) occur inland following a NW–SE line from the coast to 34!N, 6!E. Earthquakes in this zone occur mostly at shallow depths (\10 km). In Tunisia, there are no reports of large earthquakes even during the historical period. Moderate-size earthquakes occurred in 1863 and 1881 (AMBRASEYS 1962; VOGT 1993). 6. Focal Mechanisms The focal mechanisms of the instrumental period earthquakes with magnitude 6.0 or greater, taken from various authors, are listed in Table 5 and plotted in Fig. 3. In the GC zone, the solutions correspond to thrusting motions with planes oriented NE-SW and dipping either NW or SE. On both sides of the Strait of Gibraltar, the mechanisms change to strike-slip motion with NNW-SSE and ENE-WSW vertical planes and right lateral motion for the 1960 and 1994 events, and more NW–SE and NE-SW for the 2004 event. The Agadir earthquake also has a strike-slip solution with a thrusting component. For all the events, the pressure axis is horizontal and oriented within a NNW-SSE to N–S range. In Algeria the thrust mechanism reappears in the El Asnam and Boumerdes areas, with NW–SE horizontal Date Time Lat (N) Lon (E) Intensity/ Magnitude 1716/02/03 1790/10/09 1825/03/02 1856/08/22 1858/03/09 1867/01/02 1869/11/16 1885/12/03 1887/11/29 1891/01/15 1910/06/24 1922/08/25 1924/03/16 1934/09/07 1943/04/16 1946/02/12 1954/09/09 1980/10/10 2003/05/21 02:00 01:15 07:00 22:00 04:10 07:13:56 12:45 36.7 35.7 36.5 37.1 36.5 36.47 34.9 35.7 35.58 36.5 36.30 36.28 35.5 36.25 36.08 35.75 36.28 36.18 36.93 13:30 04:00 13:27:01 11:47:49 10:18:08 03:39:17 11:43:16 02:43:24 01:04:37 12:25:25 18:44:30 3.10 0.6 W 2.9 5.7 2.9 2.83 5.9 4.8 0.33 1.8 3.70 1.27 5.9 1.71 4.55 4.95 1.47 1.53 3.58 Ref X 9 X 9 X–XI 9 X 9 IX 9 X–XI 9 IX 9 IX 11 IX-X 9 X 9 X 9 X 9 IX 9 IX 9 IX 9 VIII–IX 9 X–XI (6.9 Mw) 15 X (7.1 Mw) 10 12,16 X (6.8 Mw) Lat latitude, Lon longitude, Ref reference 9 BEZZEGHOUD and BENHALLOU (1994); 10 DESCHAMPS et al. (1982); 11 HARBI et al. (2003); 12 DELOUIS et al. (2004); 15 BEZZEGHOUD et al. (1995); 16 HARBI et al. (2007) compression and NE-SW fault planes, dipping to the NW and SE. In the Southern Iberian Peninsula, there have been no large earthquakes in the instrumental period, and hence no information about the mechanisms of its large, historical shocks. We exclude the focal mechanisms of the 1909 Benavente and 1910 Adra earthquakes, with magnitudes of about 6, because they were derived from analogical records with very poor azimuthal coverage and very few stations (Teves COSTA et al. 1999; STICH et al. 2003). We also exclude both the 1954 (Mw = 7.5) and the 2010 (Mw = 6.2) earthquakes because their very deep foci (h & 650 km) have little influence on crustal processes (BUFORN et al. 2011). Earthquakes of magnitude &5 in the same given area can have different focal mechanisms. In particular, there are mechanisms of strike-slip for Bullas 2002 and 2005 and of thrust for Mula 1999 and Lorca 2011 (BUFORN et al. 2005; MANCILLA et al. 2002; BENITO et al. 2007; LÓPEZ COMINO et al. 2012; PRO et al. 2014). Author's personal copy E. Buforn et al. Table 5 Mijtotal ¼ Focal mechanisms Date Strike (8) Dip (8) Slip (8) M0 (Nm) 1960/12/05 1964/03/15 1969/02/28 2007/02/12 1960/02/29 1994/05/26 2004/02/24 1954/09/09 1980/10/10 2003/05/21 73 276 215 246 217 355 298 253 225 70 86 24 52 64 63 79 83 61 54 40 -178 117 58 51 22 2 179 104 83 95 – 291018 60091018 0.8 – 791017 691018 2891018 5091018 2991018 Pure Appl. Geophys. N X M0k mkij ð1Þ k¼1 h (km) Ref 15 12 22 30 3 8 6 10 5 6 3 3 13 4 14 7 8 15 16 17 M0 Scalar seismic