G?omorphologie : relief, processus, environnement, 2003, n" 4, p. 211-226 Nile delta margin: failed and fluidized deposits concentrated along distributary channels La marge du delta du Nil : concentration de d?p?ts fractur?s et fluidis?s le long des pal?ochenaux fluvi?tiles Jean-Daniel Stanley* Abstract Previous sed?uent studies on the Holocenc Nile delta margin have emphasized fluvial input and coastal processes, but have overloo- ked the failure of depositiortal sequences by syn- and post-dcpositional events. Herein, petrologic analysis of sediment sections in cores collected along the Nile delta margin indicates the presence of disturbed strata. These modified deposits are not randomly distributed, but are concentrated along former Holocene distributary channels, and generally absent between these Nile branches. Sediment failure involving fluidization, mass flow, and fault-offset prevailed from the mid- to late Flolocene, especially since -4,500 years before pre- sent. Bedding disturbance was caused by deposition of sediment with high pore water pressure at and seaward of Nile distributary mouths, with flooding along channel mouths believed to be the major trigger of failure. These geohazards were responsible for des- truction and partial submergence of the ancient cities of Herakleion and Ea.stern Canopus in western Abu Qir Bay from -100 B.C. to 741 A.D. Although floods no longer occur since the closure of the Aswan High Dam, some cities constructed on older water-saturated underconsolidated sediment along the modern delta coast remain prone to failure. Protection measures for expanding population cen- ters on the Nile margin warrant thorough civil engineering surveys, including analyses of physical properties of the sediment substrate, with implementation of viable building codes. Key words; bottom currents, cores, diapirs, distributaries, disturbed strata, fluidization, sediment failure, Herakleion, Holocene. R?sum? Les ?tudes r?alis?es jusqu'? pr?sent sur la marge du delta du Nil ont surtout mis l'accent siu- la contribution des apports fluviaux et l'importance des processus entiers, sans v?ritablement s'int?resser aux ph?nom?nes de d?formalions-fracturations des d?p?ts par des ?v?nements syn- et post-s?dimentaires. L'analyse stratigraphique de plusieurs carottes s?dimenlaires pr?lev?es le long de la marge del- ta?que indique clairement la pr?sence de conciles dont la structure s?dimentaire a ?t? modifi?e. Ces d?p?ts perturb?s n'ont pas une r?partition g?ographique al?atoire: ils se concentrent pr?f?rentiellcment le long des pal?ochenaux fluvi?tiles originels, alors qu'ils sont g?n?ralement absents entre les anciens bras du Nil. La pr?sence de d?formations-fiaciiiralion.-i .-i?dimentaircs, qui impliquent ?les ph?- nom?nes de fluidisation, des mouvements de masse et des d?formations d'ordre tectonique, dues ? l'activit? de failles, est av?r?e pour ce qui concerne l'Holoc?ne moyen et r?cent, plus pr?cis?ment depuis 4500 ans B.P. La modification des structures s?dimenlaires est rendue possible par la mise en place, sur la marge cxteriw des embouchures du Nil, de s?diments enclins ? de fortes pressions par ks eaux interstitielles. ? ce titre, les eaux de crue dans les embouchures sont consid?r?es comme l'?l?ment d?clenchant des d?jorma- tions/fructurations. Ces al?as g?ophysiques furent responsables de la destruction et de la .submersion partielle de l'ancienne cit? d'H?- rakl?ion et de la partie orientale de Canopiis, et l'ouest de la Baie d'Abu Qir, entre - ?00 av .l.-C. et 741 ap. J.-C. M?me si les crues .sont devenues plus rares depuis la mise en place du haut barrage d'Assouan, les quelques villes ?difi?es .sur les anciens corps s?di- menlaires meubles et satur?s d'eau, que l'on trouve le long de la frange c?ti?re du delta moderne, peuvent ?tre expos?es ? ces al?as. La protection des noyaux de population en expansion sur les marges du Nil passe par des op?rations d'ing?nierie adapt?es, qui devront prendre en compte les propri?t?s physiques des s?dimeiUs, afin de respecter les normes de construction viable. Mots cl?s : courants de fond, carottes, diapirs, effluents, structure d?form?e, fluidisation, fissure/fracture s?dimentaire, H?racl?ion, Holocene. ' * Gcoarchacology-Global Change Program, E-206 NMNH, Smithsonian las?tution, ConstiliUion Avenue and luth Street. NW, Washington, DC 20560 USA. E-mail; stanley.danieK?nninh.si.cdii Jean-Daniel Stanley Version fran?aise abr?g?e Les principaux r?sultats de cette recherche portent sur 1 ?) la niise en ?vidence de d?formations et de fracturations affectant les s?diments plastiques situ?s le long des pal?o- bras et des pal?o-embouchures holoc?nes du Nil (fig. 1) et 2?) la quasi-absence de telles perturbations syn- et post- s?dimentaires, dans les zones situ?es entre les anciens [your- relets alluviaux. Deux hypoth?ses sont envisag?es pour expliquer les perturbations s?dimentaires : 1) les crues du Nil et 2) les tremblements de terre et les tsunamis. L'?tude privil?gie la premi?re hypoth?se, ci savoir que les crues annuelles du Nil ont jou? un r?le majeur dans la mise en place des fractures et le d?clenchement d'?coulements en masse des s?diments rendus instables par leur saturation en eau. En outre, cette hypoth?se permet d'expliquer pourquoi les perturbations s?dimentaires sont observ?es pr?f?rentiel- lement le long des anciens chenaux fluvi?tiles, ci proximit? de la marge delta?que. Si des tremblements de terre et des tsunamis (hypoth?se 2) ?taient ci l'origine des perturbations s?dimentaires, leurs impacts fi.e., d?formation des strates) auraient d? ?tre enregistr?s al?atoirement, et ce sur des sec- teurs plus ?tendus de la marge delta?que. En d'autres termes, les perturbations s?dimentaires fi.e., structures flui- dis?es, micro-failles) auraient d? ?tre obsen'?es non seule- ment le long des anciens bourrelets alluviaux mais aussi entre ces derniers, ce qui n'est pas le cas. L'utilisation pluri-s?culaire des "nilom?lres" par les Egyptiens a permis d'obtenir une somme d'informations colossale sur les crues pass?es du Nil, c/ui se produisaient chaque ann?e ci la fin de l'?t? et ? l'autotnne (Popper, 1951). L'occurrence de crues annuelles exceptionnellement hautes (CAEH) a ?t? attribu?e ? des fluctuations pal?oelimaticjues majeures affectant les t?tes de bassin du Nil situ?es en Afrique centrale et orientale (Hassan, 1981 ; Humid, 1984 ; Shahin, 1985). Les travaiLX de Said (1993) ont montr? que les CAEH du Nil cmt ?t? plus fr?quentes ? partir du ?V mil- l?naire B. P. et plus pr?cis?ment entre 3840 B. P. et 3 770 B. P. puis Cl partir de 2900 B.P. L'arriv?e des CAEH est syn- chrone de la mise en place des perturbations .s?dimentaires, au cours de l'Holoc?ne moyen et r?cent. Ev?nements hydro- logiques de haute magnitude, ces CAELl ont pu provoquer des perturbations s?dimentaires au niveau des diff?rents sous-lobes form?s le long de ?a c?te et ?tre ? l'origine de d?fliiviations et du d?placement des chenaux sur la plaine delta?que. A partir d'une synth?se des donn?es de base maintenant disponibles, il appara?t que les effets de charge diff?rentielle, les ph?nom?nes de fluidisation et le d?clen- chement des ?coulements en masse sont ?troitement li?s aux CAELl et ? la rapide progradation de la marge delta?que, en particulier durant l'Holoc?ne moyen et r?cent (Stanley et al., 2003). Les stratifications inclin?es, soulev?es et/ou d?ca- l?es, les plis diapirs et les structures fluidis?es ont pu se for- mer