C Springer 2005 Journal of Seismology (2005) 9: 99–110 Reevaluation of the earthquakes of 10 March and 19 May 1951 in southern Spain A. Udı́as1 , D. Muñoz1 , E. Buforn1 , C. Sanz de Galdeano2 , C. del Fresno1 & I. Rodriguez1 1 Departamento de Geofı́sica y Meteorologı́a, Fac. Ciencias Fı́sicas. Universidad Complutense, 28040 Madrid; Instituto Andaluz de Ciencias de la Tierra, Fac. de Ciencias, Universidad de Granada, 18071 Granada; e-mail: audiasva@fis.ucm.es 2 Received 7 June 2004; accepted in revised form 5 November 2004 Key words: earthquake damage, focal mechanism, intensity, seismicity, Spain 1951 earthquake, tectonics Abstract Damage and parameters of the earthquakes of 10 March and 19 May 1951 in southern Spain have been reevaluated. Data available do not allow accurate depth determinations and previous estimates of larger depths are not confirmed, so depths have been fixed at 30 km for both shocks. Magnitudes (Ms) have been determined as 5.4 and 5.6, respectively. Intensities estimated at 22 and 29 sites from contemporary documentary sources give maximum values of VI–VII and VI (EMS Scale), lower than previous estimates. The focal mechanism for the May shock is rightlateral strike-slip with a normal component of motion, with planes with strikes 273◦ and 169◦ ; seismic moment 1.9 × 1016 Nm and dimension 6 km (radius of circular fault). Shocks are located near the boundary between the Iberian plateau and the Guadalquivir Basin and may be related to faults connected with this boundary. Introduction Southern Spain is a region of moderate seismic activity, where earthquakes of magnitude above 5 are infrequent. For this reason the study of these earthquakes is of particular importance, though at a global level they may seem rather trivial. Two of these earthquakes with magnitude of approximately 5 occurred in 1951, very close in location and separated in time by two months, the first being on 10 March (shock 1) and the second on 19 May (shock 2). Both earthquakes were followed by series of aftershocks, caused moderate damage in the epicentral zone and were felt over a wide area. They were not located in the most seismic region of Granada-Malaga, but somewhat to the north, near the town of Jaén, where seismicity is lower, and shocks of this size are rare (Figures 1 and 2). The first studies of these earthquakes were made by Due-Rojo (1952a, 1952b), Bonelli and Esteban-Carrasco (1953), Rey Pastor (1954), Chacón-Alonso (1965), Munuera (1966). More recent studies are by Mezcua (1982), Mezcua and Martı́nez Solares (1983), Vidal (1986) and Bernal et al. (1991). Both shocks were considered by the earlier authors who studied them to have foci at greater than normal depths, about 100 km. Some inconsistencies in these studies regarding their location, depth and damage have motivated this new study, in which original seismograms and contemporary reports of damage have been reevaluated and the hypocentral locations, magnitudes and intensity maps have been redetermined and the focal mechanism of the event of 19 May has been found. Seismicity of the area Recent seismicity (1960–2000) for shocks mb ≥ 3 is shown in Figure 1. There are groups of shocks on both sides (East and West) of the epicentral area of the two shocks of 1951, but none are located nearby. The greatest concentration of activity is found to the south, extending west-east, from the Gulf of Cadiz to the Mediterranean coast. The distribution of epicenters for historical earthquakes, from 1500 to 1950, with 100 Figure 1. Epicenter location of earthquakes for 1960–2000, h < 40 km, M ≥ 4 (IGN Earthquake Data File). Sizes of dots are proportional to magnitudes, the smallest represent shocks of magnitude 4. Stars correspond to epicenters of 10-3-1951 and 19-5-1951. Figure 2. Epicenter location of historical earthquakes with maximum intensity greater than or equal to 6, for the period 1500–1950 (IGN Earthquake Data File). Intensities: circles 9, squares 8, triangles 7, rhombuses 6. Stars correspond to epicenters of 10-3-1951 and 19-5-1951. The