moment, h depth, Ref Reference 3 BUFORN et al. (1988); 4 PRO et al. (2013); 7 BEZZEGHOUD and BUFORN (1999); 8 BIGGS et al. (2006); 13 FUKAO (1973); 14 CHERKAOUI et al. (1991); 15 BEZZEGHOUD et al. (1995); 16 DESCHAMPS et al. (1982); 17 DELOUIS et al. (2004) 7. Discussion Figure 4 synthesizes the occurrence and focal mechanism information concerning the study region’s large earthquakes. For the four zones into which we divided the region, we estimated the total seismic moment tensor (TSMT) and the average seismic slip velocity (ASSV), and plotted the horizontal stress axes. The TSMT is defined as the sum of the moment tensors calculated from individual solutions (BUFORN et al. 2004): where k is the number of earthquakes, M0 the scalar seismic moment of each event and mij a normalized seismic moment tensor component. We prefer using this parameter rather than the Frohlich diagrams (FROHLICH and APPERSON 1992) in which all earthquakes have the same weight, independently of their magnitude, so that it is hard to quantify the stress regime in any given area. The amount of compensated linear vector dipole (CLVD) component in the TSMT is indicative of the regularity of an area’s focal mechanism. In particular, if the mechanisms of similar large magnitude earthquakes are all of the same kind, the TSMT’s CLVD component is very low. A high CLVD component in the TSMT is indicative that similar large magnitude earthquakes have different mechanisms. The ASSV was estimated from the TSMT: Du_ ¼ M0 TlS ð2Þ where T is the time period (60 years: 1954–2014), l the rigidity coefficient (4.41 9 104 MPa), and S the total rupture area. We took the values of S to be the areas of the following rectangles (the lengths correspond to the divisions made for purposes of the present study): length 444 km for GC, MR, and AR, and widths of 20 km for GC and 10 km for MR and AR (the average for large Figure 3 Focal mechanisms of large earthquakes (M C 6.0). In black, strike-slip solutions, in grey thrusting solutions Author's personal copy Large Earthquakes at the Ibero-Maghrebian Figure 4 Stress pattern for Ibero-Maghrebian region derived from focal mechanism of large earthquakes. Total seismic moment tensor (TSMT), average seismic slip velocity (ASSV), and plotted horizontal stresses for each area. Large arrows show the regional stress pattern for the whole region. GC Gulf of Cadiz, MR Morocco, SIP South Iberian Peninsula, AR Algeria. Rectangles correspond to the area used to estimate the ASSV events); and length 777 km, and width 10 km for SIP. The focal mechanisms are those plotted in Fig. 3. For the SIP zone, since there were no large earthquakes during the instrumental period, we used the focal mechanisms of the five largest earthquakes (M = 5.0–5.2: 17/04/1968, 24/06/1984, 13/09/1984, 10/12/1989, and 11/05/2011; BUFORN et al. 1995), assuming for each a scalar seismic moment of 4 9 1016 Nm as corresponded to the 2011 Lorca earthquake (Mw = 5). For the GC zone, the TSMT corresponds to thrust faulting with a pure DC component corresponding to the overwhelming relative weight of the large 1969 earthquake; the ASSV is 25 mm/yr. Near the Strait of Gibraltar (MR zone), the rupture process changes to strike-slip faulting, with a small CLVD component (1 %); the ASSV is low (0.4 mm/year). For the SIP zone, the 10 % of CLVD that we found is clear evidence for the different faulting mechanisms present in this region; the TSMT has a strike-slip component similar to that of the MR zone; the ASSV velocity is very low (0.1 mm/yr). For the AR