Evidence for Quaternary liquefaction-induced features in the epicentral area of the 21 May 2003 Zemmouri earthquake (Algeria, M w = 6.8

J Seismol (2009) 13:161–172DOI 10.1007/s10950-008-9134-y

ORIGINAL ARTICLE

Evidence for Quaternary liquefaction-induced featuresin the epicentral area of the 21 May 2003 Zemmouriearthquake (Algeria, Mw = 6.8)

Y. Bouhadad · A. Benhamouche · S. Maouche ·D. Belhai

Received: 11 February 2007 / Accepted: 18 August 2008 / Published online: 25 September 2008© Springer Science + Business Media B.V. 2008

Abstract Evidence of ancient liquefaction-in-duced features is presented in the area of the2003 Zemmouri earthquake (Mw 6.8). This earth-quake was related to an offshore unknown 50-km long fault. A 0.55-m coseismic coastal upliftwas generated and extensive liquefaction has beeninduced in the most susceptible area which corre-spond to the seaside and along the hydrograph-ic network, mainly the Sebaou and Isser valleyrivers. Field investigations allowed us to identifypast liquefaction traces in the Quaternary de-posits. The observed features are represented bysand dikes, sills, and sand vents as well as well-preserved sand boiled volcanoes. In this work,we also describe the alluvial environment, thehosted localized stratigraphic layer, the morphol-ogy and the geometry of the observed features, as

Y. Bouhadad (B) · A. BenhamoucheNational Center of Applied Research in EarthquakesEngineering (CGS), 1, Rue kadour rahim, H. Dey,Algiers, BP.252, Algeriae-mail: [email protected]

S. MaoucheCRAAG, Route de l’Observatoire, BP.63,Bouzaréah, 16340 Algiers, Algeria

D. BelhaiUSTHB, FSTGAT, BP.32, El-Alia, Algiers, Algeria

well as the observed deformation (settlement) ofthe hosted layers that are among characteristicsof the seismically induced features as describedin worldwide examples. Our observations rep-resent a step towards paleoseismological studiesin the region knowing that the May 21st 2003Zemmouri earthquake is produced by an offshorefault where a direct study of the seismogenic faultis inaccessible.

Keywords Algeria · Paleoliquefaction ·Sand dikes · Quaternary deposits · Earthquake ·Paleoseismic perspectives

1 Introduction

Pre-historical earthquakes can be retrieved eitherin geomorphic (coseismic uplifts and fault scarps)or in geologic (sedimentary features) records (e.g.,Meghraoui and Crone 2001; Obermeier 1996).Offshore located active faults are not accessiblefor a direct study, while it may be the cause ofhuge damage in coastal cities and induce geologi-cal effects such as shoreline uplifts and liquefac-tion in coastal areas. The latter can be used toinfer the long-term behavior and seismic historyof such structures. The use of paleoliquefactionfeatures in studying inaccessible fault has beenused in the case of New Madrid buried/bind fault(Obermeier 1996) and in south Carolina coastal

162 J Seismol (2009) 13:161–172

plain (Talwani and Schaeffer 2001). In general,earthquakes of magnitude Ms ≥ 5.5 may trig-ger liquefaction (Ambraseys 1988). Nevertheless,in Algeria, liquefaction is solely associated withstrong earthquakes such El-Asnam 1980 (Ms =7.3) (Philip and Meghraoui 1983) and Zemmouri2003 earthquake (Mw = 6.8) (Bouhadad et al.2004; Machane et al. 2004). Moreover, the in-vestigation of contemporary documents revealedthat the Jijeli 1856 (Io = VIII–X) earthquake(Mokrane et al. 1994; Harbi 2001) triggeredliquefaction, while there is no trace of liquefac-tion reported during the recent moderate earth-

quakes of Chenoua 1989 (Ms = 5.7), Constantine1985 (Ms = 5.7), Mascara 1994 (Ms = 5.6), AinTémouchent 1999 (Ms = 5.6), and Beni-Ourtilane2000 (Ms = 5.7).

