Abstract
 
 
 
 

The contribution of the Ibero-Maghreb region to the global GSHAP map has been the result of a fruitful cooperation among the participants in the established Working Group including representatives from Algeria, Morocco, Portugal, Spain and Tunisia and coordinated by ICTJA-CSIC, Spain. For the first time, a map of regional seismic source zones is presented, and the agreement on a common procedure for hazard computation in the region has been achieved. The computed Ibero-Maghreb seismic hazard map constitutes the first step towards a uniform hazard assessment for the region. Further joint regional efforts are still needed for earthquake hazard studies based on a homogeneous regional earthquake catalogue. Ongoing initiatives in relation to seismic hazard assessment in the Mediterranean should profit both from these results and the established cooperation among different groups in the region as well as contribute to future regional studies.
 
 
 
 
 
 

1. Introduction

The seismic hazard mapping for the Ibero-Maghreb region did not follow strictly the GSHAP implementation guidelines (Basham and Giardini, 1993) due to time limitations. The coordination of the activities in this region by the Institute of Earth Sciences "Jaume Almera" / C.S.I.C. (ICTJA/CSIC) in Barcelona, Spain, started only in December 1996, and a special time schedule was established in order to attain the expected results on time for the contribution to the global GSHAP map.
 
 

At a first Planning Meeting in Barcelona, Spain, on December 18, 1996, a plan of activities was launched to produce a preliminary seismic hazard map of the region (30N-44N, 12W-12E) to be presented in occasion of the 1997 IASPEI General Assembly in Thessaloniki, Greece. During the Planning Meeting national representatives from each country in the region accepted to be responsible for providing the necessary input data (Algeria: Djamal El Foul; Morocco: Ben-Aissa Tadili; Portugal: Carlos Sousa-Oliveira; Spain: Mariano García-Fernández; and Tunisia: M?hamed Chadi).
 
 

In order to simplify the hazard assessment procedure, and due to the limited time, all national representatives agreed on leaving to the ICTJA/CSIC team the responsibility for performing all mapping computations and on providing the necessary pre-processed input data. It was decided to use the computer code SEISRISK III (Bender and Perkins, 1987) to generate a preliminary map of peak ground acceleration (PGA) in g-units for a 90% probability of non-exceedance in 50 years, considering a single regional attenuation law. The production of such regional hazard map was performed in two steps: First, individual hazard maps for each country were obtained based on the input data provided by the national representatives, and compared with available national maps to check the influence of both, the computation algorithm and the applied data reduction. Second, after technical discussions with the national representatives, a single set of regional non-overlapping source zones was defined, based on the available individual zoning map of each country, for which new activity parameters were calculated wherever needed.
 
 

Input data requested to the national representatives were received by April 1997 and a preliminary hazard map for each country was produced by the ICTJA/CSIC team. The input data from Spain was provided by the Spanish Instituto Geográfico Nacional (IGN). To discuss these preliminary results, and to solve problems related to the definition of seismic source zones across national borders, a second Technical Workshop was held in Barcelona on May 11-14, 1997. A representative of the Spanish IGN (A. Izquierdo) attended this meeting. Based on the outcome of this second workshop a first version of the seismic hazard map for the whole region was produced on July 1997.
 
 
 
 

2. Input Data

2.1 Seismicity
 
 

The basic seismicity data from each country consisted on a national earthquake catalogue spaning different time periods and also with different earthquake size descriptor among them. Due to the above mentioned limitation in time, no work was performed in order to obtain an homogeneous regional catalogue. Based on the available national catalogues a List of Significant Earthquakes (LSE) in the region was generated by merging the information of the five countries, avoiding spatial overlapping, as detailed in Table I, and with a uniform earthquake size value on moment magnitude, M, obtained through the empirical relationships given by Johnston (1996a, b). The LSE (Appendix A) contains the seismicity from 1900 to 1989 with estimated moment magnitude M³ 4.5. The LSE is thus just a general description of the seismicity in the region and not a regional earthquake catalogue.
 
 
 
 

2.2 Source Zones
 
 

The definition of seismic source zones in the region is highly heterogeneous, ranging from very detailed small sized sources (Tunisia), and even fault sources (Algeria), to very wide zones (Morocco, Portugal). In most of the cases seismicity is the main criterion parameter based on which source boundaries are defined, although some raw tectonic information had also been included. Figure 1 shows the original source zones from each individual country.
 
 

The time limitations did not allow redefining regional source zones for the whole area, and only those sources overlapping at national borders were considered candidates to be redefined, although trying to keep as much information as possible from original sources, to avoid performing too many new calculations for which no time was available. New sources were defined based on the existing ones after technical discussions among the national representatives. Reshaping was carried out according to known main seismotectonic features and avoiding overlapping or duplicate zones. Figure 2, in which redefined sources are labeled, includes the final distribution of the Ibero-Maghreb regional seismic source zones together with the epicenters contained in the enclosed LSE. The geographical coordinates of the final source zones can be found in Appendix B.
 