pr?f?rentiellement durant la progression des efienaux vers la c?te, comme cela a pu ?tre monlr? aux embouchures de l'ancien bras de Canopie (fig. 2 et fig. 3). Des pro.spec- lions g?ophysicjues haute-r?solution et des carottages com- pl?mentaires (fig. 4, fig. 5 et fig. 6A) s'av?rent n?cessaires, de fa?on ci pouvoir confirmer (ou pas) si de telles perturba- tions s?dimentaires existent sur la marge du delta du Nil, ailleurs que sur les sites d?j? ?tudi?s. L'essentiel des apports du Nil ? la M?diterran?e a ?t? court-circuit? aussi les inondations du fleuve ne repr?sen- tent-elles plus qu'un ?piph?nom?ne le long de la frange c?ti?re (Stanley et Warne, 1998). Mais des d?formations de masse peuvent toujours se produire dans les secteurs du delta o? abondent les accumulations de s?dinients non- consolid?s fi.c, le long des chenaux fluvi?tiles et sur la marge de la plaine; fig. 6B). Alexandrie, l'une des plus importantes villes c?ti?res d'Egypte, a ?t? ?difi?e sur un substrat stable ?i.e., calcaires pleistocenes (kurlcar)], trois m?tres ou plus au- dessus du niveau de la mer, de telle sorte cpie l'on n'y rencontre praticjuement pas de perturbations s?dimentaires. En revanche, d'autres villes implant?es ? proximit? imm?diate de l'actuelle frange delta?que, en par- ticulier celles construites sur des s?diments satur?s en eau fi.c, non-consolid?s), restent vuln?rables aux al?as naturels associ?s aux mouvements du sol (Waltham, 2002). Tel est le cas Cl Baltim, (ni les constructions, b?ties au d?part ? proxi- mit? de l'ancien bras de Sebennytic, se retrouvent aujour- d'hui sur la frange delta?que, suite au recul important du trait de cine. Il en est de m?me ? Port-Sa?d, ? proximit? de l'ancien bras de Tanitic, dans une zone subsidenle cm la mont?e du niveau marin relatif est rapide. Enfin, cette ?tude insiste sur le fait que des mesures de protection doivent ?tre prises pour permettre l'extension en toute s?curit? des noyaux de population sur la frange c?ti?re du delta du Nil. Les op?rations d'am?nagement devraient donc s'appuyer sur des ?tudes approfondies, pre- nant en compte les propri?t?s physiques des s?diments, de fa?on ? pouvoir p?renniser la viabilit? des constructions. Dans les zones expos?es au risque de imnivements du sol, on pr?conise que les nouvelles constructions et autres grandes structures soient b?ties sur des pieux profond?- ment ancr?s dans le sol. En outre, il serait n?cessaire d'ap- profondir les causes et la nature des fissures observ?es sur les structures grecques, romaines et byzantines, c? H?rak- l?ion et ci l'est de Canopus (fig. 2 et fig. 3), afin de d?ter- miner s'il existe un lien entre ces jissures et les d?forma- tions qui affectent les terrains s?dimentaires situ?s le long du bras de Canopie. Comprendre comment ces anciennes cit?s ont ?t? d?truites devrait donc repr?senter bien plus qu 'un exercice strictement acad?mique. Introduction The Nile, one of the world's longest rivers (6,690 km), flows northward across 35? of latitude and drains the east- central African basin with an area of -2,880,000 km-. The Nile delta, with an area of -20,000 km', was the largest coas- tal depocenter in the Mediterranean until the emplacement of the High Aswan Dam. The coastline along the delta is 225 km in length, and the distance between the delta apex near Cairo and the coast is 160 km (fig. lA). Recent delta deve- lopment and evolution of its coastal sector and contiguous shelf (herein termed delta margin) are the results of several 212 G?omorphologie : relief, processus, environnement, 2003, n" 4, p. 211-226 Nile delta margin; failed and fluidized deposits concentrated along distributary channels key parameters: (1) seasonally variable dispersal of River Nile flow as well as fluvial sediment to the coast and sea; (2) sufficient accommodation space to trap a considerable amount of sediment at Nile distributary mouths on the lower alluvial plain and adjacent coast and shelf; and (3) extensive erosion and redistribution of sediment by strong wave- and storm-induced erosional processes near mouths of the Nile promontories and along the coast (UNDP/UNESCO, 1977, f978; Sestini, 1989; Stanley and Warne, 1998). It is recognized that the arcuate coastline and fluvial- marine environments at the delta front have been shaped lar- gely by the collective interaction of sediment input, relati- vely strong coastal and nearshorc currents, land subsidence, and sea-level rise. To date, however, little attention has been paid to other factors that may have affected the configuration of Holocene deposits on this lower delta margin. These could include failure and disruption of strata by sediment instabi- lity and mass flow by river mouth processes, as well as Nile flooding, earthquake, and tsunami events. This investigation explores the extent to which the present current-smoothed Nile margin may have been modified during its development by soft-sediment deformation and fault offset of strata. If such evidence is found, it would be useful to determine: (1) if such processes occurred contemporaneously with (syn- depositional), or after (post-depositional), active sedimenta- tion; (2) whether such processes occurred randomly along the delta margin or were site specific; and (3) times during the Holocene in which such failure occurred. For this study, an examination is undertaken of petrologic patterns of strata in a suite of sediment cores collected along the delta margin. Shaping of seafloor and sediment patterns by currents Nile delta deposits began to agrade and prograde on Egypt's coast about 7,500 years ago (Stanley and Warne, 1994). The coastline at that time was afl'ectcd by a marked decrease in the rate of world sea-level rise, from ~iO mm/yr to 1-2 mm/yr, with shoreline placement between 10 m lo 15 m below present sea-level stand (Fairbanks, ?989). Nile .sediment input and oc?anographie conditions that drive water mass flow in the SE Mediterranean have been the dominant long-term controls of Nile delta margin physio- graphy and sedimentation patterns. Dominant water flow off the coast (fig. lA) has been linked to regionally prevailing climatic conditions, including increased regional aridity since about 6,000 years before present (yrs. B.P.). The present oc?anographie regime off Egypt's Mediterra- nean margin is a response largely to wave approach (fig. 1 A), where dominant east-directed longshore currents are genera- ted with velocities from 20-50 cm/sec, and occasionally to >fOO cm/sec (Sharaf El Din, 1973; UNDP/UNESCO, 1977). Winter storm waves (fig. lA, inset), vvitli heights to 3 m, approach the coast from the northern quadrant and also acti- vely erode and displace sediment from the Nile delta coast eastward to as far as northern Israel (Carmel et ai, 1984; Nir, 1984; Fano.s, 1986; Stanley, 1989; Sharaf El Din and Mahar, 1997). Seasonal variability of wave approach and topographic irregularities along the delta coast produce locally converging and diverging current patterns that modify erosive activity and sediment displacement patterns (fig. lA). Examples include Abu Qir Bay on the NW Nile delta margin (Frihy et al., 1994) and the inner to mid-shelf sector on the NE delta coast off the Damietta promontory (Murray et al., 1981). Effects of tides (to -30 cm) on the margin are minimal. Sedimentation patterns offshore are the result of dispersal of Nile flood water and sediment from several former distri- butary channels of the Nile. From the early to mid-Holocene, perhaps