& H ISTORY IN THE ’S ANCIENT HARBOURS ISSUE TWENTY FOUR : N WINTER 2006, P P. 138-149. NICK MARRINER SEA CASTLE CHRISTOPHE MORHANGE 139 H W E

OUTER HARBOUR Wind rose 1-15 r >15 ate S knots kw knots rea Acknowledgements Abstract Excavations led by the British Museum have been underway at B IV Inner II Sidon since 1998. In tandem with these archaeological surveys, 15 cores mole We thank the Directorate General of have been drilled in and around Sidon’s ancient port areas, with three INNER Antiquities (F. Hussei- main objectives: HARBOUR ni), CEDRE (F60/L58), (1) to elucidate the evolution of the city’s maritime façade and investigate UNESCO World Heri- IX tage (CPM 700.893.1) its coastal palaeogeography; VI V I and the Lebanese (2) to compare and contrast these data with Sidon’s sister harbours, XV British Friends of the , and Tyre; and III VII National Museum for technical and financial (3) to investigate human coastal impacts, and more specifically the prob - support. N. Marriner lem of accelerated coastal sedimentation. Our geoarchaeological Aeolianite XII ridge benefitted from a datasets elucidate a complex history of coastal change and human occu - Leverhulme Study pation. Investigation of the coastal archives has detailed six phases in Abroad Studentship. XIV BRITISH MUSEUM Sidon’s maritime history, between the and Medieval periods. EXCAVATION

Introduction XIII

The great antiquity of the Sidon-Dakerman area is attested by archaeological X material dating back to the Neolithic (Saidah, 1979 ; Doumet-Serhal, 2003) CASTLE (fig. 1). During the Iron Age, Sidon evolved into one of ’s key DOMINANT city-states reaching its apogee between the sixth and fifth centuries BC, at LONGSHORE CURRENT which time it superseded Tyre as Phoenicia’s principal naval hub. While CRIQUE archaeological discovery at Sidon has a long and productive history RONDE beginning in the nineteenth century, the ancient city had never been sys - OPEN SOUTHERN tematically explored. It was only in 1998 that a team of archaeologists XI HARBOUR 0 100 200 m under the auspices of the British Museum and the DGA began large-scale SANDSTONE excavations of the ancient , elucidating a quasi-continuous stratigraphy I RIDGE from the Early Bronze Age onwards (Curtis, 2000; Doumet-Serhal, 2003). CORE

20˚ 25˚ 30˚ 35˚ 0˚N In tandem with the terrestrial excavations, 15 cores were drilled in and 4 around Sidon’s port areas, with three main objectives: (1) to expound the coastal palaeogeography (Espic et al ., 2002; Morhange et al ., 2003a; VIII ˚N Marriner et al ., 2006); (2) to compare and contrast these data with Sidon’s 35 Phoenician sister harbours (Marriner and Morhange, 2006; Marriner et al ., 2006); and (3) to investigate the problem of accelerated coastal sedimen - Sidon Eastern Tyre tation. Silting up of the harbours played a significant role in the human 30˚N exploitation of the ancient anchorages (Morhange et al ., 2003b; Raban N

e l i 1985 and 1987a). Recent research has shown that ancient societies strived N permanently with the silting problem, and indeed in areas of high sedi - CHALCOLITHIC SITE 0 200 km ment supply it was a constant endeavour to maintain a viable draught OF DAKERMAN depth (Marriner and Morhange, 2006). 1. Sidon’s ancient harbour areas and location of cores. Geomorphological context 3. Above: Sidon and Zire during the Detailed descriptions of the litho- and biostratigraphical datasets, in addi - 1940s (from [31]). Sidon’s coastal plain runs from the Litani river in the south, northwards Below: Sidon and Zire tion to the methods employed, are given in Marriner et al . (2006). Here, towards the Awali river (fig. 2). This low-lying topography, up to 2 in 2005. Note the we briefly summarise the key findings from our recent research km wide in places, comprises a rectilinear coastline and a series of 140 extensive modernisa - looking to better understand coastal dynamics and shoreline evolu - 141 tion of the coastal front faults has oriented the talwegs NW-SE (Dubertret, in both the northern

