Since the last Newsletter, our region has been devastatedby an earthquake which killed over 2000 people in Algeria,rendered 150,000 homeless and caused €100M of damage,including to boats and underwater cables. Our thoughtand sympathies are with our Algerian neighbours, whohave, once again, suffered the effects of awesome tectonicforces which are a characteristic of our European-Mediterranean region. An article in this Newsletterexplains in more detail the role of EMSC members inrapidly gathering, and making available, objectiveinformation on such events even when communicationswith the affected area, are, inevitably, difficult. A specialsession on the earthquake was convened at the mostrecent RELEMR meeting of eastern and southernMediterranean countries, in Cyprus, which wascoordinated by EMSC in association with UNESCO,USGS, the Geological Survey of Cyprus and ORFEUS. TheEMSC’s web-page has been of particular value in keepingthe community and public updated on this earthquake;web traffic increased dramatically, demonstrating the needfor this special service and its further development.

Our joint FP6 infrastructure bid, with ORFEUS, is inthe review process. We consider that, regardless of theoutcome, the wide-ranging consultation and

cooperation in preparing it will provide a platform forfurther collaboration and bids in the future. Thecomplementary roles of EMSC and ORFEUS in servingour seismological community have been further definedand endorsed in the process.

Finally, news from our sister organisation and EMSCmember by right, the International SeismologicalCentre. Following an indication last year, that RayWillemann would like to move on at the end of 2003,after six years, his successor as Director of the ISC hasbeen selected. Avi Shapira, presently the director of theGeophysical Institute of Israel, who built up itsSeismological Group and who is a long-time supporterof ISC and EMSC, will be continuing the significantdevelopments achieved under Ray Willemann both inThatcham, and in ISC/EMSC collaboration. TheEMSC’s monthly bulletin, which should go live in early2004, will be yet another vehicle for fostering thatprocess. We wish Ray every success in his future careerin the US and welcome Avi into his important new rolewith international seismological infrastructure.

Chris BrowittPresident

The magnitude Mw6.8 (ETH, INGV, CPPT) earthquake which struckAlgeria on May 21st, 2003 is the largest earthquake to have occurred inthe region since the El-Asnam earthquake, Ms7.3 in 1980. The humanand economic consequences are disastrous for the country. Thisearthquake is of particular scientific interest for the Euro.-Med. Regionbecause it shares some characteristics with a large area of theMediterranean region. It has occurred in a zone characterised by arelatively moderate and diffuse seismic hazard where the vulnerabilityremains significant. It also illustrates the difficulty of implementing anefficient prevention policy in a region where past damaging events havebeen long forgotten and where the authorities had to deal with a boom innew buildings.

This special section, as a follow-up to the EMSC special Web page(http://www.emsc-csem.org/Html/ALGER_210503.html), aims atproviding a preliminary description of the main characteristics of thisevent at a time scale located after the initial focus of the main media andbefore peer-reviewed articles have been published and when limitedinformation is available. We hope that this will contribute to a betterunderstanding of the seismic risk in the Mediterranean region and thusimproved prevention policies.

EDITORIAL

Centre Sismologique Euro-MéditerranéenEuropean-Mediterranean Seismological Centre

www.emsc-csem.org

N° 20 SEPTEMBER 2003

Newsletter

No ISSN : 1607-1980

CONTENT

• EMSC actions concerning the Boumerdes-Zemmouri event p.2

• The Boumerdes -Algiers (Algeria) Earthquake of May 21st, 2003 (Mw=6.8) p.3

• Analysis of Strong Ground Motions Recorded during the 21st May, 2003 Boumerdes, Algeria, Earthquake p.5

• A Short Note on Building Damage p.7

• The 2003 Boumerdes (Algeria) earthquake:Source process from teleseismic data p.8

• The tsunami triggered by the 21 May 2003 Algiers earthquake p.10

• Monitoring of Seismicity in Albania p.13

• The CRAAG, Algeria p.17

• Seismicity of Egypt p.19

Foreword to the special section onthe May 21st, Algeria earthquake

Alert system and Real TimeSeismicityThe Zemmouri event has triggered the EMSCalert procedure 8 minutes after the occurrenceof the event. The first reviewed alert messagewas disseminated 41 minutes after theoccurrence announcing a mb 6.0; themagnitude was re-evaluated 47 minutes laterto Ms6.6 and then completed by a Mw 6.8

The final epicentral location which includesregional arrival data (from Algeria, Tunisia,Egypt and Balearic Island) is 37.02°N, 3.76°E.It is important to note that the initialepicentral location disseminated in the firstalert message differs by less than 3 km.

The moment tensor determined by ETHZ(Zurich, Switzerland) was automatically madeavailable on our web site 2 hours after theearthquake occurrence and was followed by thesolutions provided by INGV-MedNet (Italy),CPPT (Pamataï, LDG, France), Harvard andUSGS solutions (Figure 1).

Following the main shock, 20 aftershocks werelocated by EMSC and 4 of them have been thesubject of an alert and/or information message.

Special web pageA special web page has been opened on May22nd to provide a more detailed and regularly-updated information on this event, itsseismotectonic context, the aftershock activityand the consequences. It includes maps ofseismic activity but the majority of thecontributions was kindly provided by theseismologist community. This page hasinduced a significant increase in the numberof individual daily visitors to the EMSC Webpage, from an average of about 250 before theevent to a daily average of 900 over the wholemonth of June (Figure 2). During this timeperiod 400,000 pages have been viewed (thisweb page has had a durable impact on ourvisibility with, in July, an average of 1,300individual visitors a day). It has also beenlinked by a number of sites and referenced byseveral publications. These figures clearlydemonstrate the need for such a special pagein this particular type of situation.

Conclusion The alert system has performed well for theZemmouri event both in terms of rapidity andreliability, thanks to the data contributors. Suchinformation was particularly important for thisevent as the communication with Algeria hadbeen severely hampered following the breakageof several sub-marine phone cables. Officialorganisations that EMSC deals with such asthe Council of Europe or ECHO have receivedthe source parameter information in a timelymanner. New organisations have also showninterest for closer relationship. Notably, the K9

search and rescue team (Germany) informed usthat our email message has allowed them tocontact their task force more rapidly than usualand they arrived on the field the day followingthe earthquake; from now on, they haverequested to receive messages by fax.The special web page has proved to be a veryefficient way to disseminate information on thisevent. We will do our best to systematicallyopen such a page in case of a damaging event in

the Euro.-Mediterranean region and improvethe service. For example, the regional tectonic,historical and instrumental seismicity andhazard level should be made available morequickly; Joint hypocenter determinations wouldalso help to follow the seismic activity andidentify the active structures… Today, we wouldreally appreciate to receive your feedback,comments or ideas to further improve ourservices.

CSEM / EMSC Newsletter

2

EMSC actions concerning the Boumerdes-Zemmouri eventR. Bossu, S. Godey and G. Mazet-Roux, EMSC

2

2

3

3

4

4

5

5

6

6

7

7

35 35

36 36

37 37

38 38

EMSC

Main shock

21/05 03:14 aftershock

21/05 13:57 aftershock

24/05 19:21 aftershock

27/05 17:11 aftershock

Main shockPreliminary locationRevised locationAlgerian stations usedin the revised location

AftershocksMag < 44 <= Mag < 55 <= Mag < 6Mag >= 6

Quick CMT

CPPT

INGV

ETH

USGS

HARVARD

Figure 1: Map of the seismic activity in the Boumerdes area following the May 21st mainshock.

Figure 2: Evolution with time of the daily individual visitors (individual IP) of EMSC website. Values are averaged in a centred, seven-day-long moving window.

0

200

400

600

800

1000

1200

1400

1600

1800

01/04 16/04 01/05 16/05 31/05 15/06 30/06 15/07

Num

ber

of

Ind

ivid

ual V

isit

ors

per

Day

Bo

umer

des

Ear

thq

uake

, Alg

eria

M5,

7, S

E T

urke

y

AbstractThe Boumerdes-Algiers earthquake ofmagnitude Mw:6.8 is the second major eventwhich seriously affected Algeria in the lasttwenty years. This seismic event, located 7 kmnorth of the village of Zemmouri (50 Km east ofAlgiers) happened in a region characterizedpreviously by a low to moderate seismicactivity. With a depth of 10 km, the earthquakekilled 2300 persons and caused huge damagein the epicentral area located between Dellysand Algiers. In this region, site effects such asliquefaction, minor landslides, rock falls,surface breaks have been observed. Theearthquake was generated by an unknownoffshore reverse fault, the Zemmouri fault. ThisNE-SW oriented fault extents over 50 kmalong the margin, between Dellys and Corso.

Geological investigations revealed in theepicentral area, surface breaks affecting theQuaternary sedimentary cover. These breaksdisplay two main orientations parallel to thecoastline: N35° from Corso to Dellys and N120from Corso to Tamentefoust. Along thiscoastline, an uplift of the seafloor with a meanvalue of about 60 cm has been also observed.

IntroductionOn May 21st 2003 at 19h 44’ 19”GMT, adestructive earthquake of magnitude Mw=6.8occurred in Boumerdes, a city located 50 kmeast of Algiers. The epicentre was located 7 Kmnorth of the sailors village of Zemmouri(CRAAG) (Figure 1).

The earthquake, with a maximum intensity ofX, caused important damage in Boumerdesand the main villages of the region (Zemmouri,Dellys, Thenia,...). In these urban centres,many recent buildings totally collapsed andmany of them have been seriously damaged(figure 2). In Algiers, thousands of buildingshave been affected with various intensities. Forsome of them, torsion of the pillars or deepcracks on the walls were observed. In theepicentral area 2273 persons died and about200 000 people were made homeless.

Using records of the Algerian seismologicalnetwork , the epicentre of the main shock waslocated precisely in the Algerian margin at3°58 E, 36° 91N. The magnitude of Ml=6.2was calculated using time duration of seismicsignals recorded at stations of the AlgerianSeismic Network operated by the CRAAG. Adepth of 10 Km was determined andrepresents a value close to the depth of theAlgerian earthquakes. The CRAAG epicentrecharacteristics are slightly different fromthose given by EMSC (3.76 E, 37.02N ) andUSGS (3.78 E, 36.89N).

The focal mechanisms calculated by theinternational seismological centrescorrespond to a NE-SW thrust fault ( strike:54 ; dip: 47 ; rake: 86). The seismic moment isM0=2.4x 1019 N.m. (from Yagi, 2003) .

Following the occurrence of the main shock,an important aftershock activity wasrecorded by a portable seismic networkinstalled by the CRAAG in the epicentralarea. A rapid fruitful cooperation with theUMR Geosciences Azur (France) also allowedinstallation for three weeks, of four OBS, fewmiles from the coastline. Processing of theaftershocks is still going on but we canpointed out that the spatial aftershocksdistribution is mainly concentrated offshorealong an NE-SW axis extending from Dellysto Corso. The aftershock activity was markedby the occurrence of two main events ofMl=5.8 occurring respectively on May 24th

and 28th.

The main shock has been followed by a smalltsunami, which affected the Spanish coast.During the main shock, the sailors observed atemporary withdraw of the sea of about 100m. Along the Spanish coast of the BalearicIslands, a sea wave height of about 1.5 msunk hundred small boats parking in theports.

Historical seismicityDuring history, the region of Algiers-Boumerdes and its surroundings, wasrepeatedly affected by moderate to largeearthquakes (Rothé, 1950, Benhallou,1985;Ambraseys and Vogt,1988). These events weregenerally located within the Mitidja basin.Nevertheless, some of these earthquakes wererelated to coastal faults such as theearthquake of January 2nd 1365 or the one,which occurred on February 3rd, 1716.

In the region of Boumerdes, the mostimportant seismic event, which happenedduring the twentieth century in the region,was on September 16th,1987 (M=5.2). Thisearthquake had no effects on the region as nohuge damage was reported. Other minorearthquakes occurred in the region ofBoumerdes-Thenia, sometimes felt by thepopulation.

In the western side, Algiers region wasaffected more strongly by three moderateearthquakes, the Oued Djer event on October31st, 1988, Ml=5.0; the Tipaza event onOctober, 1989 27th, Ml=6.0 and the Ain Benianearthquake on September 4th, 1996, Ml=5.7.These last two earthquakes displayed reversemechanisms and were generated by ENE -WSW reverse faults.

Historical seismicity of this region is stillpoorly known because of the lack in the pastof geological or geophysical studies. Someevidences of active faulting have been pointedout by previous studies (Meghraoui, 1988;Boudiaf, 1996).

Seismotectonic settingThe Northern region of Algeria is locatedalong the western part of the plate boundarybetween Eurasia and Africa. This boundary isunder compression due to the convergence of

3 September 2003

Figure 1: Boumerdes, Algiers (Algeria) earthquake (May 21st, 2003) location.

Figure 2: Building damages observed inthe city of Boumerdes

The Boumerdes -Algiers (Algeria) Earthquakeof May 21st, 2003 (Mw=6.8)

A.K. Yelles-Chaouche, H.Djellit and M. HamdacheCRAAG, B.P. 63, Bouzareah Algiers, Algeria

the two major plates with a calculated rate of0.6 cm/yr (Argus et al., 1989). This led to theoccurrence, in this region, of repeatedmoderate to large earthquakes, one of themost important earthquake was the El-Asnamearthquake of October 10th, 1980 (Mw=7.3).

The region of Boumerdes is located, along thecoastal region of the central part of Algeria. Itis located, more precisely in the eastern tip ofthe Mitidja basin (figure 1). The latter wasformed during the N-S Miocene extension(Guiraud, 1977; Philip, 1983). This extensionwas followed by a N-S to NNW-SSEcompression, which peaked, between UpperPliocene and Villafranchian (Guiraud,1977).The compression has continued during theQuaternary (Philip and Thomas, Meghraoui,1988) and is still active as shown by theobserved seismic activity and recentdeformation. This deformation is representedby active folding oriented NE-SW and bythrusts and strike slip faults trendingrespectively NE and SE (Meghraoui,1988 ;Boudiaf,1996).

In the Boumerdes region, the quaternarygeological formations are marked by twoseries. The first one related to the sandyCalabrian and Villafranchian formations andthe second to the quartzitic Thyrrhenianformations. In these later series, severaluplifted marine terraces at differenttopographic levels are distinguished. Theseseries are deformed and associated with foldswith a N120 orientation (from Cap Matifou toThenia). From previous geologicalinvestigations (Boudiaf, 1998), thistopographic height corresponding to theeastern flank of the Sahel region is associatedto the Thenia fault.