area corresponds to that marked in Figure 1. maximum intensities (MSK) (Mezcua, 1982) equal and larger than 6 (VI, Arabic numerals rather than Roman will be used here for intensity degrees; Bormann, 2002), near the epicentral area of the two shocks studied here, is shown in Figure 2. Greater activity is concentrated to the south, near the city of Granada, where earthquakes with maximum intensities 8 (1806) and 10 (1884) have occurred. The only earthquakes 101 near Jaén are those of 1582 (Imax . = 7) (no day given), 27-2-1712 (Imax = 6) and 21-12-1944 (Imax = 6). The 1582 earthquake has been recently added to the Spanish Earthquakes Catalogue of the Instituto Geográfico Nacional (IGN), Madrid, Spain; it is not mentioned by Galbis-Rodriguez (1932), the classic reference for historical seismicity in Spain. The occurrence of this earthquake is based on only one contemporary reference about an earthquake in Alcalá la Real (Figure 2) mentioning “fall of towers and walls”; no more details are given and there is no other reference to it. It is questionable if the shock really occurred and if its maximum intensity was 7. The earthquake of 1712, south of Jaén, was the largest of a sequence, which lasted between 2 February and 11 March. This earthquake caused some light damage in some houses in Jaén (Figure 2), but none have to be abandoned; a maximum intensity of 6 seems to be a reasonable estimation. The earthquake of 1944 has been assigned by IGN a magnitude of mb = 4. The observatories of Cartuja, Almerı́a and Ebro, which recorded it, assigned a maximum intensity between 5 and 6 in Huelma (Figure 2). Thus, the shocks of 1951 appear to be the only ones which reached intensity 7 (MSK) for the Jaén region in the last 500 years. Hypocentral locations The epicenter locations, focal depths and magnitudes determined by different authors for the two earthquakes are given in Table 1. The epicenter locations are shown in Figures 3 and 4. The first determination of the epicenters was made by Due-Rojo (1952a, 1952b); he gave the same location for both shocks. Due-Rojo stated that the epicenters had been determined using data from near stations, but did not give the details of his determinations. The depths were given as 80 km; this depth, considered to be uncommon for earthquakes in southern Spain, was justified as based on the distributions of intensities and the P wave arrival times at the station of Cartuja (Granada). No details of the use of these arrival times are given, so that we cannot verify the argument. Due-Rojo’s epicentral determination was reported by the Bulletin of the International Seismological Summary (ISS; 1951). New determinations of the hypocenters were made by Bonelli and EstebanCarrasco (1953), using a least squares numerical procedure, and considering straight rays from the source to eight local stations. They were forced to disregard the arrival times of the stations Cartuja (CRT) and Málaga (MAL), the two nearest stations, which did not fit the Table 1. Hypocentral locations and magnitudes Date Origin time Lat. (◦ N) Long. Depth (◦ W) (km) M Author 38.1 3.7 (80) 1 38.1 3.6 117 2 10: 38: 32 38.1 3.6 100 7.4 (mb ) 3 10: 38: 36 38.18 3.82 10-3-1951 10: 38: 32 4.8 (MD ) 4 10: 38: 24.5 38.12 3.99 25 5.0 (mb ) 5 10: 38: 27.1 37.98 3.86 30 5.4 (Ms ) 6 19-5-1951 15: 54: 24 38.1 3.7 (80) 37.6 4.1 140 1 2 15: 54: 19 37.8 4.0 140 7.5 (mb ) 3 15: 54: 19 37.58 3.93 5.1 (MD ) 4 15: 54: 24.6 37.58 4.01 60 5.5 (mb ) 5 15: 54: 23.6 37.69 4.12 30 5.6 (Ms ) 6 15: 54: 23.0 37.50 4.12 90 6 1. Due-Rojo (1951). 2. Bonelli and Esteban-Carrasco (1953). 3. Munuera (1966). 4. Mezcua and Martı́nez Solares (1983). 5. Vidal (1986). 