Seismic waves shaking may be the sourceof sedimentary deformations affecting soft sedi-ments and generates structures widely known as“seismite” (Seilacher 1969; Plaziat et al. 1990;Plaziat and Ahmamou 1998). The most knownfeatures of seismic origin are those associatedwith the liquefaction phenomena such as fluidescape structures and sand intrusions representedby sand dikes, sills, and sand vents (Estevez

37° 00'N

36°30' N

2° 30' E 3° 30' E

Tunis

ie

0° 0'

Po

rtu

gal

mer méditerranée

Maroc Algérie

Espagne

France

40° N

Lybie

Italie

25 km

F1

F2

F3

F4

5mm

/an

F5

4321

Dellys

Cap Djinet

BoumerdèsAlger

Thénia

Bassin d

e la M

itidja

Blidean A

tlas

Blida

TipazaChenoua

HadjoutMénaceur

m e

r M

é d

i t e

r r a

n é e

Kabylian basement

Sahel

Fig. 1 Map of active faults of the wider region of Algiers(Meghraoui 1988). The star indicates the relocated 2003epicenter (Bounif et al. 2004). F1, Chenoua fault; F2, Blidafault system; F3, Sahel blind fault (Meghraoui 1988); F4,offshore faults (Déverchère et al. 2005); F5, the causative

fault of the 2003 earthquake (Bouhadad et al. 2004; Bounifet al. 2004; Meghraoui et al. 2004; Bellabes et al. submitted.1 = Anticline axis; 2, reverse fault; 3, strike slip; 4, probablefault; 5, blind fault

J Seismol (2009) 13:161–172 163

et al. 1994; Munson et al. 1995; Obermeier 1996,1998; Alfaro et al. 2001; Bezerra et al. 2005).The interest of such induced features is recentlyhighlighted since it can be used in seismic hazardassessment through paleoseismic studies (Servaand Slemmons 1995; Talwani and Schaeffer 2001;Tuttles et al. 2003). Indeed, calculation of pale-oearthquake parameters such as acceleration andmagnitude using paleoliquefaction features is nowpossible through analytical geotechnic procedures(Green et al. 2005; Obermeier et al. 2005; Olsonet al. 2005). Liquefaction is a physical processwhereby non-compacted, fine-grained, and satu-rated sand deposits are transformed from a solidstate to a liquefied state under a sudden increaseof interstitial (pore-water) pressure in the sedi-ment. It is a process during which ground settle-ments often occur and may be spectacular enoughto cause damage to constructions and infrastruc-tures such as the case during the Niigata-Japan1964 earthquake (Youd and Perkins 1978).

The Zemmouri main shock epicenter has beenrelocated inland in a coastal area near the city ofZemmouri (Bounif et al. 2004). It was produced

by a NE–SW trending, SE-dipping fault, locatedoffshore at about 6 to10 km away from the coastalline (Fig. 1) (Meghraoui et al. 2004; Bouhadadet al. 2004; Bellabes et al. submitted). The earth-quake caused a 0.55-m coastal uplift betweenBoumerdes and the city of Dellys. The Boumerdesregion belongs to the Tell Atlas thrust and foldsranges of Algeria. It is characterized by a seismicquiescence for a long time (hundreds of years)during which only few small quakes with Ms ≤5.0 have been mentioned in the seismicity cata-logue (Benouar 1994; Mokrane et al. 1994). In-deed, the latest destructive earthquake of thearea was the 1716 earthquake which destroyedthe city of Algiers where about 20,000 humanlives were claimed. On the other hand, we shouldmention that the historical earthquakes cataloguedo not cover a long time window to include largeearthquakes that determine long return periods(Meghraoui et al. 1988; Meghraoui and Doumaz1996). In this work, we aim to present the paleoliq-uefaction features observed in the epicentral areaof the May 21st, 2003 earthquake, which triggeredextensive liquefaction.

Ain Taya

Reghaia

RouibaDar ElBeida

Ouled Moussa

Boudouaou

Corso

TidjelabineThénia

Si Mustapha

Isser

Zemmouri

M e d i t e r r

a n e a n s e a

Bordj Menail

Naciria

TadmaitDraa BenKhedda

Baghlia

Ben ChoudSidi Daoud

Cap Djinet

Takdemt Dellys

Makouda

Tizi Ouzou

O. S

ebao

u

O. I

sser

O. C

orso

O. B

oudo

uaou

O. R

egha

ia

O. H

amis O. B

oum

erde

s

Legata

10 km

Liquefaction

Hyrographic network

0

ZemmouriBahri

4˚00'E

36˚75'N

37˚00'N

O. T

atar

eg

Boumerdes

12

3˚50'E

Fig. 2 Map of the observed liquefaction features followingthe 21 May 2003 (the filled squares represent the visitedsites where several liquefactions had been observed). Themap shows also the hydrographic network of the studied

area. Arrows 1 and 2 indicate, respectively, the OuedTatareg and Oued Corso rivers sites where the paleoliq-uefaction are described in this work