as many as ten such channels flowed seaward, but not all of them were concurrently active (Toussoun, 1922). Oc?anographie factors, riverflow and sedimentation, influen- cing the coastal configuration of the Nile delta and its conti- guous shelf, are incorporated in a Nile littoral cell model derived for this region (Inman and Jenkins, 1984). The model takes into account sediment transport patterns influenced by river Nile sources, strong easterly flow, and erosion of the delta coast and inner shelf to as far as northern Israel. Climatic change affecting Africa after the mid-Holocene resulted in annual river Nile flood volumes that were redu- ced to one-third, or less, compared to those that had prevai- led in the early Holocene (Said, 1993). This caused a reduc- tion m rate of seaward progradation of the Nile delta margin that, until then, had ranged to -10 m/yr (Coutellier and Stan- ley, 1987). Also altered were avulsion rates for different Nile branches, and promontory development at distributary mouths (Frihy et ai, 1988). Sediment discharge to the coast was seasonal prior to closure in 1965 of the High Dam at Aswan: more than 80% of the Nile's total sediment load was released from August to October (fig. lA, inset), while 20% was distributed during the remaining nine months (Sestini, 1989). Average grain size of suspended load from the river Nile during past annual late summer to early fall flood per- iods was 25% sand, 42.5% silt, and 32.5% clay. Average annual discharge decreased to little more than 100 billion m' by the 19th century, and then further reduced to an average of 84 billion m' during the first half of the 20th cen- tury. At present, very little Nile water (< 5 billion m') and sediment reaches the sea (Stanley and Warne, 1998). Howe- ver, even at present, wave-related processes and erosion maintain a 2.5-7 km-wide zone of active eastwardly-moving sand and finer sediment along the coast and nearshore zone from the delta to the Levant (fig. IC) (Manohar, 1981; Inman et ai, L992; Stanley et al, 1997). Active sedimentation, in some instances, reaches shelf depths of 40 m to 50 m (Mur- ray et ai, 1981). Under such conditions, surficial depositio- nal patterns on the Nile shelf are aligned along major eddy current trends, shelf bathymetry, and coastline configuration (Summerhayes et ai, 1978; Bernasconi and Stanley, 1997). Syn- and post-depositional structures in Abu Qir Bay Although smoothing of the seafloor by bottom current ero- sion tends to mask much of the pre-modern history of coas- tal margins, study of major deltas on coasts of modern world oceans record at least some evidence of disturbed Holocene G?omorphologie : relief, processus, environnement, 2003, n 4, p. 211-226 213 Jean-Dani?l Stanley ^ Waves A Tel Aviv A l\ Nile Delta Sediment input 160 million lons/yr prior to 1964 Discharge JFMAMJJASOND 214 G?omorphologie : relief, processus, environnement, 2003, n 4, p. 211-226 Nile delta margin; failed and fluidized deposits concentrated along distributary cliannels bedding. Such features are commonly concentrated at and seaward of delta mouths, but sometimes also occur on lower land sectors of deltas and their contiguous inner to middle continental shelves. Stratal disruption of this type has been recognized on multi-beam topographic records, side-scan sonar profiles, and high-resolution seismic lines along which sediment cores have been collected (Coleman, 1988; Maes- tro et ai, 2002). To date, however, there have been only limi- ted investigations of the late Quaternary on the Nile coastal margin and its contiguous inner shelf. Most surveys off this delta system, primarily designed for oil and gas exploration, have obtained deep seismic data and borings that focus on strata below Pleistocene deposits (Schlumberger, 1984). Low resolution, shallow penetrating, echo-sounding profiles (12 kHz) have been collected across parts of the Egyptian shelf to help identify surficial sediment patterns and patch reefs (Summerhayes et al., 1978, their fig. 5). Overall, such publi- shed results of the Nile margin have been of insufficient reso- lution and areal coverage to reliably identify and map distur- bed and/or offset Holocene strata. An exception is the recent geoarchaeological investigation in western Abu Qir Bay off the NW Nile delta that examined attributes of sediment strata beneath submerged ancient Greek to Byzantine cities of Herakleion (cited by Herodotus) and Eastern Canopus (Constanty, 2002; Goddio et al, 2003) now at depths of ~5 m to 7 m (fig. 2). These former trade centers were positioned at Canopic mouths of delta sublobes when, in the 7th century B.C., the coast was ~5 km north of the present shoreline. The cities were damaged and submer- ged from -100 B.C. to 741 A.D. (Stanley et al, 2003). Within this position of the bay, bathym?trie, magnetometer, and side-scan sonar exploration (conducted mostly in 1996 and 1997), coupled with a sub-bottom seismic survey (fig. 2; in April and May 2000) and a sediment vibrocore program (fig. 2; in April and May 2001), were undertaken to detail attributes of Holocene deposits. Cores were collected in 7.5 cm diameter alummum tubes. Particularly valuable for this investigation in Abu Qir Bay are high-resolution seismic subbottom profiles obtained with a Triton EdgeTech XStar system. The profiles, collected along 15 north-south and 22 east-west oriented transects, are spaced from about 100 m to 1000 m apart (fig. 2) and penetrate to depths of -10 ms to 15 ms two-way travel time (or -7.5 m to 10 m below the water- Fig. 1 - Nile delta margin. A: wave and wind regimes affecting coas- tal and shelf water flow patterns off Egypt's Nile delta. Seasonal variation of sediment discharge and storm wave power are shown in inset (after Sestini, 1989); B; map of the Nile shelf showing surficial sediment texture patterns (after Summerhayes et at, 1978); C: bathymetry (depth in m) and sediment transport paths along the Nile delta margin (after Inman et at, 1992). Fig. 1 - La marge du delta du Nil. A : r?gimes des tioules et des vents affectant la plate-forme continentale et la frange c?ti?re du delta du Nil. Les variations saisonni?res des apports solides et de l'?nergie des tioules sont indiqu?es sur le graphique (d'apr?s Sestini, 1989) ; B : carte de la plate-forme continentale en avant du delta du Nil, montrant la r?partition des s?diments de surface (d'apr?s Sum- merhayes et al., 1978) : C : bathym?trie (profondeurs en m) et dyna- mique des flux s?dimentaires le long de la marge du delta du Nil (d'apr?s Inman et al., 1992). sediment interface). These account for a total of -350 km of profiles covering nearly 100 km' in the western bay. Two of the east-west profiles are presented here (fig. 3). The location of sediment cores is shown relative to seismic lines and the ruins of the two submerged sites (fig. 2). This geological investigation of western Abu Qir Bay pro- vides valuable baseline information on syn- and post-depo- sitional deformation of Holocene strata localized along for- mer Canopic distributary branches of the Nile. East-west oriented profiles show that the Holocene cover is about 5- 7.5 meters thick, near the Abu Qir peninsula (western bay), and thickens (>10 m) toward the east. A hard basal layer, probably consolidated carbonate deposits of Pleistocene age, disappears at depth along eastern segments of profile lines. This configuration was caused by downward depres- sion, folding, and/or