zone (northern Algeria), the focal mechanism is thrusting motion as in GC; the estimated TSMT has a small CLDV component (2 %); the ASSV is 9 mm/yr. These ASSVs do not include folding, thickening, plastic deformation, or slow slip or creep caused by non-seismic relaxation of deformation processes. In consequence, our estimations of ASSVs have different values than those estimated for this region from GPS data (NOCQUET ET CALAIS 2004; SERPELLONI et al. 2007; VERNANT et al. 2010) or predicted by the Nuvel-1A model (DE METS et al. 1994, 2010). These estimates give for the whole region from Azores to Tunisia average values between 4 and 6 mm/y. Our values give lower velocities for SIP and MR and larger velocities for GC and AR with an average of 6.6 mm/y for the whole region. This shows that deformation was seismically unevenly released in the region during the twentieth century. These ASSV are indicative of the differing behavior of earthquake occurrences during the last 60 years in the four zones we divided the region into. In GC, the rate of earthquake occurrence is lower than in AR, although their magnitudes are greater. In the instrumental period (1900–2010), there were three earthquakes in GC of magnitudes greater than 6.0 (1964, 1969, and 2007). The 1969 event (Mw = 7.9) was the largest earthquake to have occurred in the whole Ibero-Maghrebian region in the twentieth century. In AR, there were more earthquakes in this same period, but with generally smaller magnitudes (largest event, Mw = 6.8). In MR, there was less seismic activity, with a lower ASSV (only 20 % of the GC value). In SIP, although as noted Author's personal copy E. Buforn et al. above, there were no large earthquakes during this period, it is known from historical seismicity (Fig. 2) that large earthquakes have occurred in this zone with Imax = X, corresponding to magnitudes greater than 6.0 (e.g., 1680 Malaga, 1829 Torrevieja, 1884 Arenas del Rey). This zone’s lack of large earthquakes in the recent period (1900–2010) brings up the question of whether we are in a seismically anomalous calm period for this zone which may reactivate in the coming years. For the AR zone, seismic activity has increased around Oran in the last 5 years or so, with one Mw = 5.5 earthquake on 6 June 2008 after a long calm period following the large 1790 quake. It is, therefore, also an open question whether this region is undergoing reactivation, with the possibility of a large earthquake similar to the 1790 shock occurring in the coming years. Further east, activity has been high in the areas around El Asnam and Boumerdes, with thrusting motion earthquakes. Figure 4 shows the horizontal stresses in the IberoMaghrebian region obtained from the DC component of the TSMT. The whole region is seen to be under horizontal NNW-SSE compression, as is consistent with the collision of the Eurasian and African plates. However, the northern Morocco earthquakes also show horizontal ENE-WSW extension, with the consequent strike-slip motion. This situation implies a change in the stress regime around the Strait of Gibraltar that may be related to processes in the Alboran Sea such as subduction (BUFORN et al. 1988; MORALES et al. 1999), extensional collapse of thickened continental lithosphere (PLATT and VISSERS, 1989), back arc extension caused by subduction rollback (MORLEY 1993; MICHARD et al. 2002), subduction and breaking of a slab of material (ZECK 1996), or major mechanical decoupling zone within the crust (FERNÁNDEZ-IBAÑEZ and SOTO 2008), where intermediate