164 J Seismol (2009) 13:161–172

Fig. 3 Map of theBoumerdes area showingdifferent types of depositsenvironment. The recentQuaternary deposits aresusceptible to liquefactionbecause of the presenceof the table water at smalldepth, the sandy nature ofdeposits, and their grainsize. Arrows 1 and 2 andthe square indicate,respectively, the OuedTatareg and Oued corsowaterways described inthe text and theBoumerdes city area

Boum-erdes

Corso

Boudouaou

Zemmouri

Bordj Menail

Ante-Quaternary deposits

5 km

Cap Djinet

b

Mediterranean sea

12

3˚40'E

36˚75'N

3˚30'E

O.T

atar

egO

.Cor

so

O.B

oum

erde

s

recent Quaternary deposits

ancient Quaternary deposits

1.1 Seismotectonic and geologic setting

The studied area lies within the central Tell Atlasof Algeria characterized by compressional move-ments. The seismic activity is due to active defor-mation and seismogenic thrust faulting attributedto 4 to 6 mm/yr convergent movement along theAfrica–Eurasian plate boundary (Meghraoui andDoumaz 1996; Nocquet and Calais 2004). TheAfrican plate is moving northwestward (∼ 315◦)relative to the Eurasian plate. Many poten-tially active faults have been recognized inland(Meghraoui 1988) and offshore (Déverchère et al.2005) in the vicinity of Algiers region. Figure 1shows the major active reverse faults of Sahel andthe Blida fault system whose NE offshore con-tinuation likely ruptured during the ZemmouriMay 21st 2003 (Mw = 6.8) earthquake (Bouhadadet al. 2004; Bounif et al. 2004; Meghraoui et al.2004). The Algiers region experienced in the pastseveral disastrous seismic events and in particularthe large earthquakes of Algiers in 1716 whichleft about ∼20,000 deaths and Algiers 1365 whichtriggered sea waves flooded the lower parts ofAlgiers, as well as the Blida earthquake of1825 with about ∼8,000 victims (Benouar 1994;Mokrane et al. 1994). Nevertheless, the eastern re-gions of Algiers hit by the Zemmouri earthquake

in 2003 was previously considered as an area oflow level seismicity and the offshore causativefault was an unknown active geological struc-ture. This offshore reverse fault is about 50 kmlong and NE–SW trending (Delouis et al. 2004;Meghraoui et al. 2004; Semmane et al. 2005).The coseismic movement represented by 0.55 mof coastal uplift between Boudouaou and Dellysand sea waves affected the Spanish coast. Thedrainage pattern system of the studied area isconstituted by many waterways trending N–S in

4

5

7

8

9

10 100FAULT DISTANCE (Km)

Youd & Perkins, 1978

Ambraseys,19882003 Zemmouri earthquake

1

6

SU

RF

AC

E W

AV

E M

AG

NIT

UD

E (

Ms

)

Fig. 4 Comparison of the farthest point of liquefactionduring the Zemmouri earthquake with the worldwide dataregressions of Youd and Perkins (1978) and Ambraseys(1988) (redrawn and modified from ISSMFE 1993)

J Seismol (2009) 13:161–172 165

their upstream parts where it rains toward theMediterranean sea (Fig. 2). The Isser and Sebaourivers are among the most important in Alge-ria with large Holocene active alluvial plains. Itconstitutes a susceptible area for liquefaction asconfirmed during the 2003 earthquake. The sea-side beach area is also susceptible to liquefac-tion. The geological setting of the studied areaexhibits three sedimentary units (Fig. 3): (1) Theante-Quaternary deposits include the metamor-phic basement which outcrops in Thenia, the

magmatic rocks of Dellys and Thenia and, fi-nally, the Neogene (Miocene and Pliocene marls)near Thenia and Dellys. (2) The Ancient Qua-ternary deposits include the red sand which out-crops as high terraces in Corso, Boumerdes andZemmouri, and old alluvial terraces. The red sandcontains rounded quartz pebbles. (3) The recentQuaternary deposits correspond to Holocene al-luvial terraces constituted by sand, gravels, clayand mud, and the sandy beach deposits along thecoastline.