fault offsetting of Pleistocene and Holocene strata east of the peninsula (Hassouba, 1980), resulting from rapid accumulation of delta deposits released from the Canopic branch. After the early part of the 1st mil- lennium A.D., downward-bowing of recent sediment sec- tions shifted toward eastern Abu Qir Bay. Seismic surveys indicate that more than half of the wes- tern bay is covered by near-horizontal stratified muds and sands oriented as sheets lying parallel or subparallel to the seafloor surface (fig. 3A and 3B). These are normally posi- tioned strata of Canopic delta sublobe deposits that accumu- lated above, and buried, irregular (faulted, folded, eroded) late Pleistocene basal sections. Typically, modern bay floor deposits such as these are of low relief (<3 m), the result of erosion and sediment redistribution by strong bottom cur- rents (Fanos, 1986; Smith and Abdel-Kader, 1988). Of special interest to the present investigation are seismic profiles that indicate disturbed, non-horizontal strata in localized, well defined portions of the bay. Two such areas are those that lie just beneath the submerged archaeological sites of Herakleion and Eastern Canopus (fig. 2). One ano- malous surficial feature is the large domed horizon located just east of the submerged Herakleion ruins (fig. 3A). This feature is nearly 800 m wide and 4 m high, with its steeper face sloping to the east. Core 2, collected on this mounded topography, indicates that the upper part of the feature is for- med of sand (at least to the core base at a depth of 3.2 m; fig. 4). In cross-section, the configuration of the asymmetri- cally-shaped mound resembles a broad, low-lying dune. The eastern margin is bordered by a 250-300 m wide channel- like depression (location of core 1; fig. 3A). When viewed together, E-W oriented profiles crossing the dome feature at several places (fig. 3B) identify an elongate N-S oriented sand bank, a probable coastal sand ridge thai once lay paral- lel to one of the Canopic delta sublobe channels lying east of Herakleion. The sand ridge appears closely associated with the bay floor area of offset strata immediately to the west of it. It is of note that these highly deformed strata form the sediment substrate just beneath the Herakleion ruins (fig. 3A and 3B). Seismic profiles indicate that strata directly under the ruins at both Herakleion and Eastern Canopus in the western bay (fig. 2) are disturbed, tUted, and uplifted to the seafloor. Also G?omorphologie : relief, processus, environnement, 2003, n- 4, p. 211-226 215 Jean-Daniel Stanley observed at depth are probable diapirs, large domed post- depositional sediment features that have been squeezed upward (fig. 3A). These now-deformed Holocene strata were originally deposited horizontally at and seaward of Canopic delta mouths, and then modified by geologically recent events. Side-scan sonar images collected in the western bay also provide evidence of anomalous stratification locally cropping out at the bay floor surface, especially north of Herakleion. What appear as large undulations comprise par- tially-exposed strata, probably tilted and mud-rich, that were uplifted to the bay floor. These pockets of older Holocene strata were subsequently eroded by bottom currents, and are now covered by a thin veneer of modern rippled marine sand. A nuclear resonance magnetometer survey made across the western bay area also records distinct anomalies indica- ting offset Holocene stratification at, and just beneath, the seafloor surface in the vicinity of the two submerged cities. Well-defined straight and curvilinear features are mapped at both localities, some trending N-S at Eastern Canopus and ENE-WSW at Herakleion. Diver excavation of one such anomaly at Eastern Canopus showed it to be a long (-100 m), buried, sand-filled trench formed in the under- lying Holocene mud substrate. Cleared of sand, this curvili- near trench is V-shaped in cross-section, about 5 m wide at the top and 2 m deep. A well-defined, rectilinear fault-like break (to -50 cm wide and deep) is present at its base. Such E30M' E30?5' E 30? 6' E 30? 7' E 30? 8' E 30? 9' E30?10' E30?ir 12000 ^ 10000 8000 ^M)u Qir 6000 4000 2000 N31?22' N31?2r N3r20' N31?19' 1480280 Eastern Canopus N31?18' N3r 17' 4000 6000 lamra 8000 10000 N31?16' Fig.2 Map of western Abu Qir Bay showing high-resolution seismic lines (including two profiles shown in bold, see fig. 3), 17 vibro- oore sites and location of submerged Herakleion and Eastern Canopus. Fig. 2 - Carte de la zone occidentale de la Baie d'Abu Qir, montrant la localisation des profils sismiques haute-r?solution (incluant deux profils dessin?s en gras, voir fig. 3), les dix-sept sites o? ont ?t? r?alis?s les carottages vibro-percut?s et les anciennes cit?s submerg?es d'H?rakI?ion et de Canopus (zone orientale). 216 G?omorphologie : relief, processus, environnement, 2003, n? 4, p. 211-226 Nile delta margin: failed and fluidized deposits concentrated along distributary channels breaks at the seafloor, termed crown cracks, are interpreted as surficial expressions of growth and listric faults (Cole- man, 1982; Elliot, 1986). Diver observations indicate this Fig. 3 - Selected portions of west-east oriented seismic lines 01510481 (A) and 01480280 (B) and vibrocores collected in tlie area where the ancient coastal city of Herakleion (location in fig. 2) was damaged and submerged largely by substrate sediment failure. Sectors with horizontal stratification are distinct from those with tilted and deformed strata and diapiric features. Fig. 3 - Extrait des profils sismiques (orient?s W-E) n? 01510481 (A) et n? 01480280 (B) et carottes pr?lev?es par vlbro-percus- sion dans la zone o? l'ancienne ville c?ti?re d'H?rakI?ion (loca- lisation fig. 2) a ?t? d?truite et largement submerg?e ? cause de la fracturation des s?diments sous-jacents. ? noter l'opposition entre les secteurs o? pr?dominent les stratifications horizontales et ceux caract?ris?s par des stratifications inclin?es, d?form?es et affect?es de micro-plis et diapirs. trench formed naturally while the city of Eastern Canopus was still above sea level: the trench walls were lined by human-emplaced mats of fresh to brackish water plant mate- rial, principally phragmites and reeds collected in adjacent wetlands, and the trench was then filled with dune and beach sand. In addition, well-preserved, cloved-hoof bovid (pro- bably cow) tracks and bones of an antelope were present at the mud base of the trench. Disturbed strata in Abu Qir Bay cores as example Sediment cores obtained in western Abu Qir Bay also pro- vide evidence of syn- and post- depositional deformation of Holocene strata. The seventeen vibrocores were collected at and between now-submerged sites of Herakleion and Eastern Herakleion area 0 ms 5 ms 10 ms i????S?i^i.,W^^ 15 ms Horizontal strata n 20 ms 500 m 1000 m 1500 m 2000 m 2500 m Herakleion area 3000 m r 1 0 ms 5 ms 10 ms 15 ms R ^ ^-1 I 01480280 D Core 11 | ^^^^^ ^ deformed '% Horizontal strata ' v., ?y\ 20 ms "Vl ? 500 m 1000 m 1500 m 2000 m 2500 m G?omorphologie : relief, processus, environnement, 2003, n? 4, p. 211-226 mf Jean-Daniel Stanley S-47 S-8 S-18 -13.9 m S-37 S-59 ^^?