A E S 1955 and 1975; Sanlaville, 1977). The most impor - harbour and southern

N A E d bay. In the foreground, N e 80 tant watercourse in the Sidon vicinity is the mA h

0R s R0 r Sidon’s outer harbour B e i r u t 1 e E m t T m I a 0 Awali, which tran

3 0 D w lies in the shadow 2 E 6 3 i M n a m zone of Zire island. t sits ~130 x 10 m of sediment per year. m i 0 m 40 0 L 1 5 The promontory of A w a l i

A Sidon separates two w a l i Sidon’s coastal physiography makes it an ideal S i d o n coves, the northern L i t a n i 40 location for the establishment of three natural harbour and Poide- T y r e N 0 50 km bard’s Crique Ronde, 80 anchorage havens. Two pocket beaches lie lee - Poidebard & Lauffray, 120 ward of a Quaternary sandstone ridge, partially 1951). N drowned by the Holocene marine transgression Z i r e (Morhange et al ., 2006; fig. 3). To the south of the

W E ancient city this ridge has been breached by the 4 9 2 6 sea to form a small embayment named the Crique 1 5 r 3 u o 7 rb 15 a 12 H Ronde by Poidebard and Lauffray (1951). Whether 14 Wind rose 1-15 13 S i d o n S knots >15 knots 10 or not it was ever artificially protected by harbour - 8 11 m works has never been unequivocally demonstrat - 0 0 m

1 0 5 m ed, a question we elucidate later in this paper. 0 m 3 120 0 2 80 m

0 m

1 m

5 0 North-west of the promontory lies a second bay, protected from the open sea by a prominent sandstone ridge. 580 m in length, this coastal tion in the northern harbour and ridge shields a shallow basin still used to this day. southern cove. a S

t y 40 i y a h 80 120 a n 160

d This northern harbour was the centre of Sidon’s

c

i

r

t activity in antiquity. Poidebard and Lauffray (1951) Results e

m

y identified the vestiges of a closed ancient port The northern harbour h

t

a comprising: b The coastal stratigraphy from N

o 120

160 N (1) a reinforced sandstone ridge; and the northern harbour is com - 0 1000 m (2) an inner mole, perpendicular to the ridge, and Core site posed of five facies. separating two basins. 2. Sidon’s coastal bathymetry. 1 Pocket beach ( ca. 3500-1500 A third harbour area, the offshore island of Zire, is a unique feature of the cal. BC) Sidonian coastal façade (fig. 3). First described by Renan, it was not until The marine flooding of the cove is Poidebard and Lauffray (1951) took charge that a preliminary plan of the dated ~6000 BP. A basal unit com - island was drawn up. A double shelters a series of quarries and prising subtidal sands is analogous to a medium/low energy pocket harbour quays on its leeward side. Underwater surveys by Frost (1973) beach, sheltered by the sandstone ridge. uncovered a collapsed jetty and numerous scattered masonary blocks on the sea bottom. She concluded that the island had not only served as a 2 Artificial Bronze Age cove ( ca. 1500-1000 cal. BC) quarry and harbour but also supported a number of constructions. A fall in energy dynamics is translated by a rise in the silts fraction. This Carayon (2003) has recently undertaken new research on the island, and facies corresponds to the Middle to Late Bronze Age proto-harbour describes six quarry zones. We recently dated an uplifted marine notch (Frost, 1995, Raban, 1995), with possible reinforcement of the sandstone (+50 cm) on these quarry faces, pertaining to a short-lived sea-level oscil - ridge improving the quality of the anchorage. Small boats would have lation around 2200 years BP (Morhange et al ., 2006). These data are in con - been hauled onto the beach face, with larger vessels being anchored in trast with Tyre, where submergence of ~3 m is recorded since the embayment. (El Amouri et al ., 2005; Marriner et al ., 2005). Calibrated radiocarbon years 3 Closed Phoenician to Roman har - 2 Bronze Age pocket beach bours (ca. 1000 cal. BC-300 cal. AD) The molluscan fauna is diverse, with tests from a range of ecological con - 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 0 Between the Phoenician to Roman texts including subtidal sands, the upper muddy-sand assemblage in periods, a net change is sheltered areas, the upper clean-sand assemblage, the silty or observed in sedimentary condi - 142 muddy-sand assemblage and the lagoonal assemblage. Marine 143 tions with a marked shift to silts and lagoonal ostracod taxa persist into this unit and are accompanied by a