6. This study. rest of the observations. The solutions gave an epicenter similar to that of Due Rojo for the first shock, and one displaced southwest for the second, with depths of 117 and 140 km. Though the method is questionable, because of its use of a half-space with constant velocity and direct arrivals for all distances, the epicentral locations, with shock 1 about 50 km northeast of shock 2, are similar to those found by more recent determinations. Munuera (1966), in a study of intermediate depth earthquakes in southern Spain, gave similar results for the epicenters and found depths of 100 km and 140 km (no details of his determinations are presented). The epicenters were recalculated for the Spanish Earthquake Catalogue of IGN by Mezcua and Martı́nez Solares (1983). The calculation were made by means of the HYPO71 program, using 8 and 10 observations (P and S arrival times) for shocks 1 and 2, respectively. Two different epicenters are given, with shock 2 about 60 km to the south-west of shock 1. No focal depths are given for either shock. A more recent determination by Vidal (1986) gave epicenters for both shocks slightly displaced to the west with respect to those of the IGN Catalogue. Focal depths were given as 25 km for shock 1 and 60 km for shock 2. This is the first time the first shock was given a normal depth. No details are given of 102 Figure 3. Intensity map for the earthquake of 10 March 1951 and location of epicenter according to several authors. Figure 4. Intensity map for the earthquake of 19 May 1951 and location of epicenter according to several authors. this determination, or what data have been used to change the determination of the IGN Catalogue, nor, more important, how the focal depths were determined. The greater depth of shock 2 was said to correspond to the form of the isoseismal lines, with no further discussion. We have recalculated the two hypocenters using the readings taken from the stations catalogues and 103 original readings from some of the seismograms from near stations. The epicenters obtained are not far from those given in the IGN Catalogue. For the calculation we used the program HYPOINVERSE-2000 (Klein, 2000) and a 7 layer crustal structure for the South Spain region (Dañobeitia et al., 1998). In this determination readings with residuals larger than 7 s (all far stations and most S) were eliminated, which left only P and S arrival times from local stations. The final solution was obtained from readings of 7 stations (7 P and 2 S). The stations used in each event, with their corresponding distances and azimuths, are listed in Table 2. For both events the distribution of stations does not allow a good determination of depth, since there is only one station (CRT, Cartuja) at a distance less than 100 km. The residuals for shock 2 are considerably lower (1 s) than those of shock 1 (5 s). We must keep in mind that all seismograms of local stations are of smoked paper with paper velocity of 30 mm/min, so that readings, at best, are only accurate to 0.5 s. With depth free in the determination, the program did not converge. With the depth fixed from 10 to 110 km, at 5 km intervals, the root mean square error (RMSE) of the residuals decreases from 5.7 s to 5.6 s (shock 1) and from 1.5 s to 0.6 s, (shock 2). For both events the epicenter move systematically to the south, away from the center of maximum intensity (Figures 5, and Figures 3 and 4). For shock 1, changing the depth of the focus, the RMSE values were Table 2. Epicentral distances and azimuths of stations used in hypocentral determination Azimuth (◦ ) Station Distance (km) CRT MAL ALM TOL ALI LIS EBR AVE 77.6 138.2 167.8 222.0 298.6 473.7 495.6 603.7 183 202 132 356 78 283 47 214 CRT MAL ALM TOL ALI LIS EBR 19 May 1951 44.1 93.1 166.1 260.4 332.4 460.0 540.7 150 196 116 2 73 289 46 10 March 1951 practically constant (5.65). Hence, we have selected the solution for a fixed depth of 30 km, in agreement with depth obtained from the intensity attenuation, as will be seen later. For shock 2, a minimum RMSE between 0.8 s and 0.6 s corresponds to depths between 75 km and 100 km. For depths less than 30 km the RMSE are larger than 1.5 s. In Table 1 we have chosen the solutions for fixed depths of 30 km (RMSE = 1.5) and 90 km (RMSE = 0.63). Though the argument is not conclusive, owing to the poor distribution of the available stations, most of them to the south of the epicenter, shock 