depth (40–150 km) earthquakes take place (BUFORN et al. 2004). An open question is whether this stress regime also extends to southern Spain. The lack of focal mechanism data for large shocks in this area does not allow one to give a definite answer. For the 1884 earthquake, UDÍAS and MUÑOZ (1979) propose an E–W rupture of 20 km length associated with a complex fault system, but it is unclear whether the motion was predominantly horizontal. SANZ DE GALDEANO (2013) and Grützner et al. (2013), in Pure Appl. Geophys. detailed studies of the Zafarraya fault as the origin of that 1884 event, show that there was extensional E–W motion with a right-lateral component compatible with horizontal NW–SW compression. This change in stress regime along the region was already noted by BEZZEGHOUD and BUFORN (1999) as being responsible for the change from thrust faulting in GC to strike-slip in MR and back to thrust in AR. But in that work, it was proposed that the change starts at around the Al Hoceima area (3!W). Instead, the present analysis shows that the change in stress pattern appears to start already to the west of the Strait of Gibraltar with the 1960 earthquake (Fig. 3), and continues to the east to the Al Hoceima area (3!W). However, the lack of large earthquakes between this longitude and the El Asnam area at 1!E, where there is clear thrusting motion, means that it is not possible to situate exactly where the stress regime changes back again. However, this stress regime change is compatible with the overall regional stress pattern of horizontal NW–SE compression due to the plate collision. It is difficult to extrapolate this general stress pattern south to the Agadir area. Although the focal mechanism of the 1960 earthquake is coherent with horizontal NNW-SSE compression, it may have been a local effect, and a connection may exist with the Al Hoceima area from the seismic activity along the Atlas Range. In sum, the Ibero-Maghrebian region is very complex, and is one in which large, damaging earthquakes have occurred and will continue to occur in the future. An EEWS for this zone could be very useful in helping to mitigate the damage produced by such large earthquakes. The complex characteristics of earthquakes in this region mean that many questions need to be addressed in order to design an appropriate EEWS. One that is particularly important is whether a single EEWS could cover the whole region, or instead, if more than one EEWS would be needed given the different seismic behaviors found in the four zones studied. A preliminary feasibility study for an EEWS in the GC zone has been carried out as part of the ALERT-ES project coordinated by the Universidad Complutense de Madrid (CARRANZA et al. 2013), with the participation of the Real Instituto y Observatorio de la Armada de San Fernando, Cádiz and the Institut Geologic de Catalunya, Barcelona. The extension of that work to an EEWS for the entire Author's personal copy Large Earthquakes at the Ibero-Maghrebian region is planned in the ALERTES-RIM project that will begin in 2015 with the same participants. Acknowledgments This work has been partially supported by MINECO, projects CGL2010-19803-C03-01 and CGL201345724-C3-1 and by the INNOCAMPUS project (Ministerio de Economı́a y Competitividad, Orden CIN/1934/2010). REFERENCES AMBRASEYS, N. (1962), The seismicity of Tunis. Annals of Geophysics 15, 233–244, doi:10.4401/ag-5431 BAPTISTA, M.A., J. M. MIRANDA, F. CHIERCI, and N. ZITTENI (2003), New study of the 1755 earthquake source based on multi-channel