Coseismic surface breaksCoseismic surface breaks were observedonshore in several areas of the epicentralregion as Boudouaou, Zemmouri port andDellys (figure 3). They affect the

Thyrrhenian marine terraces of the easternlittoral of Algiers. Surface breaks correspondto a clear set of cracks with a NE-SW (N35°)orientation along the Zemmouri bay. Alongthe Tamentefoust-Thenia, axis, breaks withN120 direction affects the marine terraces.

Offshore, an uplift of the seafloor has beenobserved, marked by the appearance at thesea surface of small rocky blocks (figure 4).In fact, this uplift concerns the whole areaaround Boumerdes as this phenomenon hasbeen observed along the coast, ranging from0.4 m at Reghaia beach to 0.8 m at Zemmouriembankment port.

In addition, the earthquake induced someminor landslides as observed along thesandy dunes of the beaches or along theroads (Figure 5). Some liquefactionphenomenon was also observed. It would beresponsible of the destruction of severalbuildings. Some other effects as creation ofnew water springs has also accompanied theearthquake.

The Zemmouri fault From the geophysical and geologicalinvestigations, we can assume that theBoumerdes Algiers earthquake has beengenerated by a new revealed offshore faultcrosscutting the margin. It is oriented N 35 andextents from Dellys to Corso (about 40 km)(Figure 1). This fault allowed the uplift of thecoastal region. This fault is associated with theThenia onshore fault that appears as a majoractive structure in this area (Boudiaf, 1996,1998). The evidence of folding and faulting inthe region of Boumerdes are now wellunderstood by the offshore Zemmouri fault.

Evaluation of the seismic hazardThe occurrence of the Boumerdes earthquakedrastically changes the evaluation of theseismic hazard of the Algiers region. Indeed,before the earthquake, the maximum PGAwas evaluated 0.15g. This is explained by thepoor knowledge of the seismicity of the region.

Using the earthquake design scenario,assuming that the maximum possiblemagnitude is the magnitude of earthquake at20 km ± 5 km from Algiers or Boumerdes. Themaximum PGA is now:

Discussion and ConclusionsThe Boumerdes earthquake demonstrates thatnorthern region of Algeria remains highlyactive. Indeed, during the last twenty years,this region experienced six moderate events(Constantine, October, 27th, 1985, Ml=6.0;Tipaza, October, 29th, 1989, Ml=6.0 ; Mascara,August, 18th, 1994, Ml=5.7; Ain Benian, (Algiers)September, 4th, 1996, Ml=5.8; Ain Temouchent(Oran region), December, 22th, 1999, Ml=5.8 ;Beni Ouartilane (Bejaïa region), November,10th, 2000, Ml=5.4) and two importantearthquakes (El Asnam of October 10th, 1980,Mw=7.3 and Boumerdes earthquake of May, 21,2003). All these earthquakes are more preciselylocated in the Tellian mountains which remainsthe more active zone of Algeria. This regioncould be the location of some violentearthquakes of the western part of the

G-K=0.3172g,Asup

Pi=0.2352g,Asup

K-S=0.2367gAsup

CSEM / EMSC Newsletter

4

Figure 3: Surface breaks in the epicentral area. (a) general view near Corso.(b) details of the thrust.

Figure 4: Uplift of the seafloor observed along the coastline.

5 September 2003

IntroductionThe regions of Algiers and Boumerdes, Northof Algeria, were shock by a destructiveearthquake of magnitude Mw=6.8 (EMSC/USGS)on Wednesday 21st, May 2003 at 19:44:40(GMT+1) which claimed about 2300 humanlives and injured about 11000 peoples. It’stherefore, the worst seismic event since the ElAsnam October 10th, 1980, Ms=7.3.

The main shock caused widespreaddestructions mainly in the area of the Easternpart of Algiers, the capital, and Boumerdes,about 50 km NE of Algiers. The main shockwas felt as far as 250 km from the epicentrewhich was located offshore.

Many records were made by the nationalaccelerograph network monitored by CGS. Themaximum pick ground acceleration recordedat 20 km from the epicenter reached 0.58 g.This earthquake induced liquefaction mainlyalong the Isser and Sebaou river valleys,tsunamis in the Balearic islands, and siteamplification mainly in the quaternaryMitidja basin, as well as uplift of the coastalline. The main shock was followed by severalaftershocks among them those of May 27th,2003 (Ms=5.8) and May 28th, 2003 (Ms=5.8)causing panic among the population whichfled their homes, few people left Algiers forother regions of Algeria. The earthquake of

May 27th, 2003 injured about 100 people, two ofthem threw themselves from windows, killed 3and destroyed the mostly affectedconstructions, particularly the R+10 buildingat Reghaia and the minaret of Zemmourimosque.

Strong ground motion dataanalysisIn addition to the epicentral and surroundingareas, the earthquake was felt in a radius ofmore than 250 km where it was recordedaccelerations of about 0.02 g. The main shockwas recorded by many stations of the nationalaccelerograph network monitored by CGS.

Mediterranean region and the Boumerdesearthquake is among these major events.

The Boumerdes earthquake remains the firstmajor seismic event in the Algiers region sincethe February, 3rd, 1716 earthquake. If weconsider the seismic destructive event of theJanuary 2nd, 1316, we can consider that theAlgiers region is repeatedly affected by majorstrong shakings.

The Boumerdes offshore earthquake is thethird recent seismic event, which happenedalong the Algerian margin. Its destructivecharacter indicates the urgent need ofinvestigation on the active structures of theoffshore region of Algeria whose recent tectonicis still poorly known.

The next Maradja French Algerian surveywhich will be carried out in August in thisregion will be an excellent opportunity toimprove the knowledge of this region and toextent the onshore seismotectonic map. On theother hand, results of this survey will allow tobetter understanding the uplift mechanism ofthe Algerian margin due to the compressionalstress regime of the region.

References:Ambraseys, N. N. and Vogt, J., 1988- Material forthe investigations of the seismicity of the regionof Algiers (Algeria). European EarthquakeEngineering, 3, pp. 16-29.

Argus, D. F., Gordon, R. G., Demets, C. andStein S., 1989- Closure of the Africa-EurasiaNorth America plate motion circuit andtectonics of the Gloria Fault. J.G.R.,94, 5585-5602.

Benhallou, H., 1985- Les catastrophessismiques de la région d'Ech Cheliff dans lecontexte de la sismicité historique de l'Algérie.Université des Sciences et de la Technologie,Houari Boumediène, Institut des Sciences dela Terre, Alger, 05 octobre 1985.

Boudiaf, A., 1996- Etude sismotectonique de larégion d’Alger et de la Kabylie. Thèse dedoctorat de l’université de Montpellier(France), May 1996.

Boudiaf, A., Ritz, J. F. and Philip, H., 1998-Drainage diversions ad evidence ofpropagating active faults : example of the ElAsnam and Thenia faults, Algeria. Terra Nova,Vol 10, N°5, 236-244.

Meghraoui, M., 1988- Géologie sismique duNord de l’Algérie: paléosismologie, tectoniqueactive et synthèse sismotectonique. Thèse dedoctorat es-sciences, Université de Paris 11,France.

Meghraoui, M., 1986- Blind reverse faultingsystem associated with the Mont Chenoua-Tipaza earthquake of 29 October 1989 (north-central Algeria). Terra Nova, 3, pp. 84-93.

Philip, H., 1983 - La tectonique actuelle etrécente dans le domaine Méditerranéen et sesbordures, ses relations avec la sismicité. Thèsede Doctorat de l'Université des Sciences etTechniques du Languedoc présentée le 15décembre 1983.

Philip, H. and Thomas, G.-1977-Détermination de la direction deraccourcissement de la phase de compressionquaternaire en Oranie (Algérie). Rev. Géograp.Phys et Géol. Dynam., (2), vol. XIX, fasc.4, pp.315-324.

Rothe, J. P. -1950- Les séismes de Kherrata etla sismicité de l’Algérie. Bulletin de la Soc.Histo. Nat. Af. Nord. 3-4, 211-228.

Figure 5: Landslides along the coastline in the epicentral area.

Analysis of Strong Ground Motions Recorded during the 21st May, 2003Boumerdes, Algeria, Earthquake

Nasser Laouami, Abdennasser Slimani, Youcef Bouhadad, Ali Nour and Said LarbesCGS, National Center of Applied Research in Earthquake Engineering, 1 Rue Kaddour Rahim, Hussein Dey, B.P. 252, Algiers, Algeria,

E-mail : n_laouami@hotmail.com

CSEM / EMSC Newsletter

6

A temporal and frequential analysis wasperformed (Laouami et al., 2003).

Figure 3 show the map of the stations locationwith the pick ground accelerations recordedduring the main shock in the E-W, N-S andVertical directions.

The closest stations to the epicentre, in freefield, which recorded the main shock are thoseof Keddara at 20 km. From figure 1, one canunderline the following observations:

• At Keddara site, the maximum accelerationsrecorded by two stations distant from eachother approximately 150 m are as follows:

Station 1: E-W: 0.34 g Ver: 0.25 g N-S: 0.26 g Station 2 : E-W : 0.58 g Ver : 0.22 g N-S : 0.35 gThe very significant variation observedbetween stations 1 and 2 concerning moreparticularly accelerations in E-W directionsuggest the presence of site effectparticularly significant at station 2. Acurrent geotechnical study will quantify thislocal site effects.

• PGA in E-W direction are more significantthan those in the N-S direction.This observation is valid almost at themajority of the stations, and is probablyrelated to the directivity effect of the fault.

• Because they are located on the Mitidjaquaternary basin classified as soft soils, TheDar El Beida and El Afroun stations recordsshow high acceleration level compared tothose located on firm soils.

Figure 2 shows the E-W, N-S and Verticalaccelerations recorded at Keddara 1 station.The strong part duration given by the Husiddiagram is about 10 s.

CGS

Long: 3.53 E; Lat:36.81N

3 ° 00 ' 3 ° 30 ' 4 ° 00 ' 4 ° 30 ' 5 ° 00 '2 ° 30 '2 ° 00 '36 ° 00 '

36 ° 30 '

37 ° 00 '3 ° 00 ' 3 ° 30 ' 4 ° 00 ' 4 ° 30 ' 5 ° 00 '2 ° 30 '2 ° 00 '

Digital stations ETNA Digital stations SSA-1 Analogue stations SMA-1

AZAZGAE-W:0.12gV :0.05gN-S:0.09g

TIZI OUZOUE-W:0.20gV :0.09gN-S:0.19g

HUSSEIN DEYE-W:0.27gV :0.09gN-S:0.23g

KOUBAE-W:0.31gV :0.15gN-S:0.16g

DAR EL BEIDAE-W:0.52gV :0.16gN-S:0.46g

KEDDARA (1)E-W:0.34gV :0.25gN-S:0.26g

KEDDARA (2)E-W:0.58gV :0.22gN-S:0.35g

TIPAZAE-W:0.03gV :0.014gN-S:0.03g

BLIDAE-W:0.05gV :0.03gN-S:0.04g

EL AFROUNE-W:0.16gV :0.03gN-S:0.09g

MELIANAE-W: 0.03gV : 0.016gN-S: 0.026g

HAMMAM RIGHAE-W:0.10gV :0.06gN-S:0.07g

Medea

Blida

Tipaza

Algiers Boumerdes

Tizi Ouzou

Mediterranean sea

Figure 1: Location of the recording stations and the corresponding recorded PGA in the E-W, N-S and Vertical directions. The star indicates the epicentre

determined by CGS using accelerograms : Long: 3.53 E, Lat: 36.81 N

0 5 10 15 20 25 30 35 40-400

-300

-200

-100

0

100

200

300

400

E-W Component

Time (sec)

Acc

eler

atio

n (c

m2 /

s)

0 5 10 15 20 25 30 35 40-400

-300

-200

-100

0

100

200

300

400

N-S Component

Time (sec)

Acc

eler

atio

n (c

m2 /

s)

0 5 10 15 20 25 30 35 40-300

-200

-100

0

100

200

300

V Component

Time (sec)

Acc

eler

atio

n (c

m2 /

s)

Figure 2: E-W, N-S and Vertical accelerations recorded at Keddara 1 station.

1 10 1000

1

2

3

4

5

N-S Component

Frequency (Hz)

1 10 1000

1

2

3

4

5

V Component

Frequency (Hz)

1 10 1000

1

2

3

4

5

E-W Component

Frequency (Hz)

Pse

udo

-Acc

eler

atio

n S

pec

tra

Pse

udo

-Acc

eler

atio

n S

pec

tra

Pse

udo

-Acc

eler

atio

n S

pec

tra

Figure 3: E-W, N-S and Vertical pseudoaccelerations response spectra

at Keddara 1 station.

0 20 40 60 80 100 120 140 160

0

0.1

0.2

0.3

0.4

0.5

0.6Recorded PGA Ambraseys, 1995, mean Ambraseys, 1995, mean + std

Peak Ground Acceleration (g)

Hyp

oce

ntra

l dis

tanc

e (k

m)

Figure 4: Comparison between the Recorded PGA attenuation with the mean and the meanplus std empirical Ambraseys attenuation law (1995).

7 September 2003

Response spectraThe other parameter which characterizes theseismic movement is the frequency content.The study performed on the recordingsobtained during the main shock may explainsthe observed damages in the epicentral region.

Figure 3 shows the response spectra of theE-W, NS and Vertical components for thestation of Keddara1. The plotted curves showhigh frequency content for the Keddarastation (near field). On the other hand, themost damaged buildings in the epicentralregion have a fundamental frequency close tothe earthquake frequencies.

Attenuation lawA comparative study is carried out betweenthe recorded data during the main shock of theMay 21st, 2003 Boumerdes earthquake withthe attenuation law of Ambraseys (1995),developed on the basis of a sample of 1260seismic records generated by 619 shallowearthquakes including Algerian data. Figure4 shows the recorded maximum groundaccelerations (E-W component) and theAmbraseys (1995) empirical curves. From theplotted curves, one can deduce the followings :

• The mean curve of the Ambraseys lawunderestimate largely the recordedaccelerations for all distances.

• The envelop curve of Ambraseys lawunderestimate the recorded acceleration upto 70 km, while it captures the trend of therecorded accelerations from 70 km.