2 may, in fact, have been a deeper event. However, the distribution of intensities larger than 6 for shock 2 (Figure 4) is concentrated around the calculated epicentral area, similarly as for shock 1 (Figure 3), suggesting rather a shock of normal depth. As will be seen later, the attenuation of intensities agrees well with a focal depth of 30 km for both shocks. We may conclude that, with the arrival-time data available, we cannot clearly decide on the depth of the foci. It is probable that both shocks have focal depth at the base of the crust, though shock 2 may be deeper. Both shocks were recorded also at teleseismic distances, ISS Bulletin lists 74 readings for shock 1 and 97 for shock 2, including observations from North American stations. These readings have not been used in the epicentral determination due to their very large residuals. A number of aftershocks were recorded at Cartuja (CRT), the closest station. The seismograms of this station have been lost, so we could not check them. According to Due-Rojo (1953) there were 35 aftershocks recorded at CRT between 10 March and 8 April and 64 between 19 May and 26 July. The number of aftershocks supports the assumption of a crustal focus for both events. Magnitude determination The first estimates gave the same maximum intensity 8 in the Forel–Mercalli scale to both shocks (Due-Rojo, 1952a). The first magnitudes determined by Munuera (1966) gives values of mb equal to 7.4 and 7.5 for shocks 1 and 2, respectively (Table 1). No details are given of how these magnitudes were determined and they are obviously too large in comparison with the damage caused. Different formulas were used by Munuera for magnitude determination and we do not know which one was used in this case (Lopez and Muñoz, 2003). However, this is the first indication that shock 2 was slightly larger than shock 1. Mezcua and Martinez 104 Figure 5. Change in the epicenter location of the earthquakes of 10-3-1951 and 19-5-1951, for different focal depths. Solares (1983) assigned also a smaller magnitude to shock 1 than to shock 2 (mb = 4.8 and 5.1); their magnitudes were based on signal duration adjusted to the values of mb (Lg) (formulas are given in the reference). Vidal (1986) corrected these values to 5.0 and 5.5, but he does not describe how these magnitudes have been determined. In a recent study, Mezcua et al. (2004), using the isoseismal maps of Bernal et al. (1991), have determined a value of MW = 6.1 for both shocks (based on the distribution of intensities). This value is too high and does not agree with magnitudes obtained from instrumental data prior to and in this study. Seismograms from the near stations in Spain were all out of scale, so they could not be used for magnitude determination. Using the seismograms from Strasbourg, De Bilt and Clermont Ferrand, with the IASPEI-1964 formula, we have calculated the Ms Figure 6. Seismograms of the N-S component of the 19 tons horizontal inverted pendulum seismograph at Strasbourg for the earthquakes of 10 March 1951 (top) and 19 May 1951 (bottom). 105 values, with the results of 5.4 and 5.6. The only station where we have found good seismograms for both events, with the same instrument, is Strasbourg (19 Tons inverted pendulum horizontal seismograph). The seismograms are shown in Figure 6; showing clearly that shock 2 was larger than shock 1; the difference in the logarithms of the maximum amplitudes of surface waves of the two shocks is 0.21. This agree with the damage, number of aftershocks and number of station reporting to ISS. Damage and intensity maps Due-Rojo (1952b), using the Forel–Mercalli scale, presented the first evaluation of seismic intensities with a single isoseismal map, corresponding to the combined damage produced by both shocks. The Forel–Mercalli scale, used by Spanish authors, is a 12 degree scale similar to the Mercalli–Cancani–Sieberg scale (MCS). Table 3. Intensities—10 March 1951 A B C Alcaudete 6–7 6–7 7 Baena 6–7 7 7 Carolina, La 6–7 5–6 7 Jaén 6–7 6 7 Montefrı́o 6 7 5 Vilches 