Fig. 5 Geographicalsetting of investigatedsites. The shaded zonecorresponds globally tothe urbanized city ofBoumerdes.1 = Waterways,2 = roads, 3 = railway.The arrows indicate theinvestigated sitesmentioned inthe text

TIdjelabine

Draa_Khodja

Boumerdes

Rocher Noir

Domaine Saint Marie

Ou

ed

Co

rso

Oued T

ata

reg

Figuier

0 600m

SITE 1

SITE 2

1 2 3

Oued B

oum

erd

es

France

Algeria

Mediteranean Sea

Morocco 0° 0'

Spain

Po

rtu

gal

Tu

nis

iaLybia

40° N

Italy

166 J Seismol (2009) 13:161–172

2 Induced liquefaction during the May 21st 2003earthquake

During the May 21st, 2003 Zemmouri earth-quake, the liquefaction was extensively observedin the active flood Holocene plains which corre-spond to the seaside and to the waterways valleys(Bouhadad et al. 2004; Machane et al. 2004). At

A

1 m

Clays

Red limons

Red limons

Pedogenic layer

Fine grained sand

paleoliquefaction features (dikes)

Pebble alluvial layer

1

2

1m

Pedogenic layer

Grey sand with pebble

Fine grained yellow sand

B

Fig. 6 a Schematic cross section of the Oued Tatareg out-crops indicating the stratigraphic position of the describedfeatures of Figs. 7 to 10. All the described and observedfeatures (dark lines for dikes) are concentrated just abovethe sand bed which may be the source bed. b Schematiccross section of the Oued Corso outcrop indicating thestratigraphic position of the described features of Figs. 11and 12

20 cm

Clays

Sand dike

red limons

Fig. 7 Photograph and its schematic depiction showingT-shaped liquefaction feature in section view representedby vertical ascents and lateral spreads of liquefied sandmaterial

the surface, sand material has been boiled eitherthrough fissures reaching 20 m in length and 20 cmin width or through sand volcanoes of 10- to30-cm diameter size. In terms of mechanism, weobserve either hydraulic fracturing due to waterpressure or lateral spreading due to riverside slid-ing. Surface breakings are often associated withthe under pressure sand causing damage mainlyto roads. The spatial distribution of the observedliquefaction shows that the liquefied area corre-sponds to a zone of recent Quaternary depositsand river channels (Figs. 2 and 3). On the otherhand, the farthest liquefaction point during theZemmouri earthquake is observed at Tadmait,at about 28.5 km far from the fault rupture (weconsider the location provided by Meghraoui et al.2004 at about 10 km far from the coastline)(Figs. 1 and 2). This observation is compatible

J Seismol (2009) 13:161–172 167

with worldwide data base (Youd and Perkins1978; Ambraseys 1988) (Fig. 4).

3 Description of the observed paleoliquefactionfeatures

Field investigations of paleoliquefaction fea-tures undertaken in the epicentral area of theZemmouri May 21st 2003 earthquake (Mw = 6.8)allowed us to identify several features in Quater-nary deposits. Two sites have been investigatedin detail: the Oued Tatareg and the Oued Corsorivers sites (Fig. 5). Hereafter, we describe theoutcrops and related features.

Pebble alluvial terrace

Red limons

Clays

15 cm

sand dike

Fig. 8 Photograph and its schematic depiction showingsand dike in plan view crossing through pebble alluvialterrace and red limons. The black arrows indicate defor-mation of the hosted layer

white silt

dike

Red limons

15 cm

Fig. 9 Photography and its depiction (plan view of theoutcrop) showing a silt dike crossing through red limon andpebble alluvial terrace layers

3.1 Geographic setting and descriptionof the outcrops

In both investigated sites, the observed pale-oliquefaction features are not accessible on thesurface, and these were revealed by trenchesexcavated during road works. The Oued Tataregoutcrop displays 60 m long and 8 m high (sectionview), where observations are possible in bothsides of the road. On the other hand, the OuedCorso outcrop is 6 m long and 2 m high, whereobservation is possible only on the eastern side. InOued Tatareg site, the paleoliquefaction featuresare very abundant and shows tens of features in-cluding sand dikes, sills, and sand vents observedin the same stratigraphic layer. The section showsfrom the bottom to the top: 4 m fine-grained sand,0.4 m pebble terrace, 1 m of red limons, 0.1 m ofclays, 1.5 m of red limon, and finally 0.4 to 0.5 m of

168 J Seismol (2009) 13:161–172

pedogenic layer (Fig. 6a). In Oued Corso site, thefeatures are less abundant, but we particularly, ob-served a well-preserved 30-cm diameter sand boilvolcanoes and sand dikes. The Oued Corso sitecross section shows a sandy clay layer overlain bygrey sand with a millimeter-sized pebble alluvialterrace (Fig. 6b).