seismic survey data and tsunami modeling. Nat. Hazards Earth Syst. Sci. 3, 333–340. BENITO, B., R. CAPOTE, P. MURPHY, J. M. GASPAR-ESCRIBANO, J. J. MARTÍNEZ-DÍAZ, M. TSIGE, and D. STICH (2007), An Overview of the Damaging and Low Magnitude Mw 4.8 La Paca. Bull. Seism. Soc. Am., 97, 671–690, doi:10.1785/0120050150. BEZZEGHOUD, M., and H. BENHALLOU (1994), Les seismes en Algeries de 1365 a 1992. Algiers: Centre de Recherche en Astronomie, Astrophysique et Geophysique. BEZZEGHOUD, M., D. DIMITRO, J. C. RUEGG and K. LAMMALI (1995), Faulting mechanism of the El Asnam (Algeria) 1954 and 1980 earthquake from modeling of vertical movements. Tectonophysics 249, 249–266. BEZZEGHOUD, M., and E. BUFORN (1999), Source parameters of the 1992 Melilla (Spain, Mw = 4.8), 1994 Alhoceima (Morocco, Mw = 5.8) and 1994 Mascara (Algeria, Mw = 5.7) earthquakes and seismotectonic implications. Bull. Seism. Soc. Am. 89, 359–372. BIGGS, J., E. BERGAMN, B. EMMERSON, G. FUNNING, J. JACKSON, B. PARSONS, and T. WRIGHT (2006), Fault Identification for Buried Strike-Slip Earthquakes using InSAR: The 1994 and 2004 Al Hoceima, Morocco Earthquakes. Geophy J. Int. 166, 1347–1362. BUFORN, E., A.UDÍAS and M. A. COLOMBÁS (1988), Seismicity, source mechanisms and seismotectonics of the Azores-Gibraltar plate boundary. Tectonophysics, 152, 89–118. BUFORN, E., C. SANZ de GALDEANO, and A. UDÍAS (1995), Seismotectonics of the Ibero-Maghrebian region, Tectonophysics 248, 247–261. BUFORN, E., M. BEZZEGHOUD, A.UDÍAS, and C. PRO (2004), Seismic sources on the Iberia-African plate boundary and their tectonic implications, Pure Appl. Geophys, 161, 623–646, doi:10.1007/ s00024-003-2466-1. BUFORN, E., B. BENITO, C. SANZ DE GALDEANO, C. del FRESNO, D. MUÑOZ, and I. RODRÍGUEZ (2005), The Murcia (SE Spain) region: Seismotectonic and seismic risk implications, Bull. Seism. Soc. Am. 95, 549–567. BUFORN, E., C. PRO, S. CESCA, A. UDÍAS, and C. del FRESNO (2011), The 2010 Granada, Spain, Deep Earthquake. Bull. Seism. Soc. Am., 101, 5, 2418–2430. doi:10.1785/0120110022. CARRANZA, M., E. BUFORN, S.COLOMBELLI and A. ZOLLO (2013), Earthquake early warning for southern Iberia: A P wave threshold-based approach. Geophys. Res. Lett. 40, 1–6, doi:10. 1002/grl.50903. CHERKAOUI, T.E., F. MEDINA and D. HATZFELD (1991), The Agadir earthquake of February 29, 1960. Examination of some parameters, In J. Mezcua and A. Udı́as (eds), Seismicity, Seismotectonics and Seismic Risk of the Ibero-Maghrebian Region., Madrid: Instituto Geográfico Nacional, 133–1148. DELOUIS, B., M. VALLÉE, M. MEGHRAOUI, E. CALAIS, S. MAOUCHE, K. LAMMALI, A. MAHSAS, P. BRIOLE, F. BENHAMOUDA, and K. YELLES (2004). Slip distribution of the 2003 Boumerdes-Zemmouri earthquake, Algeria from teleseismic, GPS and coastal uplift data. Geophys. Res. Lett. 31, 1–4, doi:10.1029/2004GL020687. DE METS, C., GORDON, R. G., ARGUS, D. F., and STEIN, S. (1994), Effect of recent revisions to the geomagnetic reversal time scale on estimates of current plate motions, Geophys. Res. Lett. 21, 2191–2194. De Mets, C., R.G. GORDON and D. ARGUS (2010), Geologically current plate motions, Geophys. J. Int. 181, 1–80, doi:10.1111/j. 1365-246X.2009.04491. DESCHAMPS, A., Y.GAUDEMER, and A. CISTERNAS (1982), The El Asnam, Algeria, earthquake of 10 October 1980: multiple source mechanism determined from long period records. Bull. Seism. Soc. Am. 72, 1111–1128. El MRABET, T. (2005), The Great Earthquakes in the Maghreb Region and their Consequences on Man and Environment. Rabat: CNRST-LAG. FERNÁNDEZ-IBÁÑEZ, F. and J. SOTO (2008), Crustal rheology and seismicity in the Gibraltar Arc (western Mediterranean), Tectonics 27, 2, TC2007, http://dx.doi.org/10.1029/2007/TC002192. FROHLICH, C. and K.D. APPERSON (1992), Earthquake focal mechanisms, moment tensors and consistency of seismic activity near plate boundaries, Tectonics 11, 279–296. FUKAO, Y. (1973). Thrust faulting at lithospheric plate boundary. The Portugal earthquake of 1969, Earth Plan. Sci. Lett. 18, 205–216. GODED, T., E. BUFORN, and D. MUÑOZ, (2008), The 1494 and 1680 Málaga (Southern Spain) earthquakes, Seism. Res. Lett. 79, 707–715, doi:10.1785/gssrl.79.5.707. GRIMISON, N., and W. CHENG, (1986), The Azores-Gibraltar plate boundary: focal mechanisms, depths of earthquakes and their tectonic implications. J. Geophys. Res. 91, 2029–2047. GRÜTZNER, C.,P. RUANO, A.JABALOY, J. GALINDO-ZALDIVAR, P. BECKER-HEIDMANN, C. SANZ de GALDENAO, A.RUDERSDORF, and K.REICHERTER (2013), Late Holocene rupture history of Ventas de Zafarraya Fault (Southern Spain), Cuaternario y Geomorfologı́a 27, 52–61. HARBI, A., S. MAOUCHE and H.BENHALLOU, (2003), Re-apprisal of seismicity and seismotectonics in the north-eastern Algeria. Part I: Review of historical seismicity, J. Seismology, 7, 115–136. HARBI A., S. MAOUCHE, F. OUSADOU, Y. ROUCHICHE, A. YELLESCHAOUCHE, M. MERAHI, A. HEDDAR, O. NOUAR, A. KHERROUBI, H. BELDJOUDI, A. AYADI, and D. BENOUAR (2007), Macroseismic Study of the Zemmouri Earthquake of 21 May 2003 (Mw 6.8, Algeria). Earthquake Spectra, 23, 2, 315–332. JOHNSTON, A. C. (1996), Seismic moment assessment of earthquakes in stable continental regions, III New Madrid 1811–1821, Charleston 1886 and Lisbon 1755. Geophys. J. Int., 126, 314-344. JUSTO, J.L. and C. SALWA, (1998), The 1531 Lisbon earthquake, Bull. Seism. Soc. Am. 319–328. Author's personal copy E. Buforn et al. LÓPEZ MARINAS, J. M., and R. SALORD, (1990). El periodo sı́smico oranés de 1790 a la luz de la documentación de los archivos españoles. Madrid: Instituto Geográfico Nacional. LÓPEZ COMINO J, F. MANCILLA, J. MORALES and D. STICH (2012), Rupture directivity of the 2011, Mw 5.2 Lorca earthquake (Spain), Geophys. Res.Lett., 39: L03301. doi:10.1029/2011GL050498. MANCILLA, F.L., C. AMMON, R. HERRMANN, and J. MORALES (2002), Faulting parameters of the 1999 Mula earthquake, southeastern Spain. Tectonophysics, 354, 139–155. MARTÍNEZ SOLARES, J. M. (2001). Los efectos en España del terremoto de Lisboa. Madrid: Instituto Geográfico Nacional. MARTÍNEZ SOLARES, J. M. and J. MEZCUA, (2002). Catálogo sı́smico de la Penı́nsula Ibérica (880 a.C.- 1900). Madrid: Instituto Geográfico Nacional. MENDES-VICTOR, L., C. S. OLIVEIRA, J. AZEVEDO, and A. RIBEIRO (eds.) (2009), The 1755 Lisbon earthquake:revisited. Berlin: Springer. MICHARD, A., CHALOUAN, A., FEINBERG, H., GOFFEÅL, B., and MONTIGNY, R. (2002), How Does the Alpine Belt End between Spain and Morocco? Bull. Soc. Geol. France 173, 3–15. MIRANDA, J., J. BATLLÓ, H. FERREIRA, L.M. MATIAS, and M.A. BAPTISTA, (2012), The 1531 Lisbon earthquake and tsunami, Proc. 15 WCEE Meeting, Lisbon, 2012. MORALES, J., SERRANO, I., JABALOY, A., GALINDO-ZALDIVAR, J., ZHAO, D., TORCAL, F., VIDAL, and GONZALEZ-LODEIRO, F. (1999), Active Continental Subduction beneath the Betic Cordillera and the Alboran Sea, Geology 27, 735–738. MOREIRA, V. S. (1985), Seismotectonics of Portugal and its adjacent area in the Atlantic, Tectonophysics, 117, 85–96. MOREIRA DE MENDONÇA, J. J. (1758), Historia Universal dos Terremotos, Lisbon: Antonio V. de Silva. MORLEY, C. K. (1993), Discussion of Origins of