6 – 8 Andujar 5–6 5–6 8 Linares 5–6 7 8 Úbeda 5–6 6 6 Alcalá la Real 5 6–7 6 Baeza 5 5–6 7 Córdoba 5 5–6 6 Málaga 5 4–5 4 Granada 4–5 5–6 4 Iznajar 4–5 5–6 5 Sevilla 4–5 4–5 5 Alhama de Granada 4 5 5 Antequera 4 – 4 Bélmez de la M. 4 4–5 5 Madrid 4 3–4 3 Martos 4 5 7 Torredonjimeno 4 5 6 A: This study. B: Bernal et al. (1991). C: Mezcua (1982). Most 12 degree scales are roughly equivalent to one another in actual values (Gruenthal, 1993). Maximum intensity was 8, and there was a large area covered by intensities 5 and 4, up to north of Madrid. The large extent of these two intensity areas led him to assign a large focal depth to both shocks. Mezcua (1982), using Table 4. Intensities—19 May 1951 Alcalá la Real Jaén Baena Martos Montefrı́o Alcaudete Córdoba Almadén Andujar Arcos de la F. Bélmez de la M. Carchalejo Huelma Pozo Alcón Cabra Doña Mencı́a Mancha Real Antequera Carolina, La Cazorla Granada Guadix Linares Loja Lucena Madrid Marmolejo Torredonjimeno Úbeda A B C 7 6–7 6 6 6 5–6 5–6 5 5 5 5 5 5 5 4–5 4–5 4–5 4 4 4 4 4 4 4 4 4 4 4 4 7 6 – – 4 – 6 6 – 5–6 6 4 5 5 – – 4 6 – 5 5 – 4 – 4 – 4 – 8 8 7 8 7 8 6 4 6 5 6 5 7 6 7 6 7 5 5 6 6 6 7 6 6 3 5 7 6 A: This study. B: Bernal et al. (1991). C: Mezcua (1982). Table 5. Fault plane solution of earthquake of 19 May, 1951 Plane A: Plane B: T-axis P-axis Strike Dip Slip 273 ± 35 169 ± 30 Plunge 7±3 31 ± 8 58 ± 6 69 ± 5 Trend 223 ± 33 128 ± 31 −155 ± 5 −35 ± 6 106 the MSK scale, presented two isoseismal maps based on the data collected in the Archives of the Instituto Geográfico Nacional (Madrid). The maps do not specify the locations where the intensities were assigned. It is not clear, whether values of intensity were estimated at each of the location given in the maps. For both shocks the highest intensity is 8 (MSK), over an area of about 60 by 30 km. Bernal et al. (1991), using the reports which appeared in four contemporary newspapers, for 40 locations for shock 1, and in three newspapers for 25 locations for shock 2, reevaluated the intensities. Their intensities are, in general, lower than those of Mezcua (1982). For both shocks the maximum observed intensity was 7 (MSK). We have reevaluated seismic intensities for each shock, using contemporary information from up to ten different local newspapers, from administrative reports of town halls and from church documents. From this information we have been able to get a better estimate of the damage at a larger number of towns and villages than in the previous studies. We estimated intensities for each shock, using EMS-1992 scale (Gruenthal, 1993); These are shown in Figures 3 and 4. The Figure 7. Attenuation of intensity with distance. (a) 10 March 1951. (b) 15 May 1951. Dashed line shows the attenuation corresponding to values: a = 6, h = 30 km. 107 distribution of intensities is very similar for both events and the maximum values coincide with the instrumental epicentral area. For shock 1 greater damage happened at the following localities: At Alcaudete the Church of El Carmen, suffered large cracks and the fall of part of the roof, and many houses suffered damage in roofs and chimneys. In Baena several houses suffered fall of parts of the facade and fall of staircases. In Jaén the Cathedral had cracks in the central vault and in several arches, with fall of stones and plaster; and similar damage happened in the church of San Ildefonso; in old buildings many cracks were formed. In La Carolina, some roofs suffered failures and on walls many cracks developed. We estimated the intensity at these localities to be between 6 and 7 (EMS). For shock 2, greatest damage happened at the following localities: In Alcalá la Real, the top of the tower of the church of San Antón, which had already some damage from shock 1, fell; a corner of the wall of the castle of La Mota fell, part of the wall of the bullring fell, in the church of La Consolación cracks occurred and large pieces of plaster fell; cracks, fall of plasters and light damage occurred in many houses. In Jaén, plaster and ornamentation of the