3.2 Description of the identified paleoliquefactionfeatures

The Oued Tatareg channel reveals a wide varietyof paleoliquefaction features including sand dikes,sills, and sand vents. Geometrically, the observeddikes in section are vertical or steeply dipping,and having a tabular form that ranges from a fewmillimeters to 5-cm width. In Fig. 7, one can seea typical T-shaped structure corresponding to avertical 3.5-cm width sand dike. This feature istypical of vertical ascents and horizontal spreads

10 cm

Spreaded white silt

Scraps of peble terrace

Red limon

Fig. 10 Photography and it’s depiction showing decimeter-sized vented silt and dikes through red limons (white color)

30cm

pedogenic layer

c

a

sandy clays layer

b

Fig. 11 Photograph and it’s depiction showing detail ofboiled sand volcanoes features (arrow a) and sand dike(arrow b). Also we can remark bedding deformation(settlement–arrow c). Detail of the boiled sand structure“a” is shown on Fig. 12

of sand liquefied material similar to liquefac-tion features observed during the 1980 El-Asnamearthquake (Philip and Meghraoui 1983). Carefulobservations allow us to note that the dike-filledmaterial is made of millimeter-sized pebbles likelyfrom the sidewall crossed bed. A few meters awayfrom this observation, another sand dike in paral-lel position (Fig. 8) consists of a 5-cm width sanddike crossing through a gravel alluvial terrace andred limon bed. The latter is comparable to thebottom sand layer material of Fig. 6a. Detailedobservations show firstly the folded host crossedpebbles terrace layer and secondly decreasing inpebble size from bottom to the top. Such defor-mation is likely due to settlement of the hosteddeposits during the shaking. Furthermore, Figs. 9and 10 show examples of sand dikes seen in planview and 5-cm diameter silt vents crossing throughthe same layer, respectively. The deposited silt

J Seismol (2009) 13:161–172 169

boiled sand

sandy host layer

sand dike15 cm

Fig. 12 Photograph and its depiction showing detail ofboiled sand volcanoes feature

around the vent ruled out the roots trace expla-nation of these figures. These silt vents presenta tabular internal structure similar to the 2003features which may support their seismic origin(Obermeier 1998). For the case of Oued Corsosite outcrop, we observed a well-preserved 3-cmdiameter sand boiled volcanoes accompanied bycentimeter-width sand dikes (Figs. 11 and 12).The host sandy clay layer shows deformation as-sociated with ground settlement comparable tothose observed during moderate or larger currentearthquakes. This observation is comparable tothat made after the 2003 earthquake and impliesthe seismic origin of liquefaction features.

3.3 Location and magnitude of paleoearthquakes

In order to assess the magnitude and the loca-tion of the causative faults of paleoearthquakes,

we used the qualitative method of Youd andPerkins (1987) which link the liquefaction severityindex (LSI) to the moment magnitude and tothe horizontal distance from the energy source inkilometers (Fig. 4). The liquefaction severity is es-timated or assessed following the observed grounddisplacement in a considered area:

Log LSI = −3.49 – 1.85 log R + 0.98 Mw ≤ 100.Where R is the distance to the energy source andMw is the moment magnitude.

A qualitative description of LSI show that itranges from 0% to 100% from very sparselydistributed minor ground effects to very abun-dant ground effects, mostly sand boils coveringlarge areas. In our case, observations show agenerally sparse and a locally abundant groundeffect corresponding to (LSI = 30–50%). If weconsider the 2003 source, located 10 km awayfrom the coastline (Meghraoui et al. 2004) as thesame source of the observed paleoliquefaction,we obtain a moment magnitude Mw = 6.9 for thepaleoearthquakes.

4 Discussion and conclusion

Field investigations of paleoliquefaction featuresundertaken following the may 21st 2003 (Mw =6.8) earthquake allowed us to identify abundantfeatures in Quaternary deposits in the epicen-tral area of this earthquake. The alluvial envi-ronment, the hosted localized stratigraphic layer,the morphology and geometry of the observedfeatures as well as deformation (settlement) ofthe hosted layers are among the characteristics ofseismically induced features as described in world-wide examples. Occurrence of earthquake lique-faction depends on the depositional environmentand the level of water table. These two factorsare used to define a qualitative susceptibility mapfor the Boumerdes area (Fig. 3). This qualitativeapproach is often used in microzoning studiesfor urban planning purposes (WCC 1984). Thehigh susceptibility zone corresponds to the recentQuaternary deposits which include the recentalluvial terraces and the costal beach deposits.Liquefaction during the 2003 earthquake oc-curred in this area. Additionally, the granu-lometric curves of recent Quaternary deposits,