^'"V ^^1 ^^^B^ '^^^^1 j^^^^M 2 cm 2 cm -15.4 m S-59 -4.2 m -4.8 m -16.0 m -14.5 m 220 G?omorphologie : relief, processus, environnement, 2003, n? 4, p. 211-226 Nile delta margin: failed and fluidized deposits concentrated along distributary channels recognized (fig. 5, S-5, S-18, S-31) to those thai have been subject to extensive liquefaction and where original stratifi- cation has been almost entirely obliterated (S-8, S-37). In addition to flowage features, several sections show fault-off- sets that occurred in the strata at the time of, or after, depo- sition (fig. 5, S-31, S-59). Examples of combined flowage and fault offset are also recorded in several of the twenty- two cores. Undisturbed horizontal lamination in one boring (S-47 in fig. 5, location shown in fig. 6A), typical of most Holoccne core sections, is shown for comparison with the seven units that illustrate various deformed bedding types (fig. 5). Regional distribution of disturbed stratification Northwestern Nile delta Major characteristics of the NW Nile delta margin, bet- ween Alexandria and the Rosetta promontory, during the mid- to late Holocene are: (1) relatively modest tectonic activity, except in the Alexandria region west of the Abu Qir peninsula (Kebeasy, 1990); (2) a major lithologie change, from primarily carbonate to mostly terrigenous material, e;ist of Abu Qir peninsula (Hassouba, 1980; Stanley and Hamza, 1992); (3) rapid accretion of sediment from Nile floods and delta coastal erosion, with materials transported eastward into Abu Qir Bay and farther east onto the eastern Nile shelf (UNDP/UNESCO, 1978; Sestini, 1989; Frihy et al., 1991); and (4) active progradation of Canopic branch sediment in the bay to more than 5 km north of the present shoreline until about 1,500 years ago (Stanley et al., 2003). Lx)ng drill cores were collected landward of the coastline at twenty-three sites between the city of Alexandria and the Rosetta branch (fig. 6A). Although this is the most heavily concentrated area of core recovery on the lower delta, only three (13%) cores (S-71, S-72, S-73) south of western Abu Qir Bay present some deformed and offset layers, most less than 50 cm in thickness. It is of note that the three cores were recovered along the path of the former Canopic branch Fig. 5 X-radiographs of Smithsonian cores collected on and landward of the Nile delta coast. Section in S-47 shows undefor- med laminae; the core was collected north of Burullus lagoon bet- ween distributary branches (see fig. 6A). In contrast, seven Holocene core sections show various disturbed bedding (locations in fig. 6B): S-5, Phatmitic/Damietta branches; S-8, Mendesian branch; S-18, Pelusiac branch; S-31 and S-37, Athribitic branch; and S-59, Bolbi- tic/Rosetta branches. Note fault offsets in S-31 and S-59. Fig. 5 - Radiographies des carottes pr?lev?es par le Smithonian de Wastiington sur la frange c?ti?re et sur la plaine delta?que du Nil. La section de la carotte S-47 montre des laminations non-d?for- m?es : la carotte a ?t? pr?lev?e au nord de l'?tang de Burullus, entre deux bras fluvi?tiles (localisation fig. 6A). A contrario, sept sections de carotte (S-5 : bras fluvi?tiles de Pfiatmitic/Damiette ; S-8 : bras de Mendesian ; S-18 : bras de Pelusiac ; S-31 et S-37 : bras de Athribi- tic : S-59 : bras de Bolbitic/Rosette) montrent que plusieurs types de d?formations affectent les terrains holoc?nes situ?s ? proximit? imm?diate des pal?ochenaux du Nil. of the Nile (fig. 6B). They are also located just landward of where offshore cores and geophysical data (figs. 2 and 4) provide evidence of syn- and post-depositional phenomena at what were once migrating channels of Canopic branches. Elolocenc sections in the other twenty (87%) cores collected in this NW delta region do not record disturbed stratifica- tion. North-Central Nile delta Attributes of the north-central Nile delta margin, between Rosetta and Damietta branches, during the mid- to late Holocene are: (1) relatively low rates of land subsidence, except at the promontories of the two modern distributaries (El Askary and Frihy, 1986; Chen et al., 1992); (2) a smooth arcuate coastline that, in the western half of the sector, sepa- rates Burullus lagoon from the sea (Arbouille and Stanlej', 1991); and (3) the presence of two relict Nile branches, including the larger south-to-north trending Sebcnnytic that formerly bisected the delta, and the smaller northeast tren- ding Athribitic (Toussoun, 1922; Stanley and Warne, 1998). The margin sector has experienced considerable modifica- tion by coastal and shore-parallel current erosion. Characte- ristics of this margin are a continuous sequence of sand bar- riers and the large Baltim dune fields between the eastern shore of Burullus lagoon and Gamasa (fig. IC). A set of 35 long drill cores collected on and south of the coast between Rosetta and Damietta branches were exami- ned (fig. 6A). Of these, twelve (34%) show sediment sec- tions that are deformed and/or offset, most with soft-sedi- ment deformation ranging from 20 cm to 50 cm in thickness. Eleven of the twelve borings are positioned (fig. 6B) as fol- lows: cores S-51, S-56, and S-59 close to the SSE to NNW Bolbitic/Rosetta branches; core S-52 near an unnamed relict channel; cores S-39, S-41, S-43, and S-44 along the path of the relict Sebcnnytic branch; and cores S-31, S-33, and S-37 near the path of the former Athribitic branch. Only one of the cores with disturbed bedding (S-50) is not associated with a relict channel. Northeastern Nile delta During the Holoccne, the dominant characteristics of the northeastern Nile delta margin, from the Damietta branch to the Gulf of Tineh, are: (1) large fault-related subsidence rates (to 5 mm/yr) at the Manzala lagoon and northeastern corner (south of the Gulf of Tineh) of the delta (Stanley and Goodfriend, 1997); (2) input of large volumes of sediment by Nile floods at mouths of several branches, including Bucolic/Damietta, Mendesian, Tanitic, and Pelusiac (Tous- soun, 1922); and (3) delta coastal erosion and transport of fluvial material to the east of this region, with seaward del- taic progradation at an average rate that locally exceeded 10 m/ycar (Coiitellier and Stanley, 1987). This margin, which experiences the highest rates of land lowering, is al?ected by a major structural system involving fault motion (Ncev et ?/.,1976) and active isostatic loading during the late Quaternary. Subsidence produced a lowered G?omorphologie : relief, processus, environnement, 2003, n? 4, p. 211-226 221 Jean-Daniel Stanley 32 A 25 km Mediterranean Sea Smithsonian drill core site 31?t? ^j^?V A-- A -.^.. '^/> ^o^. '^^z^. ^ Modem Nile Delta /, ? -ii^ 30? 31? 32? 32? 31? -7- Trace of ancient Nile ?* distributary :.