cathedral fell; two large stones of the tower of the church of San Ildefonso fell and an abandoned house collapsed. We estimated the intensities at these locations to be grades 6 and 7 (EMS), respectively. In Figure 3 we can see that shock 1 has two distinct regions of highest intensity (6–7), the first in its epicentral area and the second to the southwest, in the towns of Baena, Alcaudete and Montefrio; this latter region corresponds to the epicentral area of shock 2 (Figure 4). This is not so for shock 2, which has a single area of intensity between 6 and 7, in its epicentral area. In this area surface rocks are unconsolidated sediments of the Subbetic formations, in contrast to more consolidated sediments to the north (see Figure 9). Therefore, the intensities felt at Baena, Alcaudete and Montefrı́o can be considered a site effect, with motion being amplified by the unconsolidated sediments. This would explain both, the presence of the second area of intensity 6–7 in shock 1, and the slightly higher intensities (7) estimated for shock 2. Another possibility for the slight higher intensities estimated for shock 2 is the weakening effects of the first shock on the structures. Both effects are probably present. Tables 3 and 4 present for both shocks the localities with intensities I ≥ 4 of our evaluation, those of Bernal et al. (1991) and those of Mezcua (1982). Bernal et al. (1991) and Mezcua (1982) used MSK scale, while we have used EMS, but the degrees of both scales correspond to each other (Gruenthal, 1993). Mezcua’s values are taken from his isoseismal maps, since no values are given for individual locations. We have assumed that the locations indicated in the map have intensity values corresponding to the isoseismal areas in which they are located. When we compare our evaluations with those from Bernal, there is fairly good agreement, with only a few values lower in our estimations. Most of the values derived from Mezcua’s isoseismal maps appear to be overestimated by one or two units, compared to our values and of Bernal et al. (1991). In particular, for shock 1, intensity 8 is given to Linares, Andujar and Vilches, while our estimates are only 5–6, and 6. For shock 2, intensity 8 is given to Alcalá la Real, Jaén, Martos and Alcaudete, while our estimates are 7, 6– 7, 6 and 5–6. Mezcua gives maximum intensity 8, for both shocks, while our estimation is 6–7 and 7, and that of Bernal et al. (1991) is 7 for both. We have also found that Bernal et al. (1991) mistake Bélmez de la Moraleda (Jaén) with Belmez (Cordoba), thereby distorting the intensity 6 area to the west in the map of shock 2. Figure 7 shows the distribution of intensities with distance for both shocks. Attenuation of intensities in both events is similar, decreasing from the maximum values (6–7) to intensity 4 over about 200 km distance. The decrease of intensities have been fitted with the Figure 8. Focal mechanism of the earthquake of 19 May 1951. 108 relation (Ergin, 1969).

I0 − I = a log( x 2 + h 2 / h) + b( x 2 + h 2 − h) The second term (anelastic attenuation) was neglected, because their influence on the calculated values for this range of distances was very small. The values of a and h fitted, for the two earthquakes give: a = 6, h = 30 km. The fit of the curve has been made by taking first average values of intensity along the same azimuths from the epicenter. In spite of the dispersion of the data, the obtained values for the depth confirm that these earthquakes were not deep (of the order of 100 km), as proposed by earlier authors (see the discussion above). Moreover, similar values of a and h have been also found for the intensity attenuation of other earthquakes in southern Spain (Muñoz, 1974). Focal mechanism and tectonics The focal mechanism of shock 2 was first calculated by Chacon-Alonso (1965) using P wave polarities. He used Byerly’s method of extended distances, but not properly and he gave non-orthogonal nodal planes. He Figure 9. Tectonic map of the region with the epicenter location of earthquakes of 10-3-1951 and 19-5-1951 and the focal mechanism of 19-5-1951. 