170 J Seismol (2009) 13:161–172

Stone Gravel Large sand Fine sand Limon Clay100

90

80

70

60

50

40

30

20

10

0200 100 80 50 20 10 5 2 0.4 0.2 0.1 0.08 0.05 0.02 0.01 5µ 2µ 1µ 0.5µ 0.1µ

Sand liquefaction domain

PE

RC

EN

T F

INE

R

GRAIN DIAMETER

1

Fig. 13 Granulometric curves of the recent alluvial de-posits of Oueds Corso and Tatareg (redrawn and modifiedfrom Mansouri 1990). It shows that the grain size of the

recent Quaternary deposits in these two waterways is fa-vorable for liquefaction

particularly in the Tatareg and Corso rivers,are favorable for the occurrence of liquefac-tion (Fig. 13). The moderately susceptible areacorresponds to the ancient Quaternary depositsand, finally, the non-susceptible area concernsthe ante-Quaternary deposits. On the other hand,liquefaction may occur also under static con-ditions (without earthquakes) (Collinson 1994;Rijsdijk et al. 1999). Hence, the seismic originneeds to be proved by satisfying the criteriadefined in well-studied areas. Among these crite-ria, one may mention (Obermeier 1996): (1) theevidence of an upward-directed hydraulic forcethat was suddenly applied, (2) the sedimentarycharacteristic consistent with historically docu-mented observations, and (3) similar features inmultiple locations. The alternative origin may bethe artesian flow and syn-depositional deforma-tion but artesian flow features have a circularinternal structure while the seismically inducedfeatures are tabular (Obermeier 1996). Staticliquefaction may be induced in particular envi-ronments such as the case of fan deltas and insliding deposits (Plaziat and Ahmamou 1998). Thepaleoliquefaction features described in this workhave tabular internal structures and are observedin alluvial deposits. The morphology and size of

the described features are similar to the hydraulicfracturing of the earthquake-induced features de-scribed by Obermeier (1996). The wide variety ofour observed features, their stratigraphic positionin the same layer, the settlement of the hostedlayer, the geologic localization in a susceptiblearea for liquefaction, the comparison with the2003 liquefaction features, and the seismotectoniccontext marked by occurrence of strong earth-quakes suggest that the observed paleoliquefac-tion features are of seismic origin. The observedfeatures are likely induced by a paleoearthquakecomparable in size to the 2003 Zemmouri earth-quake. The observed features open a paleoseismicperspective in the studied area which is character-ized by seismic quiescence during a few hundredsof years and the occurrence of the May 21st 2003(Mw = 6.8) Zemmouri earthquake produced byan offshore active fault inaccessible for direct pa-leoseismic studies. Indeed, the use of paleolique-faction features and related paleoseismic analysisis favored in areas with a lack of surface faultingsuch as the case of blind or buried faults or in thecase of coastal offshore faults location (Obermeieret al. 2005). Finally, this work may constitutea starting point for more detailed investigationswhich should include expansion of the search of

J Seismol (2009) 13:161–172 171

paleoliquefaction features in the epicentral areaand provide isotopic dating in order to identifyand characterize paleoearthquakes.

Acknowledgements The first authors would like to thankProf. M. Meghraoui from IPG Strasbourg for improvingthe manuscript through his suggestions and remarks. Ananonymous reviewer is also thanked for his comments.

References

Alfaro P, Delgano J, Estevez A, Lopez-Casado C (2001)Paleoliquefaction in the Bajo Segura basin (EasternBetic Cordillera). Acta Geologica Hispanica 36 (3–4):233–244

Ambraseys NN (1988) Engineering seismology. Earth-quake Eng Struct Dynam 17:1–105. doi:10.1002/eqe.4290170101

Bezerra FHR, Da fonseca VP, C Vita-Finzi C, Lima-Filho FP, Saadi A (2005) Liquefaction-induced struc-tures in Quaternary alluvial gravels and gravelysediments, NE Brazil. Eng Geol 76:191–208. doi:10.1016/j.enggeo.2004.07.007

Bellabes S, Wicks C, Cakir Z, Meghraoui M (submitted)Rupture parameters of the 21 May 2003, Mw = 6.8,Zemmouri (Northern Algeria) earthquake deducedfrom InSAR. J Geophys Res