It is clear that for certain accelerations(stations of Keddara, Dar El Beida, El Afroun),the amplification effect explains thesignificant variation with the empirical curvesof Ambraseys, 1995.

ConclusionThe earthquake which shook the areas ofAlgiers and Boumerdes (Mw = 6.8, EMSC)on May 21st, 2003, is seen as the majorearthquake which has occurred in this areaof Algeria for at least two centuries, and thesecond of this importance after El AsnamOctober 10th, 1980 (Ms = 7.3) earthquake.The main shock was felt 250 km far fromthe epicentre, and was recorded by severalaccelerograph stations of the nationalnetwork monitored by the CGS.

Analysis of the recordings of the strongmovements allows to conclude what follows :

I A significant variation observedbetween two stations distant from eachother approximately 150 m. A currentgeotechnical study will quantify thislocal site effects.

II Pick Ground Acceleration in E-W directionare more significant than those in the N-Sdirection. This observation is probablyrelated to the directivity effect of the fault.

III Stations located on the Mitidja quaternarybasin classified as soft soils such as theDar El Beida and El Afroun stationsrecords show high acceleration levelcompared to those located on firm soils.

IV A comparative study shows that theattenuation law of Ambraseys (1995)underestimates largely the recorded dataduring the main shock of the May 21st,2003 Boumerdes earthquake.

On the other hand, this earthquake inducedliquefaction mainly along the Isser andSebaou river valleys and tsunamis in theBalearic islands as well as uplift of the coastalline.

References:Ambraseys N. N. 1995. The prediction ofearthquake peak ground acceleration in Europe.Earth. Eng. Struct. Dyn. 24, 467-490.

Laouami N., Slimani A. Bouhadad Y. and Nour A.2003. The 05/21/2003 Boumerdes Earthquake :Preliminary analysis. Intern Report. CGS.

The last official report concerning theconsequences of the May 21st 2003 earthquakein Algeria indicates 2 278 deaths, more than 10000 injured and about 180 000 homeless.

Concerning the buildings, about 7 400 of themhave been destroyed and about 7 000 othershave been heavily damaged in the wilaya(administrative region) of Boumerdes; in thewilaya of Algiers 8500 apartments were lost andmore of 20 000 others were heavily damaged.

Damage assessment andanalysis More than 500 experts from CGS, C.T.C, OPGI,local technical administrations, engineeringoffices, etc. have been mobilised following arequest by the Ministry of Housing to rapidlyassess the damage. This is the first step torapidly provide a shelter to the homeless andto ensure the new school year next Septemberin adequate conditions of security

Damage assessment form A form containing approximately 60 differentdata has been used by the teams on the field toclassify the damaged buildings. The same formhas been in use since 1980 and it ensureshomogeneous damage evaluations. It includes

information on the type of building (e.g.,building identification, building occupancy,number of levels, etc.... ), soil conditions, anassessment of the structural and nonstructural elements damage, commentaries onthe damage causes, global building damagelevel and the recommendations of emergencymeasures if necessary. Five levels of damage

are identified and buildings are classified in 3categories:Green category (Damage levels 1 and 2): Nodamage or slight damage. The building can bereoccupied immediately.Orange category (Damage levels 3 and 4):Significant damage requiring an extensiveexpertise.

A Short Note on Building DamageMohamed Belazougui, Mohammed N. Farsi, Abdelkader Remas

National Earthquake Engineering Center (CGS), Algiers

CONSTRUCTION USEGREEN ORANGE RED

TOTALNIV. 1 NIV. 2 NIV. 3 NIV. 4 NIV. 5

Residential buildings 18130 32352 19343 11727 10183 91735

Administrative buildings 213 300 184 76 52 825

Schools 420 814 467 286 103 2090

Hospitals 94 114 44 23 10 285

Sportive and cultural buildings 106 97 90 87 32 412

Commercial buildings 189 193 140 82 137 741

Industrial facilities and hangars 85 153 98 73 66 475

Other (water tanks, etc…) 54 112 110 74 61 411

TOTAL 19291 34135 20476 12428 10644 96974

%19.90 35.20 21.11 12.82 10.97 100

55.10 33.93 10.97 100

Table 1: Level of damage in wilayas of Algiers and Boumerdes

CSEM / EMSC Newsletter

8

Red category (Damage level 5): Constructionspartially or totally collapsed.

Damage distributionUntil the end of June, 96974 buildings haveundergone damage assessment; among them91735 (94.6%) are residential buildings. Table1 gives an overview of the level of damage.

Causes of damage The three main causes of the damage are themagnitude of the event (Mw6.8, EMSC), theimportant urbanisation close to the epicentrebut also to insufficiencies in design andconstruction of the buildings. Concerningthese insufficiencies, one can mention thefollowing points:

• Poor conceptual design (buildingconfiguration and structural layout) forseismic resistant buildings.

For example, the presence of short columns in“vides sanitaires” (crawl spaces) and betweenupper windows of classrooms (Figure 1); thepresence of soft storeys, generally at the groundzero (first storey) used as car parks or stores, withlack of infill and/or height greater than for upperstoreys (Figure 2); the use of many heavyornamental elements at building facades;insufficient dimension of the seismic joint(between two blocks of the same building, orbetween two buildings), Use of important andheavy cantilevers; Use of irregular buildingconfigurations with severe discontinuities inmass, stiffness, strength, and ductility resultingin torsional effects and stress concentration; Useof “heavy girders / weak columns”; Use of veryheavy roofs; Non tied (or non reinforced) masonry(in rural or old urban buildings) (Figure 3)

• Lack of structural design or under-dimensioning (non seismic resistantdesign)

This is the case of all the buildings aged before1981 and the great majority of individual orprivate houses built after 1980.

• Poor quality of execution and poorquality of structural material

One can note the low strength of concrete(average of 14-17 MPA instead of 25 MPArequired by the standards for currentbuildings) and the inadequate restarts ofconcrete pouring in columns (Figure 4).

• Inadequate proportioning anddetailing of structural elements(Figure 5).

• Poor inspection and constructiontechniques

• Inadequate building maintenance

Concluding remarksThis short notes only intends to presentpreliminary analysis of damage following theBoumerdes earthquake. More detailed studieswill be made available to the community assoon as possible.

The aim of this article is to present anddiscuss the source process of May 21st, 2003Boumerdes earthquake (Mw=6.9) obtained bymodelling broadband seismograms recorded atteleseismic distance by the global seismicnetwork. In the absence of near sourcemeasurements, remote seismological data canprovide the information needed to retrieve the

main characteristics of the rupture process,provided that some care is dedicated to theexploration of the model space. Robustnesscan be gained by combining body wavestogether with surface waves.

After we determined the focal mechanism ofthe event (strike 57°, dip 39°, rake 83°), weinvestigated the slip distribution for the

southeast dipping nodal plane whichcorresponds most likely to the fault planeaccording to the tectonic data. We performedseveral waveform inversions to evaluate thestability of the slip pattern. The most stablefeatures of the rupture process are thefollowing: 1) the essential part of the ruptureoccurred at shallow depth (< 10 km), 2) the

Figure 1: Short column

Figure 2: Soft (first) storey

Figure 3: Non tied stone masonry Figure 4: Bad quality of materials

Figure 5: Non adequate detailing

The 2003 Boumerdes (Algeria) earthquake:Source process from teleseismic data

Bertrand Delouis1 and Martin Vallée2

1 Géosciences Azur, CNRS/UNSA, Nice — Sophia Antipolis, France (delouis@geoazur.unice.fr)2 Laboratoire de Détection Géophysique (LDG-CEA), Bruyeres-Le-Chatel, France (vallee@dase.bruyeres.cea.fr)

9 September 2003

hypocenter itself was relatively shallow, bestlocated around 10 km depth, and 3)consequent slip occurred southwest of thehypocenter, as indicated by a markeddirectivity effect. A second order feature,required nonetheless by the data, is thatsome slip propagated toward the northeast.

The model shown in Figures 1 and 2represents a median solution for the slipdistribution and displays the most stablefeatures. This model fits optimally the P, SH,and apparent source time functions derivedfrom the surface waves (see web page of theESMC-CSEM, http://www.emsc-csem.org/).Maximum slip in the model is 170 cm and islocated updip of the hypocenter. At thehypocenter itself, slip amounts to 130 cm. Themain directivity effect toward the southwestis accounted for by a shallow asperity located25-30 km southwest of the hypocenter. There,slip reaches 130 cm also. Propagationtowards the northeast is related to theexistence of a shallow slipping area 25-30 kmaway of the hypocenter, where slip reachesabout 60 cm. The total seismic moment (2.381026 dyne.cm) indicates a moment magnitude(Mw) of 6.9. The total rupture length andduration are 50 km and 15 s respectively.Details concerning the methodology that weused can be found in Delouis et al. (2002) andVallée (2003).

Yagi (2003), modelled the teleseismic P wavesalone and obtained similar results but withhigher maximum values of slip (~ 230 cm)(http://iisee.kenken.go.jp/staff/yagi/eq/algeria20030521/algeria2003521.html).Using a grid search approach, we explored themodel space and found that the optimal values(from the data fit point of view) for maximumslip may vary between 100 and 250 cm.

An important issue is the absolute location ofthe fault plane activated during theearthquake. It cannot be constrained byteleseismic data alone. Geodetic data(levelling, GPS, SAR) and strong motionrecords, when available, will help to locate therupture plane and to constrain further somedetails of the rupture process. The absolutelocation of the hypocenter may be used, but inthe absence of seismic stations very close tothe event, the uncertainty may be relativelylarge. Observations of coastal uplift exceedingseveral tens of cm near Boumerdes and on thePeninsula of Ain Taya (Meghraoui, pers.comm.) indicate that the rupture should belocated very close to the coast. On Figure 1, wehave chosen a location of the rupture whichwould be compatible with those observationsand with the location of a major thrust faultimaged on deep seismic profiles (Mauffret,pers. comm.). We conclude that the maincharacteristics of the rupture process of the

Boumerdes earthquake are well established,although some uncertainties remain regardingthe maximum slip amplitude and the absolutelocation of the activated fault. Near sourcedata (geodetic, strong motion) will help tofurther resolve these issues in the very nearfuture.

References:Delouis, B., Giardini, D., Lundgren, P., and J.,Salichon, 2002. Joint inversion of InSAR, GPS,teleseismic and strong motion data for thespatial and temporal distribution of earthquakeslip: Application to the 1999 Izmit Mainshock,Bulletin of the Seismological Society of America,92, 278-299.

Vallée, M., 2003. Stabilizing the empiricalGreen's funcyion analysis: development of theprojected Landweber method, in revision,Bulletin of the Seismological Society of America.

50km

Algiers

3030 6060 9090 130130 170170 210210

slip (cm)slip (cm)

Boumerdes

Ain Taya

Figure 1: Slip map projected onto the surface. Black arrows indicate the slip direction. Thefocal mechanism is from this study. Rupture initiate at the hypocenter indicated by the opentriangle. Absolute positioning of the fault is not constrained by the teleseismic data used in

this study, but is chosen according to independent observations (see text).

alon

gdip

(km

)

alongstrike(km)

*3.0 10dyne.cm/s

time

25

0

STF

10s

a)

b)

t=6s

t=3s

t=9s

t=12s

t=15s

SW NE

Figure 2:a) Moment rate Source Time Function (STF).b) space-time evolution of the rupture repre-

sented by the cumulative slip at differenttimes after rupture initiation (hypocenter

indicated by the open triangle).

CSEM / EMSC Newsletter

10

ContextA strong Mw 6.9 earthquake occurred on 21May 2003 on the Algerian coast, near the cityof Boumerdes, about 60 km east of Algiers. Theepicenter has been rapidly located offshore,and has been subsequently confirmed by thelack of direct observations of grounddeformation on the coast, and by the offshorelocation of the following aftershocks. Theshallow depth (10 km at most) could partly beresponsible for the important damage due tothe earthquake on the buildings and the highdeath toll (about 2500 casualties). Thisearthquake belongs to a series of severalthrusting events that have frequently struckthe Northern Algerian coast, in response to theconvergence between the African andEurasian plates.

Observations of the tsunamiFor this kind of submarine shallow thrustingearthquake a tsunami is to be expected,however the level of magnitude was not proneto trigger a high-energy tsunami. Indeed,reports of sea disturbances along the Algeriancoasts have been rather poor so far, and thecoseismic uplift of the coast has been morefrequently evoked by the eyewitnesses. Themainshock together with highly probablesubmarine slides seem to have cut severalunderwater telephonic cables, causingsignificant trouble to establishcommunications with Algeria in the daysfollowing the earthquake.

A few tide gauges have recorded sea levelvariations around the Mediterranean Sea, asin Genoa (Italy) (Figure 1 A) where amplitudeshardly exceed 5 to 8 cm, and in Nice (France)(Figure 1 B) where 10 cm amplitudes areobserved. Arrival times can be roughlyestimated to GMT 8 h 40 p.m. in Genoa(corresponding to an about 2 h propagationduration) and GMT 8 h 20 p.m. in Nice (aroughly 1 h 40 min propagation).

Sea disturbances have been clearly observedon the southeastern coast of the BalearicIslands, located about 250 km to the north, inMajorca and Minorca Islands, as well as in

The tsunami triggered by the 21 May 2003 Algiers earthquakeHélène Hébert,

Laboratoire de Détection et de Géophysique, CEA, BP 12, 91680 Bruyères-le-Châtel, FrancePierre-Jean Alasset,

Tectonique Active, EOST, 5, rue René Descartes, 67084 Strasbourg Cedex, France

Nice's tide gauge Courtesy of SHOM France

0

10

20

30

40

50

60

21/05/200312:00

21/05/200315:00

21/05/200318:00

21/05/200321:00

22/05/200300:00

22/05/200303:00

22/05/200306:00

22/05/200309:00

22/05/200312:00

Time

Sea

leve

l (cm

)

Observed TidePredicted Tide

18h44: Earthquake occured

20h20: Beginning of tide's signalpertubations

Palma de Mallorca's tide gauge Courtesy of IEO Spain

0

3,2

3

2,8

2,6

2,4

2,2

2

1,82 4 6 8 10 12 14 16 18 20 22 0 2 4 6 8 10 12 14 16 18 20 22

Time 21/22 May 2003

Sea

leve

l (m

) 19h35 : Beginning of tide's signalpertubations

-1.0

-0.5

0.0

0.5

1.0

sea

leve

l (m

)

-1.0

-0.5

0.0

0.5

1.0

sea

leve

l (m

)

960 1200 1440 1680minutes

16h GMT 20h GMT 0h GMT 4h GMT

Mainshock

Genova’s tide gauge Courtesy of APAT Italy

-0.05

0

0.05

0.1

0.15

0.2

0.25

20/05/200300:00

20/05/200312:00

21/05/200300:00

21/05/200312:00

22/05/200300:00

22/05/200312:00

23/05/200300:00

Time

Sea

leve

l (cm

18h44: Earthquake

occured

20h40: Beginning of tide's signal pertubations

Figure 1: Sea level variations recorded in A)Genoa (Italy, courtesy of APAT, Italian Agencyfor the protection of environment and thetechnical services), B) Nice (France, courtesyof SHOM, French Hydrographic Service), andC) Palma (Majorca, Balearic Islands, courte-sy of IEO, Spanish Oceanographic Institute).For the latter, data have been filtered fromthe tide signal using a high-pass filter. Theresulting data show that the first arrival inPalma occurs about 50 min after the main-shock, and 20-min periods are observed.