109 used data from bulletins from 18 stations. We tried a solution from the same set of data, but it has a low score 0.61 (11 out of 18 correct readings). Therefore, we have not used these data in our study, but have obtained copies from seismograms and read directly the P wave polarities. The solution for only clear polarities (17), using the algorithm of Brillinger et al. (1980), has a score of 0.88 (Table 5 and Figure 8). It corresponds to normal faulting with a right-lateral component of strike-slip. We have calculated the scalar seismic moment and dimension from the spectra of body waves of records from stations, Scorbysund, Pasadena and Weston, obtaining a seismic moment of Mo = 1.9 × 1016 Nm and from the corner frequency a radius of 6 km. The corresponding moment magnitude is Mw = 4.8. This value is lower than 5.7 which is obtained from the value of Ms (5.6) using the relation MW = 0.796MS + 1.280, derived for southern Europe (Bungum et al., 2003) and much lower than MW = 6.1, given by Mezcua et al. (2004). Figure 9 shows the location of the epicenters of the two shocks and the focal mechanism of shock 2 on a tectonic map of the area. No earthquakes of intermediate depth have been ever detected there, and the 100 km depth for shock 2 does not fit with the tectonics of the area. All deep earthquakes in South Spain are located further south of this area, near the coast and in the Alboran Sea, and are related to a subduction zone located there (Buforn et al., 1997). If we accept, then, a depth of 30 km for both shocks, we place the foci the base of the crust, near the southern boundary of the Iberian Plateau relating them to faults connected to that boundary. The Paleozoic basement here dips to the south under the sedimentary cover of the Guadalquivir Basin, and the different geological units of the Betic Cordillera; the shocks could be related to the bending of the basement. Both shocks are located to the north of the Cadiz-Alicante system of faults (lower right hand side of Figure 9), while the greatest part of the seismicity of southern Spain is located to the south of this system of faults (Figure 1). The mechanism of shock 2, has a nodal plane with strike east-west; this approximately agrees with the orientation of the contact between the Iberian Plateau and the Guadalquivir Basin and may be associated with faults related to the contact. Another possibility is that the second nodal plane, with strike north-south, is related to short faults with this orientation, also present in the area. The horizontal pressure axis, trending NW-SE, agrees well with the regional direction of stresses in southern Spain related with the convergent motion between Iberia and Africa (Buforn et al., 2004). Conclusions The focal parameters and damage of the earthquakes of 10 March and 19 May, 1951, near Jaén in southern Spain, have been reevaluated, in view of the different estimations given by previous authors. The focal depths, a controversial issue in previous studies, cannot be resolved only by the available first arrivals times. The depths of both earthquakes have been fixed at 30 km for both shocks, also on the evidence of the intensity distribution and attenuation, though the May earthquake could have been somewhat deeper. Values of magnitude, Ms = 5.4 and 5.6, have been calculated. Damage have been reevaluated from new contemporary documents giving maximum intensities estimated as 6–7 and 7 in the EMS scale, respectively, lower than previous estimates. The focal mechanism from polarities of first motions of the second shock shows normal faulting with right-lateral component of strike-slip motion. This mechanism may be correlated with the mapped faults at the southern boundary of the Iberian plateau. This study shows the difficulties encountered in the reevaluation of earthquakes of moderate magnitude from this epoch (1950s) for which the number and quality of available instrumental data are low. Acknowledgments The authors wish to