Benouar D (1994) 1994, Material for the investigation ofthe seismicity of Algeria and adjacent region duringthe twentieth century. Ann Geofis 38(4):860

Bounif A, Dorbath C, Ayadi A, Meghraoui M, BeldjoudiH, Laouami N, Frogneux M, Slimani A, Alasset PJ,Kharroubi A, Oussadou F, Chikh M, Harbi A, LarbesS, Maouche S (2004) The May 2003 Zemmouri (Alge-ria) earthquake Mw = 6.8: relocation and aftershocksequence analysis. Geophys Res Lett 31:L19605

Bouhadad Y, Nour A, Slimani A, Laouami N, Belhai D(2004) The Boumerdes (Algeria) earthquake of May21, 2003 Mw = 6.8: ground deformation and intensity.J Seismol 8:497–506. doi:10.1007/s10950-004-4838-0

Collinson J (1994) Sedimentary deformational structures.In: Maltman A (ed) The geological deformation ofsediments, pp 95–125

De Mets C, Gordon RC, Argus DF, Stein S (1990)Current plate motion. Geophys J Int 101:425–478.doi:10.1111/j.1365-246X.1990.tb06579.x

Delouis B, Vallée M, Meghraoui M, Calais E, MaoucheS, Lammali K, Mahsas A, Briole P, BenhamoudaF, Yelles K, (2004) Slip distribution of the 2003Boumerdes-Zemmouri, Algeria, from teleseismic,GPS and Coastal Uplift data. Geophys Res Lett31:L18607

Déverchère J, Yelles K, Domzig A, Mercier de Lepinay B,Bouillin JP, Gaulier V et al (2005) Active thrust fault-ing offshore, Boumerdes, Algeria, and its relationsto the 2003 Mw = 6.9 earthquake. Geophys Res Lett32:L04311. doi:10.1029/2004GL021646

Estevez A, Soria JM, Alfaro P (1994) Un nouveau typede seismites dans le Miocène supérieure d’alicante(Cordière bétique orientale, Espagne): les coinsdétritiques. C.R. Acad. Sci. Paris, 318, série II,507–512

Green RA, Obermeier FS, Olso SM (2005) Engineer-ing geologic and geotechnical analysis of paleoseismicshaking using liquefaction effects: field examples. EngGeol 76:263–293. doi:10.1016/j.enggeo.2004.07.026

Harbi A (2001) Analyse de la sismicité et mise en évidenced’accidents actifs dans le Nord-Est Algérien., Thèse deMagistère, USTHB, 189 pp

International Society for Soil Mechanics and FoundationEngineering (ISSMFE) (1993) Manual for Zonationon Seismic Geotechnical Hazards, TC4

Machane D, Bouhadad Y, Oubaiche EH, Hellel M,Amrouche F, Abbès K, Messaoudi M, Cheikhlounis G(2004) Description morphologique des effets induitspar le séisme de Boumerdes (Algérie) du 21 Mai2003 (Mw = 6.8). Mémoires du Service Géologiqued’Algérie 12:133–146

Mansouri M (1990) Approche de cartographie géo-technique de la région de Boumerdes, Mémoired’ingénieur, USTHB

Meghraoui M (1988) Géologie des zones sismiques duNord de l’Algérie: paléosismologie, tectonique activeet synthèse sismotectonique. Thèse doct. Univ. ParisVI, pp.356, France

Meghraoui M, Crone AJ (2001) Earthquakes and theirpreservation in the geological records. J Seismol5:281–285. doi:10.1023/A:1011413822532

Meghraoui M, Doumaz F (1996) Earthquake-inducedflooding and paleoseismicity of the El Asnam(Algeria) fault-related fold. J Geophys Res 101:17617–17644. doi:10.1029/96JB00650

Meghraoui M, Philip H, Albarede F, Cisternas A (1988)Trench investigations through the trace of the 1980 El-Asnam thrust fault: evidence for paleoseismicity. BullSeismol Soc Am 78(2):979–999

Meghraoui M, Maouche S, Chemaa B, Cakir Z, Aoudia A,Harbi A, Alasset PJ, Bouhadad Y, Benhamouda F(2004) Coastal uplift and thrust faulting associatedwith the (Mw = 6.8) Zemmouri (Algeria) earthquakeof 21 May, 2003. Geophys Res Lett 31:L19605.