11 September 2003

Ibiza. Witnesses have reported waves up to 2m high, and a mean observed period of 10-12min in Majorca. The consequences were quiteimportant in Minorca (especially in Maoharbour, located SE of the island) where all inall about 10 boats sunk and several other tenswere seriously damaged.

While the previous tide gauges in Genoa andNice do not record sea level variations at abetter rate than 1 point every 10 min, the tidegauge in Palma (Majorca, Balearic Islands)provides a very valuable data set sampled at a1-min rate (Figure 1 C), that agrees well withthe testimonies reported above. A maximumwater height of about 70 cm (or 1.2 m peak-to-trough amplitude) has been recorded, withperiods of about 20 min. The first arrival isobserved about 50 to 60 minutes after themainshock in the epicentral area, at aboutGMT 7 h 30 pm.

These data together with the reports gatheredin the other harbour communities confirmthat a tsunami was generated by the 21 May2003 Algiers earthquake, and that the seawaves have propagated across the westernMediterranean Sea. They have causedmoderate material damage in the BalearicIslands. For this magnitude level, thesignificant tsunami was quite a surprise. Thepossibility for one or several submarinelandslides should not be discarded because ofthe reported submarine cable breakings,however these slides may have been unable toproduce a tsunami.

Preliminary modeling of thetsunamiIn a first attempt we test here a source due toa coseismic seafloor deformation that wederived from preliminary seismologicalstudies of the source. The initial seafloordeformation has been first computed from theparameters of the rupture taken from theHarvard CMT solution (Table 1). Using theanalytical formula describing a simple elasticdislocation model (Okada, 1985), we obtain amean deformation on the fault plane of 0.85 m,and the initial positive deformation computedon the seafloor amounts to 0.3 m at most.

This is undoubtly lower than the valuesobtained from a recent seismological inversion

performed using body and surface waves(Vallée, personal communication; Delouis etal., 2002; Vallée, 2003), that leads to adeformation reaching 1.7 m locally on the faultplane, while the maximum seafloordeformation reaches 0.9 m locally. The depthparticularly is much shallower than the onewe tried here first.

To deal with these first seismological studies ofthe Algiers earthquake, we slightly modifiedthe fault parameters and used a shallowerdepth (10 km) as well as a seismic momentcorresponding to a moment magnitude Mw 6.9.With these values, the initial deformation isslightly higher than 0.4 m on the seafloor. Thismodel does not account for the local

2°30’

2°30’

3°00’

3°00’

3°30’

3°30’

4°00’

4°00’

4°30’

4°30’

36°30’

37°00’

37°30’

-2000-1000

21 May 2003 mainshock

50 km

-0.50

-0.20

-0.10

-0.05

-0.02

0.02

0.05

0.10

0.20

0.50m

Figure 2: Initial seafloor deformation computed from the Harvard CMT parameters modified as explained

in the text.

0 4 8

36

40

44 T = 10 minNice

Genoa

Palma

0 4 8

36

40

44 T = 20 min

36

40

44 T = 40 min

36

40

44 T = 50 min

0 4 8

36

40

44 T = 80 min

0 4 8

36

40

44 T = 100 min

-30 -5 -3 -2 -1 0 1 2 3 5 45

cm

Figure 3: Snapshots of the propagation of the tsunami, with bathymetric contours plotted at 1000m intervals. The first leading wave approaches the Balearic Islands after about 20 min, where theshallower water depth slows down the propagation. Consequently the first wave reaching Palmadoes not arrive before 45 min, in rough agreement with the tide gauge record. In Nice and Genoa,

the simulation shows leading waves arriving after 80 to 100 minutes, and then providesminimum values for the actual arrival times that could be refined through detailed modeling.

Harvard CMT solution modified modelFault dimensions (km x km) 40 x 20 40 x 20

Strike, dip, rake 56°, 46°, 71° 56°, 46°, 71°

Seismic moment (Nm) - Mw 0.201 1020 – 6.8 0.239 1020 – 6.9

Rigidity (Nm2) 30 109 30 109

Epicenter 3.52°E - 36.98°N 3.52°E - 36.98°N

Depth (km) 17 10

Table 1: Parameters used for the tsunami modeling, taken and then modified from theHarvard CMT solution.

CSEM / EMSC Newsletter

12

heterogeneities of the slip pattern, animprovement that should be considered infurther modeling.

Then we assume that the initial seafloordeformation is fully and instantaneouslytransmitted to the sea surface, where, throughrestoring gravity forces, tsunami waves begin topropagate across the sea. Since tsunamiwavelengths are much larger than the meanwater depth, we assume that the long wavetheory is valid. The simulation of thepropagation is performed using a finitedifference method that solves the hydrodynamicequations of continuity and motion, includingnon-linear terms.

Bathymetric data are required to solve theseequations. Under the shallow-waterassumption, and since the amplitude is muchsmaller than the water depth, the wave velocitycan be given by c = (gh)1/2. To account for thewave amplification near the coasts, we need torefine the bathymetric data used to compute thepropagation of the tsunami waves. We used datataken from the global bathymetry data deducedfrom altimetry (compiled by Smith and Wessel,1997), to build a 2' grid describing the westernMediterranean Sea, and a 30'' grid for theBalearic Islands. We did not refine here ourmodelings in the bays of the Balearic Islands, forthe compilation of detailed bathymetric data (inthe Balearic Islands as well as in Algeria) is nowin progress and deserves a more comprehensivestudy.

Figure 3 shows several snapshots of thetsunami propagation. The tsunami waves slowdown towards the eastern Spanish marginwhere the water depth decreases, and in theBalearic Islands where the first arrival inPalma does not occur before 45 to 50 min.We didnot try to match the tide gauge data from Palmasince the bathymetric grid used here is notdetailed enough to deal with the shoaling effect.In Nice and Genoa, the modeling shows arrivaltimes of 80 to 100 minutes, yielding minimum

values that could better agree with theobservations through an another refinedmodeling. Figure 3 also shows that, after theBalearic Islands, the Sardinia Island may havebeen struck by the tsunami as well, especiallyalong its southwestern coasts. No observationshave been however reported on these poorlyinhabited coastal areas.

Finally the maximum water heights recordedduring the propagation at each grid point revealhow the tsunami energy was radiated in theMediterranean (Figure 4). The maximumenergy is perpendicular to the fault strike, inagreement with the analysis or radiationpattern (e.g. Okal, 1988) and tsunami modelings(Hébert et al., 2001). For the Algiers tsunami,our results show that the westernmost island inthe Balearic archipelago (Ibiza) was moreexposed to the tsunami, with respect to Minorca,the easternmost one. Several bays in theBalearic (and especially Mao harbour inMinorca) are however known to be extremelysensitive to storm sturges and othermeteorological sea level perturbations (e.g.Campos, 1991), and a specific configuration ofthe bay may also account for the disturbancesobserved.

Concluding remarksThis preliminary modeling shows that acoseismic source can explain a focusing of thetsunami energy toward the Balearic Islands,especially along their SE coasts, as well astsunami waves observed further to the North, inNice and Genoa. Our first results do not providenew significant facts to describe the behavior ofthe waves along the Algerian coasts. We observethat the slopes offshore Algeria are extremelysteep, probably favoring a reflexion rather thanan amplification of the waves. Nevertheless it isworth noting that these steep slopes could alsobe at the origin of submarine slope failures thatare known to provoke tsunamis prone to belocally devastating (e.g. Heinrich et al., 2001).

Historical tsunamis observed on Spanish coastshave been compiled into a catalogue (CamposRomero, 1992; Carreño et al., 1998) that showsfrequent inundations on the southern coasts ofthe Iberian peninsula, occasionally in theBalearic. Among these tsunamis having reachedsouthern Spain, hardly one event originatingfrom Algeria is clearly identified. However seadisturbances reported on the southern Spanishcoasts after the 1954 Orleansville and the 1980El Asnam earthquakes (Campos, 1991), andobservations gathered after the late 2003earthquake indicate that earthquakes occurringon the Algerian margin can create tsunamisable to propagate across the westernMediterranean Sea and generate moderatedamage up to 200 km away from the source.

Regarding local tsunami hazard in Algeria, apreliminary investigation has stressed theoccurrence of historical tsunamis apparentlylinked to earthquakes, in 1365, 1856 and 1891(Yelles Chaouche, 1991). The 1856 tsunami,

following the Djidelli strong earthquake, wasparticularly noticeable and several flooding oflow lands were reported. But the source of thesetsunamis, although they followed earthquakes,could be related to the coseismic deformation aswell as to turbidity currents, an originfrequently proposed. The same combinedsources can be suggested for the 2003earthquake, but fine modelings of the tsunamiconstrained by observations as detailed aspossible could help, first, to refine the sourcemechanism, then to improve the tsunamihazard assessment along the Algerian coasts.Indeed, even though tsunami hazard isapparently rather moderate in Algeria, thecombination of thrusting events, occurring onsteep slopes, off densely inhabited coastal areas,makes the study worth an effort.

ReferencesCampos, M.L., Tsunami hazard on the Spanishcoasts of the Iberian peninsula, Science ofTsunami Hazards, 9, 83-90, 1991.

Campos Romero, M.L., El riesgo de tsunamis enEspaña. Análisis y valoración geográfica,(http://www.geo.ign.es/servidor/sismo/cnis/catsunami.html), Instituto Geográfico Nacional, SerieMonografías, n° 9, 1992.

Carreño, E., A. Izquierdo, and A. Suárez,Spanish tsunami catalogue, Proceedings of theInternational Conference on Tsunamis, Paris,May 26-28, p.101-110, 1998.

Delouis, B., Giardini, D., Lundgren, P., and J.,Salichon, Joint inversion of InSAR, GPS,teleseismic and strong motion data for thespatial and temporal distribution of earthquakeslip: Application to the 1999 Izmit Mainshock,Bulletin of the Seismological Society of America,92, 278-299, 2002.

Hébert, H., P. Heinrich, F. Schindelé, and A.Piatanesi, Far-field simulation of tsunamipropagation in the Pacific Ocean: impact on theMarquesas Islands (French Polynesia), Journal ofGeophysical Research, 106, C5, 9161-9177, 2001.

Heinrich, P., A. Piatanesi, and H. Hébert,Efficiency of deep submarine landslides inproducing tsunamis: the 1998 Papua NewGuinea event, Geophysical Journal International,145, 97-111, 2001.

Okada, Y., Surface deformation due to shear andtensile faults in a half-space, Bulletin of theSeismological Society of America, 75, 1135-1154,1985.

Okal, E.A., Seismic parameters controlling far-field tsunami amplitudes: a review, NaturalHazards, 1, 67-96, 1988.

Smith, W.H.F., and Sandwell, D.T., Global seafloortopography from satellite altimetry and shipdepth soundings, Science, 277, 1956-1962, 1997.

Vallée, M, Stabilizing the Empirical GreenFunction analysis: development of the projectedLanweber method, in revision for BSSA, 2003.

Yelles-Chaouche, A., Coastal Algerian earthquakes:a potential risk of tsunamis in westernMediterranean? Preliminary investigation, Scienceof Tsunami Hazards, 9, 47-54, 1991.

1 2 4 5 6 7 8 9 20 50

cm

0

0

4

4

8

8

36 36

40 40

44 44

Figure 4: Maximum water heights obtai-ned after a 1 h 40 min of tsunami propaga-tion in the Mediterranean. Given the fault

strike, the maximum values are foundtowards the westernmost island of the

Balearic islands (Ibiza).

13 September 2003

The observation and monitoring of theseismicity in Albania is carried out by theDepartment of the Seismological Network,which is part of the Seismological Institute ofAlbania, Albania Academy of Sciences. TheSeismological Network of Albania wasestablished in 1968 with the setting up of thefirst seismological station at Tirana whichbecame the Seismological Center, after settingup of the other seismological stations all overthe country in 1976. In 1994, theSeismological Center has been renamed theSeismological Institute, by decision ofAcademy of Sciences of Albania. This Instituteincorporates: Department of SeismologicalNetwork; Department of Seismicity andSeismotectonics; Department of EngineeringSeismology and Department of EarthquakeEngineering. The Seismological Networkconstituted by 10 seismological stations is themain department of this Institute and isresponsible for the seismic activitymonitoring and the relevant data collection inand nearby Albania, also ensuring themaintenance of the Albanian SeismologicalNetwork.

Albanian seismicity in theframe of the geodynamics ofthe surrounding zoneAlbania is located in the western part of theBalkan Peninsula where earthquakes are themost significant hazard on the list of naturalhazards of the country. On the worldwideseismic zonation, Albania is situated on theAlpin-Mediterranean seismic belt. This beltcomprises the wide zone of contact betweenlithospheric plates of Africa and Eurasia, fromAzores Islands up to the eastern border ofMediterranean basin. The concept of platetectonics in this zone is especially complicatedby the presence of numerous blocks and therelease of stress through plastic deformationon a large part of it. The region that surroundsAlbania comprises a wide tectonic belt withrelatively rigid blocks as Adriatic, some sectorsof Alpine belt, Alps, Carpathians, BalkanMountains, Dinarides, Helenides, the HellenicArc and Anatolian belts as well as internalbasins as Terrene, Aegenian, Panonia andBlack Sea. In the above-mentioned belt, theseismically most active region is the Aegenianand the surrounding zones, including Greece,Albania, Montenegro, Macedonia, SouthBulgaria and Western Turkey. Leaving apartthe Helenic Arc where the African plate sinksunder the Eurasian plate in the subductionform, the other contact between these twoplates and especially the part where thewestern wing of Helenic Arc already meets thewestern coasts of the Balkan peninsula, isrealized through the Adria microplate. Thisunit acts as a wedge between the Apenines,

Alps and Dinarides-Albanides-Helenidesmountain range. The origin of orogenicsystems of western Balkan as well as thosesystems surrounding on the north and west ofthe Adriatic Sea, is strongly connected withthe convergence between Eurasian andAfrican plates.