^ Sebennytic 37 Holocene Nile Delta ? o ?n j!it^i^i3 \ D. % 32? Fig. 6 Smithsonian sediment cores collected at and landward of the Nile delta coast (after Stanley ef a/., 1996). A: location of the 83 borings recovered between Alexandria and tine Gulf of Tineh and examined in the study. A: Alexandria, R: Rosetta, B: Baltim, G: Gamasa, D: Damietta, PS; Port Said are coastal cities; Mar : Mariut, I : Idku, Bu : Burullus, Man : Manzala are lagoons; GT: Gulf of Tineh. B: position of the 22 cores between Alexandria and the Gulf of Tineh with sections of disturbed stratification. Of these, 20 are positioned close to traces of former Nile distributary channels as shown by O. Toussoun (1922) and other sources. Fig. 6 - Carottes s?dimentaires collect?es par le Smithonlan de Washington sur la frange c?ti?re et sur la plaine delta?que du Nil (d'apr?s Stanley et al., 1996). A : localisation des quatre-vingt- trois carottes pr?lev?es entre Alexandrie et le Golfe de Tineh prises en compte dans l?tude. A : Alexandrie, R : Rosette, B : Baltim, G : Gamase, D : Damiette, PS : Port-Sa?d sont des villes c?ti?res. Mar : Mariut, I : Idku, Bu = Burullus, Man : Manzale sont des lagunes. GT : Golfe de Tinefi. B : Localisation des vingt-deux carottes entre Alexandrie et le Golfe de Tineh montrant des d?formations s?di- mentaires. Vingt d?ntre elles sont situ?es ? proximit? imm?diate des trac?s des pal?ochenaux fluvi?tiles du Nil, comme le montrent les travaux de O. Toussoun (1922) et d'autres sources bibliographiques. surface on which a thick sequence (to -50 m) of Nile-deri- ved sediment accumulated during the past ~7 millennia (Stanley and Goodfriend, 1997; Stanley and Warne, J998). The Holocene sediment accumulation rate of ~5 to 7 mm/yr is much higher than recorded elsewhere on the margin. The position of the Holocene shoreline, once 30 km to the south, prograded rapidly seaward. The sediment sequence in this part of the delta provides a record of overlapping delta lobes that formed seaward of several distributary mouths (Coutel- lier and Stanley, 1987; Pugliese and Stanley, 1991). Of twenty-five cores collected east of the Damietta branch (fig. 6A), seven (28%) show strata with soft-sediment defor- mation. As on delta margins to the west, most units display deformed and/or offset layers that are less than 50 cm thick. The following are identified: one (S-5) near the Phatmi- lic/Damietta branches; one (S-8) along the Mendesian branch; one (S-2?) near the Tanitic; and three (S-13, S-18, S-19) east of the Suez Canal in proximity to the Pelusiac (fig. 6B). The seventh boring with disturbed structures 222 G?omorphologie : relief, processus, environnement, 2003, n" 4, p. 211-226 Nile delta margin; failed and fluidized deposits concentrated along distributary channels (S-17) was recovered along tlie NW margin of Manzala lagoon, at some distance from distributary traces. The where, when, and why of strata disruption Petrological analysis of drill cores indicates tiic large majority of Holocenc subsurface sections with disturbed sediment stratification distributed quite specifically on the lower Nile delta. Twenty (91 %) of the twenty-two cores that show sofl-sediment and fluidized structures and/or fault off- sets arc positioned near traces of ancient Nile distributaries that once channelized water northward across the delta. Evi- dence of sediment failure is found along each distributary channel and/or mouth of these former branches, from the Canopic in the west to the Pelusiac in the easternmost mar- gin. In contrast, -99% of recovered Holocene core strata in the sixty-one other cores recovered at a distance from Nile distributaries do not show disrupted stratification. It thus can be reasonably deduced that processes and physical pro- perties of sediment most favorable for sediment failure pre- vailed along ancient distributaries. As to liming of sedijTient failure, the deformed strata are dated from early Holocene (prior to 6,000 yrs. B.P.), until near-present, but determining the exact age of strata and date of their disturbance remains problcmalic. Although all cores have been radiocarbon dated (Stanley et al., 1996), ages have generally proven to be unreliable (commonly too old) and provide only an approximate time of deposition (Stanley, 2001 ), For dating by other methods, it is necessary to take into account the considerable core-to-core variation of total Holo- cene thickness (10 m to 50 m). Herein, a rough estimate of age of core strata is determined by (1) assigning a dale of 7,500 yrs. B.P. to the base of the Holocene section, and then (2) calculating an approximate date of the disturbed layer based on its relative depth (% depth down-section from core top) in that core. This assumes a constant depositional rate and no erosion of strata, which in most cases is an unrealistic scenario. However, if this approach is used, one finds that most deformed horizons are positioned at depths from 60% to 30%< down from core tops, suggesting a very approximate age of deposition ranging roughly from mid- to late Holocenc {i.e. from -4,500 to -2,250 yrs. B.P.). For example, a disturbed stratum in core S-71 collected along the Canopic branch occurs at a relative depth of -38%. from the core top, sugges- ting an approximative depositional date of-2,850 yrs. B.P. As to the timing of deformation of that stratum, it can only be surmised that such an event would likely have occurred since 2850 yrs. B.P. In this respect, it is recalled that archaeological evidence of destruction along the Canopic of ancient cities in Abu Qir Bay, lying to the north of core site S-7f, indicates a time from about 2,000 to 1,300 yrs. B.P. (Goddio el al., 2003). Several processes may have been involved in failure of sediment, including earthquake tremor or sudden loading (weighting) by addition of new sediment during flood, tsu- nami, or storm surge. For example, earthquake and tsunami events (including the famous one in 365 A.D. and in subse- quent periods) were probably significant factors in destruc- tion and submergence of human-built structures in the eas- tern harbour of Alexandria, a seismically more active region 20 km west of Abu Qir Bay (Kcbeasy, 1990; Guidoboni, 1994). Although in recent years there has been a concerted effort to document important ancient earthquake and seis- micity-related phenomena in the Eastern Mediterranean (Guidoboni, 1994; Soloviev et al., 2000), no direct correla- tion has yet been established between such earth tremor and/or tsunami wave surge triggers and sediment failure along the delta margin east of Alexandria. More likely triggers of failure were unstable depositional conditions at delta mouth settings, especially those resulting at times of high flow and extreme high flood. In this respect, observations made in western Abu Qir Bay and described in the earlier section of this article provide information to help interpret events in delta sectors elsewhere along the Nile margin. It is probable that it was failure of unstable sediment and mass flow, mostly at Canopic river mouths, that resul- ted in subsidence and shifts of Holocene sections toward adjacent Abu Qir Bay floor settings. During flooding, mouths of powerful rivers such as the Canopic are characte- rized by hydraulic changes that involve much increased out- flow velocities, bed shear and fluid turbulence along with increased, and commonly prolonged, discharge of denser sediment-laden river waters (Bates, 1953; Wright and Cole- man, 1974). The rapidly deposited materials at and seaward of channel mouths are commonly characterized by excess pore water pressure, especially in clay and fine silt deposits, resulting in underconsolidation and low sediment strength. Rapid addition of a new sediment layer during flooding (dif- ferential loading, weighting) induces change in the particle- to-particle configuration of the underlying channel mouth deposits; this, in turn, leads to dewatering of buried clays and silts and sediment flowage (Morgan et al., 1963). Burial of organic matter, released with fine-grained particles at river mouths, can result in formation of methane gas which contributes to generation of excess pore pressure that can further weaken delta-front deposits. Such conditions, especially at and seaward of delta mouths, can induce a stratum to suddenly fail, even on near- horizontal (<1 to 2 degrees) surfaces (Morgan et ai, 1963; Coleman, 1982, 1988; Maestro el ai, 2002). Thus, even