thank Dr. J.M. Martinez Solares of the Instituto Geográfico Nacional (Madrid) for providing important information and data, the directors of the stations of De Bilt (Holland), Strasbourg (France), Weston (USA), Pasadena (USA) and others agencies which have provided data for this study and to Dr. L. Drake who revised text. This work has been supported in part by the Ministerio de Ciencia y Tecnologı́a (Spain) project REN 2003-05178-C03-01. References Bernal, A., Barrera, T. and Santiago, J. L., 1991, The earthquakes of March 10th and May 19th 1951 occurred in the province of Jaén, Spain. In: Mezcua, J. and Udı́as, A. (eds.), Seismicity, Seismotectonics and Seismic Risk of the Ibero-Magrehbian Region. Instituto Geográfico Nacional, Madrid, 149–161. Bonelli, J. and Esteban-Carrasco, L., 1953, Resultados provisionales del estudio del carácter sı́smico de la falla del Guadalquivir. Instituto Geográfico Catastral, Madrid. 110 Bormann, P., 2002. IASPEI New Manual of Observatory Practice (Vol. 1). GeoforschungZentrum, Potsdam. Brillinger, D. R., Udı́as, A. and Bolt, B. A., 1980, A probability model for regional focal mechanism solutions, Bull. Seis. Soc. Am. 70, 140–179. Buforn, E., Coca, P., Udı́as, A. and Lasa, C., 1997, Source mechanism of intermediate and deep earthquakes in southern Spain, J. Seis. 1, 113–130. Buforn, E., Bezzeghoud, M., Udı́as, A. and Pro, C., 2004, Seismic sources in the Iberian-African plate boundary and their tectonic implications, Pageoph 161, 623–646. Bungum, H., Lindholm, C. D. and Dahle, A., 2003, Long-period ground motion for large European earthquakes, 1905–1992 and comparison with stochastic prediction, J. Seis. 7, 377–396. Chacon-Alonso, L., 1965. Ensayo del método de Byerly para la determinación de fallas sı́smicas aplicado a la zona de MartosAlcaudete (Jaén), utilizando el terremoto de 19 de Mayo de 1951. Instituto Geográfico y Catastral, Madrid. Dañobeitia, J. J., Sallares, V. and Gallart, J., 1998, Local earthquakes seismic tomography in the Betic Cordillera (South Spain), Tectonophysics 160, 225–239. Due-Rojo, A., 1952a, Die südspanischen Erdbeben von März bis August 1951. Neues Jahrbuch für Geologie und Palëantologie/Monatheften 1, 4–5. Due-Rojo, A., 1952b, El periodo sı́smico de la provincia de Jaén (marzo-agosto de 1951), Las Ciencias 17(1), 49–54. Due-Rojo, A., 1953, Movimientos sı́smicos en España durante el año 1951, Bull. Real Soc. Española Historia Natural 50, 59–71. Ergin, K., 1969, Observed intensity-epicentral distance relations in earthquakes, Bull. Seis. Soc. Am. 59, 1227–1238. Galbis-Rodriguez, J. 1932, Catálogo sı́smico de la zona comprendida entre los meridianos 5 E y 20 W de Greenwich y los paralelos 45 y 25 N. Instituto Geográfico, Catastral y de Estadı́stica. Madrid. Gruenthal, G. (ed.), 1993, European Macroseismic Scale 1992 (updated MSK-scale). Conseil de l’Europe, Cahiers du Centre Européen de Géodynamique et de Séismologie 7, 79 p. Klein, K. W., 2000, User’s guide to HYPOINVERSE-2000, a Fortran program to solve for earthquake location and magnitude. U.S. Geological Survey, Open File report (preliminary 0.99). Lopez, C. and Muñoz, D., 2003, Fórmulas de magnitud en los boletines y catálogos españoles, Fı́sica de la Tierra 15, 49–71. Mezcua, J., 1982, Catálogo general de isosistas de la penı́nsula Ibérica. Instituto Geográfico Nacional, Madrid, Publ. 202, 261 p. Mezcua, J. and Martı́nez Solares, J. M., 1983, Sismicidad del área Ibero-Mogrebı́. Instituto Geográfico Nacional, Madrid, Publ. 203, 299 p. Mezcua, J., Rueda, J. and Garcı́a-Blanco, R.M., 2004, Reevaluation of historic earthquakes in Spain, Seism. Research Letters 75, 75–81. Munuera, J. M., 1966, Dos secuencias de sismos Ibéricos, posiblemente de foco intermedio, Revista de Geofı́sica 99–100, 167–185. Muñoz, D., 1974, Curvas medias de variación de la intensidad sı́smica con la distancia epicentral. Tesis Licenciatura. Universidad Complutense de Madrid. 50 p. Rey Pastor, A., 1954, Estudio morfo-tectónico de la falla del Guadalquivir. Instituto Geográfico y Catastral, Madrid, 48 p.