Hinterland Basins to the Rif-Betic Cordillera and Carpathians, Tectonophysics, 226, 359–376. MUÑOZ, D., and A. UDÍAS (1988), Evaluation of damage and source parameters of the Málaga earthquake of 9 October 1680.In W. H. K. Lee, H. Meyers, and K. Shimazaki (eds.), HistoricalSeismograms and Earthquakes of the World, San Diego: Academic Press, 208–221. MUÑOZ, D. and A. UDÍAS, (1991), Three large historical earthquakes in southern Spain. In J. Mézcua and A. Udı́as (eds), Seismicity, Seismotectonics and Seismic Risk of the Ibero-Maghrebian Region. Madrid:Instituto Geográfico Nacional, 175–182. NOCQUET, J.M. and E. CALAIS (2004), Geodetic Measurements of Crustal Deformation in the Western Mediterranean and Europe, Pure Appl. Geophys. 161, 661–681, doi:10.1007/s00024-0032468-z. PLATT, J.P. and VISSERS, R. (1989), Extensional Collapse of Thickened Continental Lithosphere: A Working Hypothesis for the Alboran Sea and Gibraltar Arc, Geology 17, 540–543. Pure Appl. Geophys. POIRIER, J. P.,and M. A. TAHER (1980), Historical seismicity in the near and middle east, north Africa and Spain from Arabic documents (VIIth-XVIIIth century). Bull. Seism. Soc. Am. 70, 2185–2201. PRO, C., E.BUFORN, M.BEZZEGHOUD, and A.UDÍAS, (2013), The earthquakes of 29 July 2003, 12 February 2007, and 17 December 2009 in the region of Cape Saint Vincent (SW Iberia) and their relation with the 1755 Lisbon earthquake, Tectonophysics 583, 16–27, http://dx.doi.org/10.1016/j.tecto.2012.10.010. PRO, C., E. BUFORN, S. CESCA, C. SANZ DE GALDEANO, and A. UDÍAS (2014). Rupture process of the Lorca (southeast Spain) 11 May 2011 (Mw = 5.1) earthquake. J. Seismology. 18, 481–495, doi:10.1007/s10950-014-9421-8. SANZ DE GALDEANO, C. (2013). The Zafarraya Polje (Betic Cordillera, Granada, Spain) a basin open by lateral displacement and bending. J. Geodynamics 64, 62–70, http://dx.doi.org/10. 1016/j.jog.2012.10.004. SERPELLONI, E., G. VANNUCCI, S. PONDRELLI, A. ARGNANI, G. CASULA, M. ANZIDEI, P. BALDI and P. GASPERINI (2007), Kinematics of the Western Africa-Eurasia plate boundary from focal mechanisms and GPS data, Geophys. J. Int. 169, 1180–1200. doi:10.1111/j. 1365-246X.2007.03367.x. STICH, D., F. MANCILLA, S. PONDRELLI and J. MORALES (2007), Source analysis of the February 12th 2007, Mw 6.0 Horseshoe earthquake: Implications for the 1755 Lisbon earthquake, Geophysical Research Letters 34. http://dx.doi.org/10.1029/ 2007GL0300127. TEVES-COSTA P, J. F. BORGES, I. RIO, R. RIBEIRO and C. MARREIROS (1999), Source parameters of old earthquakes: semiautomatic digitization of analog records and seismic moment assessment. Nat. Hazards 19:205–220. UDÍAS, A., A. LÓPEZ ARROYO and J. MEZCUA (1976), Seismotectonics of the Azores-Alboran region, Tectonophysics, 31, 259–289. UDÍAS, A. and D. MUÑOZ, (1979), The Andalusian earthquake of 25 December 1884, Tectonophysics, 53, 291–299. VERNANT, P., A. FADIL, T. MOURABIT, D. OUAZAR, A. KOULALI, J. MARTIN DAVILA, J. GARATE, S. MCCLUSKY and R. REILINGER (2010). Geodetic constraints on active tectonics of the Western Mediterranean: Implications for the kinematics and dynamics of the Nubia-Eurasia plate boundary zone. J. Geodynamics 49, 123–129. VOGT, J. (1993), Further research on the historical seismicity of Tunisia. Terra Nova 5, 475–480. ZECK, H. P. (1996). Betic–Rif orogeny: subduction of Mesozoic Tethys under E-ward drifting Iberia, slab detachment shortly before 22 Ma and subsequent uplift and extensional tectonics. Tectonophysics 254, 1–16. (Received July 14, 2014, revised September 11, 2014, accepted October 6, 2014)