Mokrane A, Ait Messaoud A, Sebai A, Menia N, Ayadi A,Bezzeghoud M (1994) Les séismes en Algérie de 1365–1992, CRAAG, Alger, Supervised by Bezzeghoud, M.and Benhallou, H., pp. 227

Munson PJ, Munson CA, Pond EC (1995) Paleoliquefac-tion evidence for a strong Holocene earthquake insouth-central Indiana. Geology 23(4):325–328. doi:10.1130/0091-7613(1995)023<0325:PEFASH>2.3.CO;2

Nocquet JM, Calais E (2004) Geodetic measurements ofcrustal deformation in the western Mediterranean andEurope. Pure Appl Geopys 161:661–681

Obermeier SF (1996) Use of paleoliquefaction-inducedfeatures for paleoseismic analysis. An overview of howseismic liquefaction features can be distinguished fromother features and how their regional distribution andproperties of source sediment can be used to inferthe location and strength of Holocene paleoearth-

172 J Seismol (2009) 13:161–172

quakes. Eng Geol 44:1–76. doi:10.1016/S0013-7952(96)00040-3

Obermeier SF (1998) Seismic liquefaction features — ex-amples from paleoseismic liquefaction studies in thecontinental United States. Open-file Rep (US GeolSurv) 98–488

Obermeier SF, Olson SM, Green RA (2005) Field occur-rence of liquefaction induced features: a primer forengineering geologic analysis of paleoseismic shaking.Eng Geol 76:204–234

Olson SM, Green RA, Obermeier SF (2005) Geotechni-cal analysis of paleoseismic shaking using liquefactionfeatures: a major updating. Eng Geol 76:235–261.doi:10.1016/j.enggeo.2004.07.008

Philip H, Meghraoui M (1983) Structural analysis andinterpretation of the surface deformation of theEl-Asnam earthquake of October 10, 1980. Tectonics2(1):17–49. doi:10.1029/TC002i001p00017

Plaziat JC, Ahmamou M (1998) Les différents mécan-ismes à l’origine de la diversité des séismites, leuridentification dans le pliocène du Saiss de Fès etde Meknès (Maroc) et leur signification tectonique.Geodinamica Acta Paris 11(4):183–203. doi:10.1016/S0985-3111(98)80004-1

Plaziat JC, Purser BH, Philobbos ER (1990) Seismic de-formation structures (seismites) in the syn-rift sedi-ments of the NW Red Sea (Egypt). Bull Soc Geol Fr4(8):419–434

Rijsdijk KF, Owen G, Warren WP, McCarrol D, Van DerMeer JJM (1999) Clastic dikes in over-consolidatedtills: evidence for subglacial hydrofracturing at KillingBay, Eastern Ireland. Sediment Geol 129:111–126.doi:10.1016/S0037-0738(99)00093-7

Seed HB, Idriss IM (1982) Ground motions and soil lique-faction during earthquakes, El Cerrito, EERI, Mono-graph Series 5:134

Seed HB, Idriss IM, Arango I (1983) Evaluation ofliquefaction potential using field performance data.J Geotech Eng Div ASCE 109(3)

Seilacher A (1969) Fault-graded beds interpreted as séis-mites. Sedimentology 13:155–159. doi:10.1111/j.1365-3091.1969.tb01125.x

Semmane F, Campillo M, Cotton F (2005) Fault loca-tion and source process of the Boumerdes, Algeria,earthquake inferred from geodetic and strong mo-tion data. Geophys Res Lett 32:L01305. doi:10.1029/2004GL021268

Serva L, Slemmons DB (eds) (1995) Perspectives in paleo-seismology, Assoc. Eng. Geologists, Spec. Publ 6:139

Talwani P, Schaeffer WT (2001) Recurrence rates oflarge earthquakes in the South Carolina CoastalPlain based on paleoliquefaction data. J Geophys Res106(B4):6621–6642. doi:10.1029/2000JB900398

Tuttles MP, Prentice CS, Williams KD, Rena LR, Burr G(2003) Late Holocene liquefaction features in theDominican Republic: a powerful tool for earthquakehazard assessment in Northeastern Caribbean. BSSA93(1):27–46, 1

Woodward Clyde Consultants (WCC) (1984) Seismic mi-crozonation of Ech-Chellif region, Algeria. Reportprepared for C.T.C., Algiers, 1, CTC, Algeria, pp 145

Youd TL, Perkins DM (1978) Mapping liquefaction-induced failure potential. J Geotech Eng Div ASCE104(GTA):433–446

Youd TL, Perkins DM (1987) Mapping of liquefaction se-verity index. J Geotech Eng ASCE 113(11):1374–1392