relatively tectonic tranquil regions, such as the Nile margin, experience rotational slumps, mud flows (Wright and Cole- man, 1974; Coleman, 1982), along with growth and normal faults (Maestro et al., 2002). Sudden differential loading by younger distributary mouth sands deposited in such settings can cause, in addition to physical disruption of Holocene strata, deformation of underlying sediment into diapiric domes (fig. 3A; Morgan el al., 1963). It is likely that diapi- ric features and associated tilted beds of delta sublobes in western Abu Qir Bay formed primarily during periods of Canopic channel outgrowth when loading and sediment flo- wage prevailed (Stanley et ai, 2003). It is suggested here that, on the basis of the present core analyses, similar condi- tions of sediment instability and failure also occurred along the other Nile channel mouths at various times during the delta's Holocene formation. G?omorphologie : relief, processus, environnement, 2003, n' 4, p. 211-226 223 Jean-Oaniel Stanley Conclusions and ramifications The major findings in this investigation are (1) the concentration of failed and disturbed stratification located along relict Holocenc Nile distributary channel and mouth settings, and (2) the low frequency of such syn- and post- depositional soft-sediment structures between channel traces. This suggests that annual floods of the river Nile per- iodically played a major role in triggering failure and mass flow of unstable water-saturated sediment that had been pre- ferentially distributed along distributary traces at the delta margin. It is postulated that if earthquakes and tsunamis had been (he significant triggers, their effects, especially defor- med strata, would be recorded more randomly over wider sectors of the delta coast and shelf. Moreover, if this were the case, disturbed bedding should be found in cores not only along, but also between, distributary channels. The use of Nilometers (structures designed to measure the level of the Nile) by Egyptians for many centuries has pro- vided considerable information on ancient floods that occur- red each year in late Summer and Fall (Popper, 1951). Exceptionally high annual floods have been attributed to major paleoclimatic fluctuations affecting Nile headwaters in central and eastern Africa (Riehl and Meitin, 1979; Has- san, 1981; Hamid, 1984; Shahin, 1985). Notable among such events since the 4th millennium B.P. are the unusually high floods that occurred between 3,840 yrs. B.P. and 3,770 yrs. B.P. and since 2,900 yrs. B.P. (Said, 1993). These dales are well within the prevailing mid- to late Holocenc period of delta margin failure indicated in the present study. Such events of large magnitude could have induced sediment fai- lure in different sublobes formed along the coast and also could have caused avulsion and switching of channels in the delta. From the preliminary data now available, it appears that differential loading, fluidization, and mass flow were closely associated with high floods and rapid build-out along the different delta margin sectors, especially in the mid- to late Holocenc (Stanley ct al., 2003). Tilted and uplif- ted, and/or offset beds, diapirs and soft-sediment structures may have formed preferentially during the outgrowth of channels at the coast, such as those recorded at former Canopic mouths. Additional high-resolution geophysical surveys and cores arc now needed to better assess whether such deformation could also have occurred elsewhere along the Nile margin. River Nile flow to the sea has now been virtually cut off and flooding is no longer a major event along the coast (Stanley and Warne, 1998), although mass failure could still affect localized sectors along distributary channels in the delta and at the margin where unconsolidated sediment remains prone to failure. Alexandria, Egypt's major coastal city built on a consolidated base of Pleistocene limestone (kurkar) three or more meters above sea-level, remains relatively safe from such failure. However, other cities close to the present low-lying delta coast, especially those built on water-saturated underconsolidated sediment, remain vulnerable to geohazards associated with sediment instability (Waltham, 2002). Examples include Baltim city undergoing construction near Ihe ancient Scbennytic branch and now positioned along a rapidly eroding coast, and Port Said near the trace of the former Tanitic branch located in an area subject to rapid land subsidence and rela- tive sea-level rise (fig. 6A). Increased protection for such expanding Nile coastal population centers requires tho- rough civil engineering surveys, including analyses of sedi- ment physical properties, and implementation of viable building codes. These precautions will result in placement of new buildings and other large structures on deeply anchored pilings, where needed. It could thus be useful, in this respect, to further investigate causes and nature of fai- lure of the Greek, Roman, and Byzantine structures at Heraklcion and Eastern Canopus as a result of substrate sediment failure along the Canopic branch. Understanding how these cities were destroyed would surely be more than a purely academic exercise. Acknowledgements ? thank the iriemhers of the Nile delta team associated with the Smithsonian's Geoarchaeology-Glohal Change Program, and especially Mr Thomas F. Jorstad for his assistance with Smithsonian cores examined in this study. Also acknowledged are Mr G. Schnepp for providing geo- physical data from Abu Qir Bay, Ms. Mary Parish for help with illustrations, and Ms. Christina Borg for technical assistance with tlic typescript. The manuscript was revie- wed by Dr G. Arnaud-Fassetta, Dr F../. Anthony, Mr T. F. Jorstad, and an anonymous reviewer Funding was provi- ded by grants from the National Geographic Society and the Smithsonian's National Museum of Natural History (Walcott Fund). References Arbouille D., Stanley D.J. (1991) - Late Quaternary evolution of the Burullus lagoon region, north-central Nile delta, Egypt. Marine Geology, ^)9, 4.S-66. Bates C.C. (1953) - Rational theory of delta formation. Tlie American Association of Petroleum Geologisis Bulletin, 37, 2119-2t()2. Bernasconi M.P., Stanley D.J. (1997) - Molluscan biofacies, their distributions and current erosion on the Nile Delta shelf .lournal of Coastal Reseurcli, 13, 1201-1212. Carmel Z., Inman D.L., Golik A. (1984) - Transport of Nile sand along the southeastern Mediterranean eoast. Coastal Enginee- ring, 19 (Vll), chapter 87, 1282-1290. Chen Z., Warne AX,., Stanley D.J. (1992) - Late Quaternary evolution of the northwestern Nile Delta between Rosetta and Alexandria, Egypl..!ournal of Coustid Research, 8, 527-561. Coleman J.M. (1982) - Deltas, Processes of Deposition and Models for Exploration. International Human Resources Deve- lopment Corporation, Boston, 111 p. Coleman J.M. (1988) - Dynamic changes and processes in the Mississippi River delta. Geological Society of America Bulletin, 100, 990-1015. Constanty H. (2002) - Hcracl?ion, les tr?sors de la ville englou- tie. GfO, 283, 148-158. 224 G?omorphologie : relief, processus, environnement, 2003, n? 4, p. 211-226 Coutellier V., Stanley DJ. (1987) - Late Quaternary stratigraphy and palcogeography o? the eastern Nile Delta, Egypt. Marine Geology, 237-275. El Askary M.A., Frihy O.E. (1986) - Depositional phases of Rosetta and Damietta promontories on the Nile delta coast../O??