 
 
 

2.3 Seismicity Parameters
 
 

The seismicity parameters of the source zones from each country were provided in different format. Either the a and b values of the Gutenberg-Richter relationship (Morocco, Tunisia), or the b-value and the annual rate for a minimum earthquake size (Portugal, Spain), or the number of events by magnitude intervals (Algeria) were given. The seismicity parameters for the reshaped sources (those labeled in Figure 2) were obtained by keeping the original source b-value from which originated, and further calculating earthquake occurrences according to the new reshaped area. A general recalculation was performed for all sources in order to obtain values in terms of moment magnitude, M, through the appropriated empirical relation from Johnston (1996a, b). In Appendix C the main activity parameters (i.e., earthquakes/year km², b-value, and maximum magnitude) of the Ibero-Maghreb regional source zones are summarized.
 
 
 
 

2.4 Attenuation laws
 
 

While in some countries only one attenuation law was used to obtain the available national maps (Algeria, Morocco, Tunisia), others considered several regional laws to apply to different sources (Portugal, Spain). A special case corresponds to the Atlantic offshore sources west of the Iberian Peninsula for which both, Portugal and Spain, assume a very specific low attenuation relation for the strong earthquakes taking place there (i.e., Lisbon 1755).
 
 

Some differences existed as well in the parameters based on which the individual attenuation laws were given. Some included PGA in terms of magnitude, M, and distance, R, (Algeria, Morocco, Tunisia), while in other cases the ground motion parameter considered was intensity, I, either in terms of magnitude and distance (Portugal), or epicentral intensity, I₀, and distance (Spain). For these last cases, it was agreed to obtain PGA in g-units through the empirical relationship log PGA = 0.30 I - 3.22, included in the Spanish Building Code NCS-94 (IGN, 1995). Table II includes all the attenuation relationships used in the region. In the case of Portugal and Spain, only the general one for land sources is included. These same laws are plotted in Figure 3 for a magnitude M=6.0 earthquake.
 
 

For the calculations in the final map, in order to keep a minimum of homogeneity, only one attenuation law was considered. Since a regional law based on good and enough acceleration data was lacking, the relationship proposed by Joyner and Boore (1981) was chosen. This relationship could roughly represent an average of the different laws used in the region, as it can be observed in Figure 3. The hazard calculations were performed considering one standard deviation in log PGA.
 
 
 
 

3. Seismic hazard computation and results
 
 

As a first step, and based on the input data provided by the national representatives, preliminary seismic hazard maps of each country in the region were produced by the ICTJA-CSIC team using SEISRISK III. The maps were compared with existing national maps when available, and if not, obtained results were discussed with the national representatives in order to achieve a general agreement on its validity. The observed differences should arise either from the computation algorithm or from those simplified procedures that had to be applied in some particular cases (e.g., using only one attenuation law for Spain).
 
 

The final seismic hazard map of the Ibero-Maghreb region in terms of PGA, in m/s², with a 90% probability of non-exceedance in 50 years is shown in Figure 4, with color-coded contouring.
 
 

The highest expected hazard corresponds to El Asnam region in Algeria (3.5 ms^-2 or 0.36g). Values between 2.5-3.0 ms^-2 are found in Alger region, Algeria, and in Tunis-Bizerte and Chardimaou-Bou Salem regions, in Tunisia. Sour El Ghozlane and Mostaganem regions, in Algeria, and Granada and Alicante-Murcia regions, in Spain show values between 2.0-2.5 ms^-2.
 
 

Although having followed a simplified procedure, this regional Ibero-Maghreb hazard map is representative as a relative hazard map, i.e., PGA values should be considered as indicative of regional relative seismic hazard. The obtained results as a whole are in good agreement with recently published regional studies (e.g., Benouar et al., 1996).
 
 
 
 

4. Concluding remarks
 
 

This computed Ibero-Maghreb seismic hazard map constitutes only a first step for a uniform hazard assessment for the region. Nevertheless, the effort in compiling all the information regarding presently used methodologies for seismic hazard practice in the region, together with available catalogues, source zoning and attenuation laws, results in invaluable information towards the GSHAP objectives of regionalized approach of hazard assessment for the Western Mediterranean.
 
 

Despite the original lack of homogeneity on the basic elements for seismic hazard assessment, fruitful cooperation among the representatives of each country allowed to produce this first version of a regional seismic hazard map. For the first time, a map of regional seismic source zones is presented, and the agreement on a common procedure for hazard computation in the region has been achieved.
 