From the focal mechanism andpaleomagnetism studies, it is revealed thatthe Adria microplate participates on aclockwise rotation with pole in the northernItaly. The conclusions of many studies on thegeodynamics and seismicity of Aegenian andgenerally of eastern Mediterranean zoneswhere Albania takes place are converged onthe point that mostly the seismicity of Albaniais strongly connected with the contactbetween Adria and Albanides orogen which ispart of a wider collision between Eurasianand African plates (Sulstarova et al, 2003).This contact which possibly takes effectthrough a continental type of collisionunceasingly accumulates deformations andpropels along the longitudinal tectonic faultsbordering it. This also creates transversaltectonic faults cutting it and penetrating tothe interior of the peninsula (Figure 1).

The strongest seismic events inAlbaniaThe seismicity of a certain region isdetermined as a function of earthquake size(magnitude, intensity, seismic moment etc.) aswell as the frequency of their occurrences. Onthis basis, keeping in mind the well knownclassification of earthquakes according to theirmagnitudes (Hagiwara, 1964, Lee et.al., 1981),the seismicity of Albania is characterized by anintensive seismic microactivity (1.0<M≤3.0),many small earthquakes (3.0<M≤5.0), raremedium-sized earthquakes (5.0<M≤7) and veryseldom by strong earthquakes (M>7.0)(Sulstarova et al, 2003).

Usually the seismicity of a country is separatedin two periods: the historical seismicity and theinstrumental one. Historical seismicity isbased on the information collected fromdifferent sources and has to do with the periodof history when the earthquakes were not yetrecorded by special instruments. Instrumentalseismicity is associated with the 20th centurywhen the implementation of seismologicalstations started in Europe and worldwide andearthquake records began to be collected andanalyzed systematically.

Monitoring of Seismicity in AlbaniaLlambro Duni, Neki Kuka, and Edmond Dushi

Seismological Institute, Academy of Science of Albania.

Figure 1: Map of active faults

The historical seismicityThe historical seismicity of Albania isdescribed in some various catalogues such as:Mihajlovic 1951; Shebalin et al., 1974,Sulstarova et. al., 1975, Makropoulos et. al.,1981, Papazachos et. al., 1989.

From the evidence we possess today it resultsthat from the period of III–II centuries B.C toour days, Albania has been stricken by 55strong earthquakes with intensities Io_VIIIdegree (MSK-64), for which 15 have had anintensity Io_IX degree (MSK-64). From these55 earthquakes, for a period longer than 2000years, 36 of them belong to the 19th century,which make us believe that the number ofdisastrous earthquakes we report isunderestimated and other disastrousearthquakes are hidden on the depth ofhistorical time.

There are reliable evidences that old town ofDurres (Dyrrahum) has been stricken severaltimes by strong earthquakes, which causedserious human and economic losses. Oldchronicles report that this town has beenalmost totally destroyed on the year 177 B.C,334 or 345 A.C., 506, 1273, 1279, 1869 and1870. Documents for the earthquake of March1273 say that the town of 25 thousandinhabitants at the time has been totallydestroyed. There were many casualties andthe people who survived left the town seekingfor other living places. After this earthquakethe importance of the town of Durresi as aharbor in the Adriatic Sea was diminished.

On the centuries III-II B.C. there are evidencesthat Apolonia, another ancient town, has beenstricken by strong earthquakes which causedlarge casualties and damages.

On the year 1153, the town of Butrinti (oldButhrot) on the south of Albania, has beendestroyed from a strong earthquake. Its tracescould be found even today on the remnants ofthis old town. Some descriptions of thestrongest historical earthquakes are givenhere:

The earthquake of October 12th, 1851

On October 12th, 1851, around 07 o’clock astrong earthquake hits the gulf of Vlora. InVlora town a part of the buildings wasdestroyed and the other part was heavilydamaged. According to the news of that timethe number of casualties was about 200.Kanina village was heavily damaged as well.The sea level in Vlora bay rose of 66 cm andfloods were observed. Before the main shock at02 h 40 min a weak foreshock was felt in Vloratown. The strong main shock (07 h) wasfollowed by a great number of aftershocks.

The earthquake of October 12th, 1851 wasstrongly felt in Italy, causing panic among thepopulation of Taranto, Bari, Barleta, Canosaand Cerignola. It was felt V degree at Ioannina(Greece) and III-IV degree at Naples (Italy).According to other data serious damage wasobserved at Berat. Previous authors did not

mention Berat, but Himara. According to ouropinion this earthquake caused damages inboth these towns (Vlora and Berat). Thehighest intensity of this earthquake had to beIX degrees and its epicenter at coordinates40o.5 N, 19o.5E.

The earthquake of October 17th, 1851

On October 17th, 1851 a very strong shockcaused lots of destructions in town of Berati.The fortress of the town was damaged andunder its ruins 400 soldiers were buried. Thisfact showed that other victims had to beobserved in the town of Berati. Cracks on theground were observed together with fountainsof sands and water mixed together, and a kindof a sulfur dust, which made the breathingdifficult, was observed. Big landslides wereobserved as well. The highest intensity for thisearthquake had to be IX degree (based mainlyon the degree of destructions of the cityfortress).

The earthquake of June 14th, 1893

On June 14th, 1893 a destructive earthquakehits the region of Himara and especially thevillage Kudhes was totally destroyed. Rock-fallsand cracks on the ground were observed over aconsiderable area. The majority of the dwellinghouses was destroyed in Kuç village (Vlora) aswell. The highest observed intensity of thisquake had to be IX degree with the epicenternearby the village of Kudhes and Çorraj. Thisearthquake was felt in Puglia, and in Italy.

The Instrumental seismicityThe establishment of seismological stations inEurope at the end of the 19th century andespecially at the beginning of 20th centurymade possible even the evidence ofearthquakes occurred in Albania and nearby.Depending on the density and modernizationof seismological stations in Europe andworldwide one can say that the earthquakes ofAlbania and nearby with magnitudes Ms≥6.0are recorded from the seismological stationssince the beginning of 20th century; those withmagnitude Ms≥5.5 since 1911; those withMs>5.0 since 1940; those with ML≥4.0 since1968 and those with magnitude ML≥2.5, since1976.

From the data collected and elaborated it isevidenced that during 20th Albania has beenstricken by many damaging earthquakes(Figure 2). The most disastrous earthquakes,also displaying the highest magnitudesoccurred in Albania during 20th century are:

The Shkodra earthquakes of 1905

It represents one of the biggest series of strongearthquakes of Shkodra valley: the highestintensity, large number of aftershocks andlarge consequences of the main shock andstrongest aftershocks. This serie had been thesubject of many seismological studies (Koçiajet. al., 1980).

The strongest shock occurred on June 1st, 1905at 4 hour 42 min 43 sec. This shock wasrecorded by a large number of seismologicalstations despite their lower sensibility. Themagnitude of this earthquake was determinedas Ms=6.6. The duration of the earthquake onJune 1st, 1905 was 10-12 sec and caused bigdamages among the population and propertiesin the town of Shkodra and the surroundingvillages (especially in the villages SE toShkodra). This shock caused about 200 deadand about 500 injured.

The highest intensity of this shock had to beIX degree (MSK-64). This earthquake was feltin the Yugoslavian territory, in the southernpart of Hungary, in Bulgaria, Greece and Italy.

The earthquake of June 1st, 1905 is famous forits large number of aftershocks, some of themvery strong. These aftershocks continued up toJuly 1906.

The earthquake of November 30th, 1967

This earthquake occurred at 07h 23m and hadMs=6.6. It caused heavy damages in thedistricts of Librazhdi and Dibra in Albaniaand in Western Macedonia. The intensity wasIX degree. In the districts of Dibër andLibrazhdi, this earthquake caused damages in13 localities and 177 villages implied thedeath of 12 people and 174 people wereinjured. Were damaged: 6336 buildings; 5664dwelling houses, 156 social-cultural objects(133 schools), 534 houses collapsed, 1623suffered heavy (of 3-4 degree) damages and4179 suffered medium (second degree)damages. Damages of first degree wereobserved over a wide zone, including districtsof Elbasan, Tirana, Rreshen, Burrel and

CSEM / EMSC Newsletter

14

Figure 2: Map of earthquake epicenters;Time span 1958-2000, Ms≥4.5.

Pogradec. Casualties were observed in theYugoslav territory as well.

The largest damages were observed during thecontact between limestone and flyschsediments and along the fault created by thisearthquake on free surface. This fault had azigzag form of general strike of NE 40 degrees.This earthquake was accompanied by manyrock-falls especially in the epicentral area. Thenumber of aftershocks arrived up to 1200. Thestrongest one was that of December 2nd, 1967at 12h.

The earthquake of April 15th, 1979

The earthquake of April 15th, 1979 (so-calledthe Montenegro earthquake), is one of thestrongest earthquakes, which occurred in theBalkan Peninsula during the 20th century. Itsmagnitude has been evaluated between 6.6and 7.2. In our catalogue we assigned thevalue of Ms=6.9 (Karnik, 1996). The epicenterof this earthquake is in the coastal area, nearPetrovac, Montenegro. It happened at 06h 19m

of April 15th, 1979. Many foreshocks occurredabout two weeks before the main shock ofApril 15th, and the aftershocks continued formore than 9 months. The strongest aftershockis that of May 14th, 1979, with magnitudeM=6.3. The intensity of this earthquake inepicenter is IX-X degree (MSK-64). The mainshock of April 15th, 1979 earthquake caused 35casualties in Albania and 382 injuries. Morethan 100.000 inhabitants (mostly in thedistricts of Shkodra and Lezha) were lefthomeless. There were destroyed almostcompletely 17.122 dwelling houses and social-cultural facilities. The most casualties andeconomic losses were observed on the coastalpart of Montenegro. The earthquake of April15th, 1979 has been felt strongly in all theterritory of Albania. It has been felt IV degree(MSK-64) in Northwestern Greece, Croatia,Slovenia and Eastern Italy.

This earthquake has been accompanied by a lotof physical-geological phenomena on theground. In Shkodra and Lezha districts manysoil cracks, liquefaction phenomena, riverbankssubsidence and rock falls were observed.

The earthquake of Tirana, January 9th,1988

On January 9th, 1988, 01h 02m (GMT), anearthquake with Ms=5.4 hits Tirana city andthe villages nearby causing relatively slightdamages. The epicenter of this earthquakewas situated ten kilometers on thesouthwestern part of Tirana city. The highestintensity reached VII degree (MSK-64). Themost damaged villages were Petrela, Arbanaon the south of Tirana. Slight damages wereevidenced in Tirana city, especially on thenorthwestern and western part.

It is worth mentioning that for thisearthquake, the strong motion instrumentinstalled in Tirana seismological stationrecorded a very high and unusual value for

maximum acceleration in E-W component:404.8 cm/s2. It can be explained by the factthat the strong motion instrument was verynear to the fault, which generated thisearthquake.

The history of instrumentalobservations in AlbaniaAs previously mentioned, the instrumentalseismological observations in Albania begunwith the setting up of the first seismologicalstation at Tirana, on 1968 as part of theGeophysical Department of the TiranaUniversity. This station was first equippedwith an intermediate-period instrument,three-component SS-1, Sprengnether typeseismograph with light paper recording. Lateron, the station of Tirana has been equippedwith one short period instrument, DDJ-1(1973), one Kinemetrics short period, three-component instrument SS-1 and a three-component, long period LP-S-5100 (1975).Since 1973 this station was named TheSeismological Center and became an integralpart of the Academy of Sciences of Albania.The main task was to perform studies onseismicity and seismic zoning of Albania,monitor the seismic activity of the country,compile the bulletins of earthquakes and toinform the government on earthquakeemergence situations. As result of the studiesperformed by this center, based on theinformation of instrumental data, someimportant works were carried out, such as:The Seismic Map of Albania on scale1:1.250.000 (Sulstarova et al., 1972); TheCatalog of Earthquakes of Albania(Sulstarova & Koçiaj, 1975); TheSeismotectonic Map of Albania on the scale 1:1.000.000 (Aliaj et al., 1973).

The year 1976 is the beginning of the AlbanianSeismological Network (ASN). The first fourseismological stations were installed around theFierza Reservoir in northen Albania. Thepurpose of this sub-network was the monitoringof the possible induced seismicity from thereservoir impoundment. The swarm ofNikaj–Merkuri in North Albania, near the highdam of Fierza HPS Reservoir and Reservoir ofKomani HPS was a special event observed andmonitored by ASN. It started abruptly onNovember 10th, 1985, reached its peak by the endof November-beginning of December 1985 andthen continued at a much lower level of activityfor some months. A total of more 17.000 microearthquakes with local magnitude ML>1.0 wererecorded by Bajram Curri SeismologicalStation–the nearest station to the swarm zone;

1800 of which were used to determine thehypocentral coordinates. About 300 of them werefelt and caused slight damages in the epicentralzone; the maximum intensity was VI–VII degreeof MSK-64 scale. The maximum magnitude ofearthquakes in this swarm was ML=4.6; thedepth of micro earthquakes in this swarm wasshallow, up to 10–15 km.

All the stations were equipped with Chinesetype analog drum record three-componentshort-period seismographs DDJ-1. The practiceof work in these stations and homogenization ofmagnitude formulas and other routineprocedures were the object of a long andcontinuous work from the responsible staff. Bythe end of the 80’s, ASN was composed of 13seismological stations.

The Seismological Center of Albania enlargedthe scope of its research and seismic monitoringof the natural and induced earthquake activity.In this period the seismological andseismotectonic studies were completed onhigher basis, begun the studies of seismichazard on regional and national scale, themonitoring of strong motion shaking andearthquake engineering studies.