/'- nul ofAfrLcan Earth Sciences, 5, 627-633. Elliott ?. (1986) - Delias. In Reading, H.G. (Ed.); Sedimentary Environments and Fades, 2'"' edition. Blackwell Scientific Publications, Oxford, 113-154. Fairbanks R.G. (1989) -A 17,0()0-year glacio-custatic sea-level record: Influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation. Nature, 342, 637-642. Fanos A.M. (1986) - Statistical analysis of longshore current data along the Nile delta coast. Water Science Journal. 1, 45-55. Frihy O.E., El Fishawi M.iVI., El Askary M.A. (1988) - Geo- morphological features of the Nile delta coastal plain: A review. Acta Adrialica, 29, 51-65. Frihy O.E., Fanos A.M., Khafagy A.A., Koraar P.D. (1991) - Patterns of ncarshore sediment transport along the Nile Delta, Egypt. Coastal Engineering 15, 409-429. Frihy O.E., Moussa A.A., Stanley D.J. (1994) - Abu Qir Bay, a sediment sink off the northwestern Nile delta, Figypt. Marine Geology 121, 199-211. Goddio R, Yoyotte .J., Stanley, J.-D., and others (2003) - Aim Qir I. Periphis, London (in press). Guidoboni E. (1994) - Catalogue of Ancient Earthquakes in the Mediterranean Area up to the 10th Century. Istituto Nazionale di Geofisica, Bologna, 504 p. Hamid S. (1984) - Fourier analysis of Nile Hood levels. Geophy- sical Research Letters, 11, 843-858. Hassan F.A. (1981) - Historical Nile floods and their implications for climtitic change. Science, 212, 1142-1145. Hassouba A.M.B.H. (1980) - Quaternary sediments from the coastal plain of northwestern Egypt from Alexandria to El Omayid. Unpublished doctoral thesis. Imperial College London, 320 p. Herodotus - TItc History, translated by David Grenc, 1987. Uni- versity of Chicago Press, Chicago, 699 p. Inman D.L., Jenkins S.A. (1984) -The Nile littoral cell and man's impact on the coastal zone of the southeastern Mediterranean. Scripps Institution of Oceanography, Reference Series, 31, 1-43. Inman D.L., Elwany M.H.S., Khafagy A.A., GolikA. (1992) - Nile Delta profiles and migrating sand blankets. Coastal Engi- neering 1992, 23rd International Conference, Venice, Italy. 23, 3273-3284. Kebeasy R.M. (1990) - Seismicity. In R. Said (Ed.) The Geology of Egypt. A.A. Balkcnia, Rotterdam, 51-59. Maestro \., Barnolas A., Somoza L., Lowrie A., Lawton T. (2002) - Geometry and structure associated to gas-charged sedi- ments and recent growth faults in the Ebro Delta (Spain). Marine Geology 186,351-368. Manohar M. (1981) - Coastal processes at the Nile delta coast. Shore and Beaches, 49, 8-15. Morgan J.P., Coleman J.M., Gagliano S.M. (1963) - Mudlumps at the mouth of South Pass, Mississippi River: sedimentology. paleontology, slrucUire, origin, and relation to deltaic processes. Coastal Studies Institute Series no. 10, Louisiana State Univer- sity, Baton Rouge, 190 p. Murray S.P., Coleman J.M., Roberts H.H. (1981) -Accelerated currents and sediment transport off the Damietta Nile promon- tory. iV?/??', 293,51-54. Neev D., Almagor G., Arad A., Ginzburg A., Hall J.K. (1976) - Tlie geology of the southeastern Mediterranean. Geological Sur- vey of Israel Bulletin, 68,1-51. Nir Y. (1984) - Recent Sediments of the Israel Mediterranean Continental Shelf and Slope. LJniversity of Gothenburg, Depart- ment of Marine Geology, Sweden, Report 2, 149 p. Popper W. (1951) - The Cairo Nilometer. University of California Press, Los Angeles, 269 p. Pugliese N., Stanley D.,I. (1991) - Ostracoda, depositional envi- ronments and late Quaternary evolution of the eastern Nile Delta, Egypt. // Qualernario, 4, 275-302. Riehl H., Meitin J. (1979) - Discharge of the Nile River: A barometer of short-period climatic varititions. Science, 206, 1178-1179. Said R. (1993) - The River Nile: Geology Hydrology and Utiliza- tion. Pcrgamon Press, New York, 320 p. Schlumberger (1984) - Well Evaluation Conference: Egypt 1984, Middle East S.A. Imprimerie Moderne du Lion, Paris, 39-43 p. Sestini G. (1989) - Nile delta: A review of depositional environ- ments and geological history. In M.G.K. Whateley and K.T. Pic- kering (Eds.): Delias: Sites and Traps for Fossil Fuels. Geologi- cal Society of London, Special Publications, 41, 99-127. Shahin M. (1985) - Hydrology of the Nile Basin. Elsevier, New York, 575 p. Sharaf El Din S.H. (1973) - Geostrophic currents in the sou- theastern sector of the Mediterranean Sea. Ada Adrialica (Sym- posium on eastern Mediterranean Sea), 18, 221-235. Sharaf El Din S.H., Mahar A.M. (1997) - Evaluation of sediment transport along the Nile Delta, ./o???v?rt/ of Coastal Research, 13, 23-26. Smith S.E., Abdel-Kader A. (1988) - Coastal erosion along tlic Egyptian delta, .tournai of Coaslal Research, 4, 244-255. Soloviev S.L., et al. (2000) - Tsunamis in the Mediterranean Sea 2000 BC-2000 AD. Kluwer Academic Publishers, Dordrecht, 237 p. Stanley D.J. (1989) - Sediment transport on the coast and shelf between the Nile Delta and Israeli margin as determined by heavy nimcr??s. Journal of Coaslal Research, 5, 813-828. Stanley D.J., Hamza F.H. (1992) - Terrigenous-carbonate se- diment interface (late Quaternary) along the northwestern mar- gin of the Nile delta, Egypt. Journal of Coastal Research, 8, 153-171. Stanley D.J., Warne A.G. (1994) - Worldwide initiation of Holo- cene marine deltas: Deceleration of sea-level rise as principle factor. Science, 265, 228-231. Stanley D.J., McRea J.E. Jr., VValdron J.C. (1996) -iVi/c Delta Drill Core Sample Databases for ?985-1994: Mediterranean Basin (MEDIBA) Program. Smitlisonian Contributions to the Marine Sciences 37, 428 p. Stanley D.J., Goodfriend G.A. (1997) - Recent subsidence of the northern Suez Canal. Nature, 388, 335-336. Stanley D.J., Mart Y., Nir Y. (1997) ~ Clay mineral distributions to interpret Nile cell provenance and dispersal: II. Coastal plain from Nile Delta to northern \si?c\. Journal of Coaslal Research, 13, 506-533. G?omorphologie : relief, processus, environnement, 2003, n' 4, p. 211-226 225 Jean-Daniel Stanley Stanley D.J., Warne A.G. (1998) - Nile delta in ils destruction UNDP/UNESCO (1977) - Pmceeilingx of Seminar on Nile Delia pilase. Journal of Coastal Research, 14(3), 794-825. Shore Processes. Alexandria, UNDP, 624 p. Stanley DJ. (2001) - Dating modern deltas: progress, problems, UNDP/UNESCO (1978) - Coastal Protection Studies. Project Fin- and prognostics. Annual Review of Earth and Planetary dings and Recommendations. Pms^VKDV/EGYn^/Obj, 483 p. Sciences, 29, 257-294. Waltham ?. (2002) - Sinking cities. Geology Today, 18, 95-10?. Stanley DJ, Schnepp G., .Jorstad T. (2003) - Submergence of Wheeler R.L. (2002) - Distinguishing seismic From nonseismic archaeological sites in Abu Oir Bay, the result of gradual long- soft-sediment structures; Criteria from seismic-hazard analysis. term processes plus catastrophic events. //; Goddio F. and others In Ettensohn, F.R., Rast, N., and CE. Brett (Eds.): .Ancient Seis- (Eds.):/\/7?? Qir I. Periplus, London (in press). miles. Geological Society of America, Special paper, 59, 1-11. Summerhayes C.P., Sestini G., Misdorp R., Marks N. (1978) Wright L.D., Coleman J.M. (1974) Mississippi River mouth Ni(e delta: Nature and evolution of continental shelf sediments. processes: effluent dynamics and morphologic development. Marine Geology, 27, 43-65. .lournal of Geology, 82, 751-778. Tous.soun O. (1922) - M?moires sur les anciennes branches du Nil, ?poque ancienne. M?moire de l'Institut d'Egypte, 4, 212. Article re?u le 21 janvier 2??3, accept? le 2 juin 2003 226 G?omorphologie : relief, processus, environnement, 2003, n 4, p, 211-226 L ._