 

Further joint regional efforts are still needed for seismic hazard assessment studies in the Ibero-Maghreb region based on a homogeneous regional earthquake catalogue. Ongoing initiatives in the Mediterranean like the CEPRIS Council of Europe OPA-center, the UNESCO IGCP-project SESAME, the UNESCO/USGS-project RELMER, and the ESC Working Group on Seismic Hazard might profit both from these results and from the established close cooperation attained in the framework of GSHAP; as well as contribute to future regional studies.
 
 
 
 

Acknowledgments
 
 

The two technical workshops held in Barcelona, Spain, were partially funded by UNESCO (contract SC/RP205.550.6) and the SESAME project (UNESCO/IGCP-382).
 
 

References
 
 

Basham, P.W. And D. Giardini (1993): Technical guidelines for global seismic hazard assessment, in GSHAP Technical Planning Volume, edited by D. Giardini and P.W. Basham, Ann. Geofis., 36, 15-24.
 
 

Bender, B. and D.M. Perkins (1987). SEISRISK III: A Computer program for Seismic Hazard Estimation, U.S. Geological Survey Bulletin, 1772, 48 pp.
 
 

Benouar, D., G.L. Molas and F. Yamaziki (1996). Earthquake hazard mapping in the Maghreb countries, Earthq. Eng. & Struct. Dyn., 25, 1151-1164.
 
 

IGN (1995). Norma de Construcción Sismorresistente (Parte General y Edificación) NCSE-94, Centro Nacional de Información Geográfica, Madrid, Spain, 105 pp. (in Spanish).
 
 

Johnston, A.C. (1996a). Seismic moment assessment of earthquakes in stable continental regions-I. Instrumental seismicity, Geophys. J. Int., 124, 381-414.
 
 

Johnston, A.C. (1996b). Seismic moment assessment of earthquakes in stable continental regions-II. Historical seismicity, Geophys. J. Int., 125, 639-678.
 
 

Joyner, W.B. and D.M. Boore (1981). Peak horizontal acceleration and velocity from strong-motion records including records from the 1979 Imperial Valley, California earthquake, Bull. Seism. Soc. Am., 71, 2011-2038.
 
 
 
 
 
 

Table captions
 
 
 
 

Table I. Earthquake catalogue available from each country in the Ibero-Maghreb region. M_L: Local Richter magnitude (different definition in each country; e.g., in Tunisia it is obtained from MCS intensity). I_MSK: MSK intensity. LSE: List of Significant Earthquakes (see text).
 
 
 
 

Table II. Attenuation relationships used in the Ibero-Maghreb region.
 
 
 
 
 
 

Figure captions
 
 
 
 

Figure 1. Original earthquake source zones from each individual country in the Ibero-Maghreb region.
 
 
 
 

Figure 2. Final distribution of earthquake source zones for the Ibero-Maghreb region, and epicenters of the generated List of Significant Earthquakes with M³ 4.5 from 1900 to 1989.
 
 
 
 

Figure 3. Selected attenuation relationships from each country in the Ibero-Maghreb region.
 
 
 
 

Figure 4. Earthquake hazard map of the Ibero-Maghreb region. PGA [m/s²] with 90% probability of non-exceedance in 50 years.
 
 

ANNEX 1: National Representatives And Contributing Organizations
 
 

Mohammed CHADI

Institute National de la Meteorologie

BP 156, TN-2035, Tunis-Carthage, TUNISIA

Phone: +216-1-773400 ; Fax: +216-1-772609
 
 

Djamal EL FOUL

CGS

Rue Kaddour Rahim pro., BP 252, 16040 Hussein-Dey, Algiers, ALGERIA

Phone: +213-2-599091 ; Fax: +213-2-776656
 
 

Mariano GARCÍA-FERNÁNDEZ

Inst. De Ciencias de la Tierra ?Jaume Almera? - CSIC

Lluís Solé i Sabaris, s/n, E-08028 Barcelona, SPAIN

Phone: +34-93-4095410 ; Fax: +34-93-4110012 ; E-mail: mgarcia@ija.csic.es
 
 

Arantxa IZQUIERDO and José-Manuel MARTÍNEZ-SOLARES

Instituto Geográfico Nacional. Área de Geofísica.

General Ibáñez de Ïbero, 3, E-28003 Madrid, SPAIN

Phone: +34-91-5333800 ; Fax: +34-91-5979616
 
 

Carlos SOUSA-OLIVEIRA

Inst. Superior Técnico, Dept. Engenharia Civil

Avda. Rovisco Pais, 1, P-1096, Lisboa, PORTUGAL

Phone: +351-1-8418201 ; Fax: +351-1-8497650 ; E-mail: csoliv@civil1.ist.utl.pt
 
 

Ben-Aissa TADILI

Université Mohammed V / Institut Scientifique

Charia Ibn Batouta, B.P. 703, Agdal-Rabat, MOROCCO

Phone: +212-7-774543 ; Fax: +212-7-774540