The strong motion network has beenestablished for the first time in 1983. During atwo years period, thirteen seismological stationsof the albanian seismological network wereequipped with SMA-1 analog accelerographs.The second phase begun in 1985 with theequipment of the major dams of the countrywith accelerographs and seismoscops of WM-IItype. At the end of 1986, thirty SMA-1accelerographs and fifty seismoscops weredistributed around the country and intended toconstitute the base for future seismic hazardassessment studies, studies concerning the soilamplification, microzonation of major towns,soil-structure interaction etc. This networkrecorded the Tirana earthquake of January 9th,1988 (ML=5.4) with acceleration as high as404.8 cm/sec2 on E-W component on sandstonesof the Tirana seismological station. The peakvalues of the corrected acceleration computedvelocity and displacement are given in Table 1(Muço et al., 2001).

The duration of the strong motion, withacceleration above 0.05g is about 6 sec. Thisvalue of peak ground acceleration is muchhigher than the value of seismicity coefficientkE=0.1 proposed by the KTP-N2-89 (Albanianaseismic design code) for this seismic zone,with an expected seismic intensity of VIIdegree according to MSK-64 scale, to whichthe area of Tirana belongs. It shouldnevertheless be mentioned that in spite of the

15 September 2003

Comp. Max. acc. Max. vel. Max. displ.(cm/sec2) (cm/sec) (cm)

Z 69.3 4.7 1.6

E-W 404.8 14.3 1.6

N-S 106.3 5.9 1.3

Table 1: Ground motion parameters of January 9th, 1988 earthquake.

large recorded peak acceleration, the damagewas not as severe as one might have expected.

The strong earthquake of April 1979 withMs=6.9, hitting the border region betweenMontenegro and Albania leaded to anincreasing attention by Albanian Governmenttoward the seismological studies. The firstSeismic Zonation of Albania in the scale1:500.000 was followed by the Technical Designand Construction Code, replacing the old Codecompiled on 1952 and updated on 1963.Important studies were carried out on thecharacteristics of Albanian earthquakes, focalmechanism, macroseismic field, energeticestimations, seismotectonics, active faultssystems, etc. The engineering seismologicalstudies on the period 1982-1991 wereconcentrated mainly on the seismic hazardassessment at local level at relatively largescale of 1:10.000 for the most importantinhabited centers situated in seismic hazardouszones of the country. The most important worksproduced during this period are: SeismicZonation of Albania in the scale 1:500.000 (Sulstarova et al., 1980); The Earthquake ofApril 15th, 1979 (Proceedings of the symposiumon this earthquake on Shkodra, 1980);Microzonations Studies of different cities ofAlbania (Group of authors, 1984-1991). Someperiodical publications on seismological studiesare issued during this period.

Latest improvements of theseismological networkWith the beginning of the 90’s, progress weremade for a modest upgrading of ASN, through afund granted by the Academy of Sciences ofAlbania for buying some spare parts such aspen-motors, amplifiers, digital timing systems,etc. In the framework of the ASPELEA project,the upgrading of the network continued, and anew seismographic station, a TeledynePortacorder has been set up in Sarandaseismological station. Nevertheless, it was clearthat this kind of instrumentation at the ASNwas not in the contemporaneous level withmany seismological stations technologically andphysically exhausted. An attempt has beenmade in the framework of PLATO-1 Project(Implementation of telemetered net-works inthe countries of Mediterranean area) by WorldLaboratory, Switzerland, to set up a telemeterednetwork constituted by 8 stations, short-periodone component system. 5 stations wereimplemented, but the political events of 1997did not allow the continuation of the project.

At present time the Seismological NetworkDepartment has substituted the analoginstruments with digital ones. Now, all thestations are equipped with short-period, threecomponent instruments of GBV-316 type,GeoSig production. Their control and datadownloading is carried out remotely by the dial-up communication. When, any event withmagnitude ML≥2.5 is triggered within thecountry, the new network provides thenecessary information for rapid hypocenter

determination, using the HYPO71 software, andmagnitude estimation. The information is sentto the Civil Protection Department of theDecentralization and Local GovernmentMinistry as rapidly as possible. The informationis disseminated to news media only forearthquakes with magnitude M>4. Themonitoring of the seismic activity is providedcontinuously twenty-four hours and onetechnician is always present, assuring thecorrect data processing and informationdistribution. We are taking the necessarymeasures for the information to be sent to theinterested international seismological agencies.The web page of the Institute that will beavailable shortly will help for a rapiddissemination of the data, including waveformsfor important events. Periodically, theDepartment issues a bulletin, composed by thepreliminary readings. This Bulletin is sent tothe neighboring Institutes, like Montenegro,Macedonia and also to the EMSC.

The challenging effort for the staff of theSeismological Institute is the implementation ofa new fully integrated digital seismographsystem using the satellite telemetry (VSAT).The Institute is under a contractual agreementwith Nanometrics (Canada) for theimplementation of this new system. The centralhub data acquisition system and two remotestations have just been commissioned. Therewill be four remote sites by the end of this yearand this number will increase by one or twoeach next year. This system offers enormouspossibilities that data collected to be shared andexchanged on-line by the Internet technologieswith the homologues neighbour institutionsabroad, especially in Balkan region. Also, thedata collected will be complementary with those

collected by other systems (dial-up and RFtelemetry networks, etc.).The new VSAT systemwill provide support for additional services suchas continuous or triggered strong motion, dualfrequency GPS reference data, meteorologicalsensing, auxiliary data communications andLAN or WAN extension to remote sites. With nodoubt, it comes apparent that this system willimprove seismic activity and in generalgeophysical monitoring capabilities in Albaniaimproving greatly the detectability and theearthquake parameter (location, magnitude,etc) estimation of the existing seismologicalnetwork (Figure 3).

References:Aliaj, Sh., Koçiaj, S., Sulstarova, E. (1973) TheSeismotectonic Map of Albania on the scale 1:1.000.000 Archive of Seismological Institute,Tirana, Albania.

Hagiwara, T. (1964) Brief description of theproject pro-posed by the Earth-quake PredictionResearch Group of Japan. Proc. U.S.-Japan Conf.Res. Relat. Earth-quake Prediction Probl., 10-12.

Karnik V. (1996) Seismicity of Europe and theMediterranean. Ed: K.Klima. Geophys. Inst. Ofthe Academy of Sciences of the Czech Republicand Studia Geo, Praha.

Koçiaj, S., Sulstarova, E. (1980) The earthquakeof June 1, 1905, Shkodra, Albania. Intensitydistribution and macroseismic epicentre.Tectonophysics,- 67, pp. 319-332.

Lee, W.H.K., Stewart, S.W. (1981) Principles andApplications of Microearthquake Networks.Academic Press Inc.,1-266.

Macropoulos, K., Burton, P.W. (1981) ACatalogue of seismici-ty in Greece and adjacentareas. Geophys. J. R. Astr. Soc., 65, pp. 741-762.

Mihajlovic, D. (1951) Catalogue destremblements de terre Epiro-Albanais. Travauxde l'Academie de Sciences Serbe, Beograd.

Muço, B., Vaccari., Panza, G. (2001) Seismiczonation of Albania using a deterministicapproach, AJNTS, 2001 (1), pp.5-16.

Papazachos, B., Papazachou, C. 1989 On GreekEarthquakes. Thessaloniki, Greece, 355 p.

Shebalin, N.V., Karnik, V., Hadzievski, D. (1974)Balkan Region–Catalogue of Earthquakes,UNESCO Office, Skopje.

Sulstarova, E., Koçiaj, S., Aliaj, Sh. (1972) TheSeismic Map of Albania on scale 1:1.250.000.Archive of Seismological Institute, Tirana, Albania.

Sulstarova E., Koçiaj S. (1975) The catalogue ofearthquakes of Albania. Publ. SeismologicalCenter of Academy of Science of Albania, Tirane,(in Albanian) 224 p.

Sulstarova, E., Kociaj, S., Aliaj, Sh. (1980)Seismic Zonation of Albania in the scale1:500.000. Published by the SeismologicalCentre of Academy of Science, Tirana. 297 p.

Sulstarova, E., Muço, B., Aliaj, Sh., Kuka, N.,Duni, Ll. (2003) Earthquakes, seismic hazardand seismic risk in Albania. Internal Study,Archive of the Seismological Institute, ArkiviInstitutit Sizmologjisë, Tiranë, p. 82.

The Earthquake of April 15 (1980) Proceedingsof the symposium on this earthquake onShkodra. Published by the Seismological Centreof Academy of Science, Tirana. 531 p.

CSEM / EMSC Newsletter

16

Figure 3: The seismographic and strongmotion networks of Albania

OrganisationThe CRAAG was founded in 1985 after themajor El Asnam earthquake of October 10th,1980. The occurrence of this big eventdemonstrated the urgent need to promote theresearch activities in the earth and universesciences.The CRAAG inherited from both activities ofthe Algiers Astronomical Observatory created

in 1890 and the IMPGA (Institut deMétéorologie et de Physique de Globed’Alger) established in 1931.

The CRAAG located at Algiers has threemain departments: the AstrophysicsDepartment where universe studies aremade, the Geophysics Department which isformed by Geomagnetism, PaleomagnetismGravimetry and Electromagnetic laboratories

and the Seismological Monitoring and studiesDepartment. This later is divided in two mainservices: the first one is responsible for themaintenance and regular operations of theAlgerian Seismic Network, the second isdevoted to the research activities with fourlaboratories: seismology, seismotectonic,geodetic deformations and the seismic hazardassessment.

In parallel of its research activities, the ESSdepartment is involved in the nationalprogramme of seismic hazard reductionlaunched in 1985 after the El Asnamearthquake. For instance, the CRAAG playsan important role for schools and to informthe population. Some cooperations areestablished with the Earthquake EngineeringInstitute for the elaboration of the zoningmap of Algeria.

The Algerian Seismic Network(REALSAS)

Seismic activity in Algeria

Algeria is affected by a moderate to highseismicity. From history, the first knownearthquake occurred in Algiers in 1365. Itcaused huge damages to the town and manypeople died. Since that time, several moderateto strong seismic events happened in Algeria(Algiers, 1716, Blida, 1825, El Asnam, 1980,Boumerdes, 2003).

The Algerian seismicity is mainly located inthe northern part of Algeria, more precisely inthe Tellian chain (Figure 1).

Some offshore epicentres have been recordeddemonstrating that the margin is also active.The High Plateau is marked by a low seismicactivity. In the Saharian Atlas, someepicentres have been recorded and might begenerated by the south Atlasic flexure.

The Algerian earthquakes are superficial, nodeeper than 20 kilometres, explaining their

destructive character. This is in agreementwith the lack of subduction zone along themargin.The seismicity of northern Algeria ischaracterised by a compressional NNW-SSEstress regime (Groupe de Recherche deNéotectonique, Phillip, 1977). The focalmechanisms of the recent importantearthquakes indicates that the compressionalmovements are due to reverse or thrust faults

related to folding, trending generally NE-SW(the El Asnam, Tipaza, Ain Temouchent,Boumerdes faults).

History of the network

In Algeria, the instrumental seismologystarted in 1910 when the first Algerianseismological station was installed at Algiers,more precisely at the Observatory (ABZ). Thisstation was equipped with a mechanical Bosch-

17 September 2003

The CRAAG, AlgeriaA.K.Yelles, S. Haned and A. Deramchi,

Centre de Recherche en Astronomie Astrophysique et en Géophysique

ESS Department

Research LaboratoriesSeismology

SeismotectonicsHazard Assessment

Géodetic Deformation

Risk Mitigation(Education, Earthquake Engineering,…)

Earthquake Monitoring Service

Data Analysis Service

Figure 1: Seismicity of northern Algeria

The Seismological Department

Mainka seismograph. Later, other seismicstations with a short- period Grenet-Coulombseismographs were installed in severallocalities as Tlemcen, Setif or Relizane; some ofthem in the dams. The Oued Fodda (El Asnam)station was installed in 1935, the Benaouda(Relizane) station in 1955. The Setif stationstarted to be in operation from 1958. Althoughthese stations recorded the main Orleansvilleearthquake of September 1954 (M: 6.7), onlyfew studies relative to this important seismicevent were carried out.

In that period, the first Algerian seismiccatalogues were edited by Grandjean in 1950and Rothé in 1955. These catalogues reportedmain seismic events with some information ontheir effects.

The autonomous stations operated more or lesscontinuously until 1980 when the El Asnamearthquake occurred. This major eventdemonstrated the urgent need to install a real-time seismic network. In 1985, the CRAAGobtained a financial support, which allowed itto be equipped in 1988 with a telemetric (radio-link stations) seismic network of 32 stations.The Kinemetric Telemetric seismic network(REALSAS Réseau Algérien de Surveillance etd’Alerte Sismique) was installed in 1990.Unfortunately, at that time the internalsituation of Algeria was marked by politicalevents and the network underwent with forcethis bad environment.

Since 1998, as the situation improved, theAlgerian network has been progressivelyreinstalled. Today, the network is composed byfour subnetwork arrays (Algiers, Oran,Constantine, Chleff) located in the four majorseismic regions of Algeria.

The Algerian Seismic Network

Each subnetwork is composed by a threecomponent regional station and seven one-component satellite stations. These later are

installed in the mountains of the Tellian chainbecause of the quality of the signal and thecapacity of a good transmission. Each stationis equipped with a SS1 short periodseismometer of 1s frequency. The signal istransmitted continuously from the remotesites to the regional station by radio-link andthen to the central station located at Algiers(figure 3). There, the real time data arerecorded in analogue forms on drums anddigital forms.

The Algerian network has been recentlyexpanded by five new stations installed inareas (Guelman Batna, Medea, Tlemcen,Tipaza) not yet covered by the initialconfiguration of the net. These stations areconnected to the central station of Algiers bytelephone link.

In the Central station, seismological data areinterpreted (P and S arrivals identification,first onsets, signal duration) and processed bycomputer. For the focal parametersdetermination, the widespread HYPO71software is used.

At the Central Station of the REALSAS, twopersons are on duty 24 hours per day. Thesetwo scientists belongs to a team which is incharge of:

-Permanent high-quality recording andanalysis of the seismic signals

-Determination of the earthquake parameters

-A daily service to government agencies (CivilProtection, Wilaya…) and to media.

In the Data Bank Service, data are collectedfrom the monitoring service and InternationalCentres. All data contribute in edition of themonthly bulletin which is transmitted towardsthe national and International data Centressuch as CSEM, ISC, NEIC by mailing and nowInternet facilities.

The data bank service has also in charge thepublication of the catalogues. A new one

(Yelles et al., 2002) for the seismic eventsbetween 1992-2002 has been publishedrecently. This catalogue is an extension ofthe previous one published in 1994 byMokrane et al.

The earthquake monitoring operated by theREALSAS promotes several scientific projectssuch as:-Earthquake source mechanisms-Seismotectonic studies-Seismic hazard assessment-Seismic zonation-Seismic prediction (precursors, electro-magnetic ,…)

-Geodetic deformation-Seismic waves propagation-Sites effects-International Project (Geoscope, Midsea…)

For the field investigations, 30 short periodportable stations are available allowingmicroseismic investigations in the interestedareas.

Future developments for the AlgerianSeismic network

After the occurrence of the Boumerdesearthquake of May 21 , 2003 (Mw: 6.8), anopportunity to develop the Algerian SeismicNetwork is given by all the new investmentsproposed by the government and thecooperation channel. Near future strategy willinclude several actions. Among them,densification of the network by installation offive new stations obtained after the AinTemouchent earthquake, installation ofseveral new stations in the northern region ofAlgeria and installation as a complementarysystem to the REALSAS network, a newBroad-Band stations Network (10) with itshigh dynamic range and stable transfercharacteristics. Because of a poor quality ofthe telephonic network, the use of a satellitetechnology (VSAT) in the data transmission isalso expected. This improvement of the seismicmonitoring system will permit a betteracquisition of the seismic data, a good dataanalysis and a better knowledge of theAlgerian seismicity. Very relevant seismicstudies should be expected, allowing a bettercontribution of our Institute in the cooperationwith the International data Centre and in theseismic risk mitigation in Algeria.

CSEM / EMSC Newsletter

18

Figure 2: The Algerian Seismic Network

Figure 3: A view of the Central Station ofthe seismic network (Algiers station)

IntroductionEgypt is located close to one of thecontinental fracture system (Hellenic arc) atthe convergence boundary of two biglithospheric plates (Eurasia and Africa).Also, Egypt is affected by the opening of theRed Sea (Mid Oceanic System) and its twobranches (the Gulf of Suez and the Gulf ofAqaba-Dead Sea transform system). Thus theseismicity is due to the interaction betweenthe three plates of Eurasia, Africa andArabia. Within the last decade, Egypt hasbeen shaken by several distant and neardamaging earthquakes. Such events wereinterpreted as the result of the interaction ofthe African, Arabian, Eurasian plates andSinai subplate. Some large earthquakes arelocated near the plate boundaries as far asCyprus and were felt in Egypt. Damagingearthquakes effects in Egypt are mostlyconcentrated on the highly populated areas ofthe Nile Valley and Nile Delta. Earthquakelosses from such events started to diminishafter the establishment of the EgyptianNational Seismic Network (ENSN) from 1997on. Also, Egypt is affected by the opening ofthe Red Sea (Mid Oceanic System) and its

two branches (the Gulf of Suez and the Gulfof Aqaba-Dead Sea transform system). Therelative motion along these plates createsareas of high seismicity in Egypt. Some areashave been stricken by immense earthquakes,causing considerable damage (Sieberg, 1932;Maamoun et al., 1984; Ambraseys et al.,1994; and El-Sayed and Wahlström, 1996).Therefore, it could be concluded thatalthough the damaging earthquakes occurredinfrequently, its risky consequences could notbe ignored (Osman and Ghobarah, 1996).

In the last century, Egypt has been shakenby many large earthquakes (Ms= 5.8-7.3).These earthquakes caused considerabledamage in both historical and recentconstructions. The most recent damagingearthquakes were caused by the 1955(Alexandria) and 1992 (Cairo) earthquakes(see Woodward-Clyde Consultants 1985,Ambraseys et al., 1994, El-Sayed, A. andWahlstrom, R., 1996, El-Sayed et al., 1999and 2000, and El-Hadidy, 2000). Reports forsuch earthquakes reveal that the deathcaused by seismic activity exceeds the sum ofvictims caused by other natural hazards.

Moreover, Faro, Romanic and Islamicconstructions suffered damage. Damageswere concentrated mainly in Cairo as well asin cities located within Nile Delta and NileValley. Historically, many similar cases werereported. El-Sayed (1996) and Moustafa(2002) investigated the factors thatcontributed to the damage caused byearthquakes in Egypt. The result of thisinvestigation shows that building quality,site effect and local geological condition arethe main contributing factors to damage.

Tectonic settingsThe primary features of active plate tectonicsin the vicinity of Egypt are discussed in theliterature and are reported on Figure 1. Threemajor plate boundaries are recognized: theAfrican-Eurasian plate margin, the Levanttransform fault, and the Red Sea platemargin. A piece of the African plate, calledSinai block or subplate, is partially separatedfrom the African plate by the spread –apart orrifting along the Gulf of Suez. In addition tothese plate boundaries, there are twomegashear zones running from southernTurkey to Egypt.

African-Eurasian Plate Margin

The African and Eurasian plates areconverging across a wide zone in the NorthernMediterranean Sea. The zone is characterizedby folding within the Mediterranean Sea floorand subduction of the northeastern Africanplate to the north beneath Cyprus and Crete(Mckenzie et al., 1970). To the north of themargin, there is a complex zone of convergence(folding and reverse faulting) and strike-slipfaulting (Mckenzie et al., 1970). The effects ofthe plate interaction are mainly north of andremote from the Egyptian coastal margin.Some secondary deformation appears to occuralong the northern Egyptian coast, as shownby earthquake activity (Woodward, 1985).

Levant Transform

The Sinai block and Arabian plate areseparated by the Levant transform fault zone,along which left-slip faulting has beenobserved (Mckenzie et al., 1970). North of thehead of the Red Sea, the Levant transformfault extends along the lineament marked bythe Gulf of Aqaba and the Dead Sea.Prominent left-slip faulting has been observedon this zone (Garfunkel et al., 1981). The totalamount of left-slip was proposed by Freund etal., (1970) to be 110 km. Recently, Mart andHall (1984) suggested that much of the Levantmotion can be explained as geologically recentopening along the Dead Sea Rift and itsextension to the north-northeast. According to

19 September 2003

Seismicity of EgyptSalah EL-Hadidy Youssef,

National research Institute of Astronomy and Geophysics, Helwan, Cairo, Egypt

Figure 1: General Tectonic Setting of Egypt (Abu Elenean, 1998)

this model, the present-day movement of theLevant zone is both left lateral andextensional. The geomorphologic expressionsshow a small component of spreading as wellas left lateral motion. In general, the boundaryis seismically active. The activity along thiszone may materialize as swarms, e.g., theswarms of 1983 and 1993, or as foreshock-main-shock-aftershock events e.g., the event ofNovember 22nd, 1995.

Red Sea Plate Margin

The Arabian plate is continuing to rotate awayfrom the African plate along the Red Seaspreading centre. Current sea-floor spreading

has been identified as far north as 20° to 22°north, from the continuous presence of basalticcrust in the axial rift of the Red Sea, and thegeophysical signatures of newly emplacedoceanic crust (Cochran, 1983). This axial riftrepresents the boundaries between theArabian and African plates. In the northernmost Red Sea, north of 25°, Cochran (1983)notes the presence of rifted continental crustthat has been introduced by discontinuousvolcanic rocks. The seismic activity in thenorthern part of the Red Sea is concentratedat the entrance of the Gulf of Suez and Gulf ofAqaba. Earthquakes of magnitudes up to 6.9were reported, e.g., the event of March 31st,1969. Daggetta et al. (1986) conducted a

microseismic survey in eastern Egypt in early1980 and showed that there is a large numberof earthquakes in the most northern part ofthe Red Sea, averaging 25 events per day, withswarms as many as 200 events in one dayoccurring about once every seven days. Theseearthquakes occurred at depths varying from5 to 22 km and had a local magnitude varyingfrom 1 to 3.1.

Suez Rift

The Gulf of Suez is controlled by a rift thatextends north west into the African plate fromthe junction of the Red Sea Rift and theLevant transform fault. It separates the

CSEM / EMSC Newsletter

20

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

0 50 100

kmMain center of ENSN

Sub-centers of ENSN

Single component station

Three components station

Very broadband station

Single component station

Three components station

Very broadband station

Single component station

Three components station

Stations using Satellite

Stations using Radio

Stations using Telephone

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

Sinai

Red Sea

Mediterranean Sea

Western DesertEastern Desert

Figure 2: Geographic Distributions of Egyptian National Seismic Network (ENSN)

northeastern part of Egypt from the SinaiPeninsula. The Suez Rift may have becomeactive as early as Late Mesozoic or late Eoceneand established its depression by EarlyMiocene. The Suez Rift is a tectonically activestructure that is considered to be a sub-plateboundary that formed as relict of the earlyopening of the Red Sea. According to Sfratome(1984), a large part of the deformation due toplate interaction is concentrated along theSuez Rift and its continuation toward thenorth in the Cairo Alexandria fault zone.

Megashear Zone

A magnetic and gravity trend statisticalanalysis using the results of all areomagneticflown and gravity surveys in Egypt was carriedout. One of the results is a confirmation ofmajor transcurrent shear zones that belong tothe Pelusiem Megashear system across Africa.It extends from Turkey to the south Atlanticrunning subparallel to the eastern margin ofthe Mediterranean Sea then curving southwestacross Africa from the Nile Delta to the Deltaof Nigar. These two zones may correspond tothe eastern Mediterranean-Cairo-Fayum zonein Northern Egypt.

Brief History of Seismographsin Egypt

Helwan Standard Station

Instrumental earthquakes recording in Egyptstarted as early as 1899 with a single stationinstalled at Helwan. In 1962, the Helwan seismicstation has been selected as one of the AmericanWorld Wide of Standard Seismograph Network(WWSSN). It was equipped with long periodSpringnether and short period Benioff system ofseismographs. Seismic data are exchangedregularly with World Data Centres and similarstations in many countries around the world.

Helwan Visual Recording Station

In 1971 a visual recording seismic station waserected at Helwan with a short period frequencyanalyzer. The equipment was denoted to theinstitute from Japan International cooperationAgency (JICA).

Aswan Station

It has been erected in 1975 and it is equippedwith long period and short periodseismographs of Kemos type. The equipmenthas been denoted from UNESCO.

Matrouh and Abu Simble stations

These stations are equipped with shortperiod seismographs of Kemos type. The firstwas erected at Matrouh at the northwestcoast of the Mediterranean Sea and thesecond at Abu Simble furthermost south ofthe country. UNESCO has also donated thesestations in 1975.

Aswan Telemetred SeismographNetwork

After the occurrence of the 1981 Kalabshaearthquake in Southern Egypt, a seismicnetwork has been established around theKalabsha fault and Naser Lake to monitorcontinuously the earthquake activity aroundthe northern part of Naser Lake. It consists ofa 13 radio-telemetry vertical seismographstations network with a frequency analyzerof 0.2 to 30 Hz within the area. Two of the 13stations have 3-component sensors.

Kottamia Broad Band Station (KEG)

As a result of the cooperation between(NRIAG) and Italian scientists, a broad bandseismic has been erected in 1990. Thisstation belongs to MEDNET.

Hurghada seismograph network

As a result of fruitful cooperation betweenEgypt (NRIAG) and Japan InternationalCooperation Agency (JICA), in August 1994,telemetric seismological network at the bothsides of the Gulf of Suez was installed. Itconsists of eight stations, four of them in thesouthern Sinai and four others in theWestern Side of the Gulf of Suez. The maincenter was located in Hurghada City.

The Egyptian National SeismicNetwork (ENSN)

After the occurrence of the Dahshourearthquake in Oct. 1992, it was decided toestablish the Egyptian Seismic Network(ENSN). This network had to be atechnologically sophisticated system in orderto meet the important needs for public safetyand emergency management, quantificationof hazard and risk associated with bothnatural and man-made earthquakes, andengineering applications related to this, aswell as basic research. ENSN is a multiphaseprogram. It consists of a centralizedrecording and analyses system based at theDC at Helwan and 6 sub-networks thatconsist of different types of seismic stations(e.g. Broad-Band, short period, Bore Hole,Strong Motion) represented on Figure 2. TheENSN is a digital network with a duplexcommunication network which reliesprimarily on three major elements: - a highresolution digitizer (HRD series) thatprovides a resolution of 24 bits (132-DBdynamic range) and GPS timing, - a NAQS32-Pacquisition and monitoring software with

21 September 2003

22

22

24

24

26

26

28

28

30

30

32

32

34

34

36

36

22 22

24 24

26 26

28 28

30 30

32 32

34 34

36 36

0 100 200

km

22

22

24

24

26

26

28

28

30

30

32

32

34

34

36

36

22 22

24 24

26 26

28 28

30 30

32 32

34 34

36 36

Mediteranean Sea

Red SeaSinai

EgyptLibya

undefined mb

4 <= mb < 5

5 <= mb < 6

6 <= mb =< 7

Figure 3: Location of Historical Earthquakes from 2200 BC to 1900 AD

CSEM / EMSC Newsletter

22

reliable error detection and correctionmechanism and - the monitoring of the vitaltechnical parameters of the remote andrepeater stations, eliminating theunnecessary visits to remote sites.

Objectives

• Monitoring local and regional activityincluding artificial events.

• Data analysis and interpretation andresearches in the field of seismology andgeophysics.

• Studying the microearthquake activities ofcertain areas in Egypt, especially, the areasof dense population.

• Site effect studies

• Strong motion studies.

• Assessment of seismic hazard.

• Estimating the expected future earthquakeeffects.

• Protecting strategic buildings, high damand archaeological sites.

Configuration

The network consists of 62 remote sitestransmitting the data to the main centre atHelwan and five sub-centres at Hurghada,Burg El-Arab, Mersa Alam, Aswan andKharga (Figure 2). The main centre receivesthe seismic data from the near distancestations through a telemetry communicationand from the remote stations and the sub-centres via the telephone lines and thesatellite communications. The distribution ofthe seismic stations and the strong motionunits are chosen to cover the known seismicsources as much as possible. Also thisdistribution covers some regions with knownhistorical earthquakes without any evidenceof instrumental activity (e.g. Siwa seismicstations).

Brief History of the EgyptianHistorical EarthquakesAlthough the occurrence of 2200 BCearthquake is not confirmed (Ambraseys etal., 1994), no historic information on the

effect of earthquakes are available beforethat period. The ideas of the causes ofearthquakes had a mythological character.Until the beginning of the 20th century,information on earthquakes is coming frommacroseismic effects of large shocks.According to Badawy (1999) the total numberof historical earthquakes till 1900 is 83including the non-confirmed ones (Figure 3).Some of these earthquakes are initiated byseismic sources outside the Egyptianterritory. He attributed the lack of reportedearthquakes in some areas to the politicalsituations at these times and thedisappearance of the documents. Despitethis, observations can be classified into fourtime intervals in terms of studiedearthquakes: 2200 BC to 1300 AD, 1300 to1800, 1800 to 1900 and from 1900 on.

The first time interval covers the periodfrom the beginning of the 22nd century BC tothe beginning of the 14th century. For thisperiod, information for a total of 34 strongearthquakes is available, while the realnumber of felt earthquakes per century ismuch larger. The small number of shocks maybe due to the fact that people of such periodpaid their attention toward the importantearthquakes and therefore towards the rareevents, which caused extensive damage,destroyed important cultural centres. Thebeginning of the second time intervalcoincides with the beginning of the Mamlukperiod. This time interval includes theEuropean scientific revolution and theindustrial revolution. The total number ofstrong earthquakes of this period is 36 andthus the frequency of occurrence per centuryis 6. It is obvious that this rate is much largerthan the corresponding rate of the previousperiod. This is due to the fact that theobservational research started to dominateand to become the base for the new science.The beginning of the third time intervalcoincides with the end of the Ottoman Empireperiod. This time interval includes the actualbeginning of the instrumental recording. Thetotal number of strong earthquakes for thisperiod is 17; i.e. the frequency of occurrenceper century is 17. This rate is much largerthan the corresponding rate in the previoustwo periods. This may be due to thecontribution of Cairo newspaper (Badawy,1999). The historical earthquakes arestrongly related to the populationdistribution, thus we believe that the realnumber of the relatively strong earthquakesis much larger than the previously recordedones. Many earthquakes were not felt due toits generation in the desert or in the sea awayfrom the concentrated population. Figure 3shows the map of historical events till 1900AD. The fourth time interval covers theperiod 1900-2002. There are observations foralmost all strong shocks. Although manyearthquakes with magnitude about 4 are feltby many in the last decades, only earthquakes

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

0 100 200

km

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

26˚

26˚

28˚

28˚

30˚

30˚

32˚

32˚

34˚

34˚

36˚

36˚

22˚ 22˚

24˚ 24˚

26˚ 26˚

28˚ 28˚

30˚ 30˚

32˚ 32˚

Mediteranean Sea

Western DesertEastern Desert

Sinai

Red Sea

0 <=M < 2

2 <= M < 3

3 <= M < 4

4 <= M < 5

5 <= M < 6.1

Figure 4: Location of Local Earthquakes Recorded by ENSN in the period 1997-2002

with magnitude 5.0 which occurred in Egyptare considered to represent the feltearthquakes in the fourth time interval. Thetotal number of earthquakes with magnitudegreater than 5.0 in this period is 53. It meansthat the mean rate of strong earthquake inthis area is about 51.96 per century.

Recent Instrumental Seismicity

Generally, the seismic activity of Egypt wasdiscussed and studied by many authors, e.g.,Sieberg (1932); Ismail (1960); Gergawi and El-Khashab (1968); Maamoun and Ibrahim(1978); Maamoun et al., (1980); Poirier andTaher (1980), Maamoun et al., (1984);Kebeasy (1984 & 1990), Abu El-Enean (1997),Deif (1998) Badawy (1999) and El-Hadidy(1995 & 2000) to delineate the seismotectonicsources and zones.

The temporal distribution of the last decadereported seismicity shows three periods ofobservation. These periods are represented bythe early instrumental (1900-1960),intermediate instrumental (1961-1997) andrecent instrumental period (1997-now). Theinstrumentally observed seismicity in Egypthas the following characteristics:

• Concentration of earthquakes in four majorzones. These zones are known as NorthernRed Sea-Gulf of Suez-Cairo-Alexandriatrending NW-SE, Gulf of Aqaba-LevantFault, NNE-SSW, Eastern Mediterranean-Cairo-Faiyum, NE-SW and Egypt-Mediterranean Coast, E-W.

• Noticeable clusters or spots of activities atthe north end of the Gulf of Aqaba, theentrance of the Gulf of Suez into the RedSea and the eastern Nile Delta.

• Occurrence of both mainshock-aftershocks(e.g. the earthquake activity in the Gulf ofSuez starting on March 31, 1969) andswarm (e.g. the sequences in the Gulf ofAqaba in 1983 and 1993) types of activity.

• A number of small, scattered earthquakesoutside the main seismic zones. Many ofthese events were located using a singlestation (Maamoun et al., 1984), makingplotted epicentres uncertain.

Damaging earthquakes and developmentof seismic networks in Egypt

There are many damaging historical andrecent earthquakes that have been reported inEgypt. The total reported damage dependsmainly on the size of the earthquake, localgeology and location with respect to populatedareas. Comparing the earthquake size with theassociated damage shows that small local orlarge distant earthquakes could cause largedamage in different areas in Egypt, e.g., Cairo,the Nile Delta and the Nile Valley.

Egypt has a historical record of earthquakesactivity extending over the past 4,800 years. Inaddition to the magnitude 5.9 Dahshourearthquake of 12 October 1992, which damagedover 1,000 schools and killed or injured over7,000 people, the three most significantearthquakes of this century include: 1) themagnitude 6.7 earthquake offshore ofAlexandria on 12 September 1955, whichdestroyed over 300 buildings, 2) the magnitude6.8 Shadwan Island earthquake on 31 March1969, and 3) the magnitude 5.2 Aswanearthquake of 14 November 1981, which calledattention to the importance of earthquakemonitoring in the High Dam area and of hazardand risk studies in the country. More recent isthe Aqaba earthquake of 22nd November 1995,magnitude 7.2, which killed at least 8 peopleand injured 30 in the epicentral area. Damageoccurred in many parts of northeastern Egyptas far away as Cairo. Few injuries and somedamage were reported in Saudi Arabia.Substantial damage with power outages andliquefaction occurred at Eilat, Israel. Somedamage also occurred in Jerusalem and Aqaba.It was felt from Sudan to Lebanon.

Local Seismic activity

The local seismic activity recorded by ENSNfrom Aug. 1997 (start time of ENSNrecording) to Dec., 2002 is located at specificseismic zones that reflect their tectonicactivity (Figure 4). These zones are:

• Suez–Cairo shear zone, which extendsfrom the apec of the Suez Gulf towardCairo City and is characterized bymoderate activity.

• Northern Part of Eastern Desert, whichreflects a tendency of WSW-ENE active,faults (these faults transverse the Gulf ofSuez main faults). A cluster of seismicactivity from the Suez gulf to Beni SuefCity corresponds to one of these faults.

• South–West Cairo zone (Dahshour area),which is characterized by moderateseismic activity.

• High seismic activity is also related tothe Gulf of Aqaba and its extensiontowards the north.

• A cluster of seismic activity comes clearlyfrom the central part of the Suez Gulf,which indicates that the entire Gulf isactive. - Another cluster of seismicactivity is shown at the northern part ofthe Red Sea at the triple junction area.This activity tends to lie at the Aqabafault extension to the south. - A cluster ofseismic activity is as well clearly shownalong the central part of the Red Seaafter the operation of the ENSN seismicstations, which were located along thewestern Red Sea Coast. This activity maybe related to the active rifting of theNorthern Red Sea. - A new cluster ofseismic activity is recorded along thesouthern part of Naser’s Lake. Thiscluster tends to align with the E-Wdirection in agreement with the majorfault trends in this area.

• Moderate seismic activity occurs alongthe northern part of Naser’s Lake to thesouthwest of Aswan High Dam.

• Another seismic zone was identifiedthanks to the increasing detectability ofENSN through the installation of somestations in the central part of Egypt(ADB, EDF, DK2 and SFG). This zoneextends from Abu Dabab area along theRed Sea coast to the western Directionuntil the Nile River.

• The Network recorded a scatteredactivity from the Western Desert to thewest of El-Minya and Sohag cities.Scattered events also lie along theMediterranean coast but this does notreflect any specific trends.

23 September 2003

EMSC informationEMSC/CSEM - c/o LDG, Bât. Sables - BP 12 91680 Bruyères-le-Châtel, France

Chris Browitt

President president@emsc-csem.org

Rémy Bossu +33-1-69267814

Secretary General bossu@emsc-csem.org

Gilles Mazet-Roux +33-1-69267813

Alert system / Data Exchange mazet@emsc-csem.org

Stéphanie Godey +33-1-69267813

Euro.-Med. Bulletin godey@emsc-csem.org

Véronique Guerdet

Administration +33-1-69267808

Fax +33-1-69267000

Alert messages alert@emsc-csem.org

urg@ing.es

Bulletins csem@emsc-csem.org

Other matters csem@emsc-csem.org

24

CSEM / EMSC Newsletter

Syn

op6

Nan

cy -

ww

w.s

ynop

6.fr

501949 1999

CONSEILDE L’EUROPE

COUNCILOF EUROPE

EMSC membersInstitute Country CorrespondantActive MembersSeismological Institute, (ASN) Albania Dr. Edmond DushiCentre de Recherche en Astronomie, Astrophysique et Géophysique (CRAAG) Algeria Dr. A. Karim Yelles ChaoucheNational Seismological Centre (NSC) Armenia Dr. Sos Margaryan Central Institute for Meteorology and Geodynamics (ZAMG) Austria Dr. Edmund FiegweilCentre of Geophysical Monitoring of NAS of Belarus (CGM) Belarus Dr. Arkady AronovObservatoire Royal de Belgique (ORB) Belgium Dr. Roland VerbeirenBulgarian National Operating Telemetric System for Seismological Information (NOTSSI) Bulgaria Dr. Boyko RanguelovGeophysical Institute and Croatian Seismological Survey (AMGI & CSS) Croatia Dr. Andrija MohorovicicGeological Survey Department (GSD) Cyprus Dr. George PetridesGeophysical Institute of the Academy of Sciences (GFU) Czech Republic Dr. Jan ZednikInstitute of Physics of the Earth, Brno (IPE) Czech Republic Dr. Jan SvancaraNational Survey and Cadastre, Copenhagen (KMS) Denmark Dr. Soren GregersenNational Research Inst. for Astr. and Geophysics (NRIAG) Egypt Prof. Ali TealebInstitute of Seismology (ISUH) Finland Dr. Pekka HeikkinenSeismic Risk Evaluation for the Safety of Nuclear Facilities (BERSSIN) France Dr. Catherine Berge-ThierryBureau Central de Sismologie Français (BSCF) France Dr. Michel CaraBureau de Recherches Géologiques et Minières (BRGM) France Dr. Pascal DominiqueLaboratoire Central des Ponts et Chaussées (LCPC) France Dr. Pierre-Yves BardInstitute of Geophysics (TIF) Georgia Prof. Tamaz ChelidzeBGR Seismologisches Zentralobs. Gräfenberg (BGR) Germany Dr. Klaus KlingeNational Observatory of Athens (NOA) Greece Dr. George StavrakakisUniversity of Thessaloniki (AUTH) Greece Dr. Manolis ScordilisInstitute of Engineering, Seismol., and Earthq. Engineering (ITSAK) Greece Dr. Christos Papaioannou Icelandic Meteorological Office (IMO) Iceland Dr. Ragnar StefanssonDublin Institute for Advanced Studies (DIAS) Ireland Dr. Peter ReadmanGeophysical Institute of Israel (GII) Israel Dr. Yefim GittermanOsservatorio Geofisico Sperimentale (OGS) Italy Dr. Marino RussiStoria Geofisica Ambiente srl (SGA) Italy Dr. Emanuela Guidoboni Geophysics Centre at Bhannes (SGB) Lebanon Dr. Alexandre SursockDirection Environnement Urbanisme et Construction (DEUC) Monaco M. Philippe MondielliCentre National de la Recherche (CNR) Morocco Dr. Nacer JabourNorwegian Seismic Array (NORSAR) Norway Dr. Jan FyenUniversity of Bergen (BER) Norway Dr. Jens Havskov Instituto de Meteorologia (IMP) Portugal Dr. Maria-Luisa SenosInstituto Superior Tecnico (IST) Portugal Dr. Joao FonsecaNational Institute for Earth Physics (NIEP) Romania Dr. GheorgheKing Abdulaziz City for Sciences and Technology (KACST) Saudi Arabia Dr. Tariq Al-KhalifahGeophysical Institute of the Slovak Academy of Sciences (GI-SAS) Slovakia Dr. Peter LabakAgencija Republike Slovenije za Okolje (ARSO) Slovenia Dr. Ina CecicUniversidad Politecnica de Madrid (UPM) Spain Dr. Belen Benito OterinoInstitut Cartografic de Catalunya (ICC) Spain Dr. Antoni RocaSchweizerischer Erdbebendienst (SED) Switzerland Dr. Manfred BaerRoyal Netherlands Meteorological Institute (KNMI) The Netherlands Mr. Reynoud SleemanKandilli Observatory and Earthquake Research Institute (KOERI) Turkey Prof. G. BarbarosogluEarthquake Research Institute (ERD) Turkey Dr. Ramazan DemirtasBritish Geological Survey (BGS) United Kingdom Ms. Alice WalkerMontenegro Seismological Observatory (MSO) Serbia and Montenegro Dr. Branislav Glavatovic

Key Nodal MembersLaboratoire de Détection et de Géophysique (LDG) France Dr. Bruno FeignierGeoForschungsZentrum (GFZ) Germany Dr. Winfried HankaIstituto Nazionale di Geofisica (INGV, Roma) Italy Dr. Marco OliveiriIstituto Nazionale di Geofisica (INGV, Milano) Italy Dr. Massimiliano Stucchi Center of Geophysical Computer Data Studies (CGDS) Russia Dr. Alexei Gvishiani Instituto Geografico Nacional (IGN) Spain Dr. Emilio Carreno

Corporate Members Mediterranean Re Ireland Ms. Karen Crawford

Members by Right European Seismological Commission (ESC) - Ms. Alice WalkerObservatories and Research Facilities for European Seismology (ORFEUS) - Dr. Bernard DostInternational Seismological Centre (ISC) - Dr. Ray Willemann

EMSC,coordinator

of an E.C. funded project

EMSC,specialized European Centre

for the Open Partial Agreement

Marmureanu