200 200 fine sands, concomitant with a low gradual rise in coastal taxa. Sporadic tests of marine species indicate con - energy environment. A rise in lagoonal tinued communication with the open sea. molluscs and ostracods is in compli - ance with anthropogenic sheltering of At no point during antiquity do our sedimentological and palaeoecological the environment by harbourworks. data show evidence for artificial harbourworks in the southern cove. 400 400 O N Persistent age-depth anomalies pro - During the Bronze Age this embayment would have served as a fair I D S vide strong chronostratigraphic evi - weather harbour for the inhabitants of Sidon and Dakerman. The wide

l dence for dredging from the Roman sandy beaches prevalent in the southern cove would have accommodated a ) c i m g c

( period onwards (fig. 4). smaller vessels, drawn from the water onto the beachface. This phenome - o l l o e i

v 600 600 b

non is still practiced today throughout the Mediterranean by fishermen e l w a o l e 4 Closed late Roman and early with light, shallow draught vessels. Larger vessels would have sought shel - e s b

n h a t Byzantine harbours (ca. 300-600 cal. ter in the better protected northern harbour. e p e m

D AD) R E T Y 800 800 Accentuation of low energy condi - 3 Prograding shoreline tions is manifested by a plastic clays The final unit comprises shelly sands with an important rise in the gravels unit. The silts and clays fraction com - fraction concomitant with a prograding shoreline. The molluscan fauna prises >90 % of the sediment. comprises badly preserved tests and shell debris, reworked by wave Radiocarbon ages are often stratigraphi - action. 1000 1000 cally incoherent for this facies. Lagoonal and muddy sand taxa support an Under a context of stable relative sea level and moderate to high sedi - extremely well protected, lagoon-like ment supply, Sidon’s southern cove gradually prograded seawards, harbour. dimishing the cove to its present disposition. The absence of artificial har - 1200 1200 bourworks and relative persistence of middle energy coastal dynamics 7000 6000 5000 4000 3000 2000 1000 BC/AD 1000 2000 5 Exposed Islamic harbour means that this littoral deformation is less pronounced than the northern Calibrated radiocarbon years B.C./A.D. A sharp rise in the gravels and coarse harbour. Natural cove/protoharbor phase sands fractions appears concomitant Roman and Byzantine protected and dredged harbor with a gradual demise in harbour Discussion Sea-level envelope (after Fleming et al., 1998) maintenance. Biostratigraphic data Sidon dates and 95% confidence levels Tyre dates and 95% confidence levels corroborate a reopening of the envi - Our geoarchaeological datasets elucidate a complex history of coastal ronment to the influence of offshore change and human occupation. Harbour management advances are clear - ~3 m post-Roman tectonic collapse (Tyre) 4. marine dynamics, with numerous taxa ly translated by distinctive sedimentary facies and faunal suites. The sedi - Chronostratigraphic from the subtidal sands assemblage. This facies could correspond to the mentary history of the northern harbour details six periods. evidence for Roman gradual demise of Sidon as a commercial centre during the Islamic peri - and Byzantine dredg - ing of Sidon and Tyre’s od. 1 Bronze Age proto-harbour phase ancient harbours. The At the time of Sidon’s foundation, during the third millennium BC, mari time older radiocarbon Southern cove harbour technology was still very much in its infancy (Wachsmann, 1998; group corresponds to a naturally aggrading The southern cove is characterized by three lithostratigraphical units. Marcus, 2002 ; Fabre, 2004-2005). Existing Bronze Age evidence from the marine bottom. Quasi- shows a clear pattern of environmental determinism, whereby absence of chronos - 1 Holocene transgression (ca. 4500 cal BC) populations founded settlements in proximity to naturally occurring tratigraphic record between BC 4000 to The sandstone substratum is overlain by a pebbly sand unit marking the anchorages such as leaky lagoons, estuaries and protected pocket beaches 500, coupled with per - Holocene marine transgression of the cove. This subsequently grades into (Raban, 1987b, 1990). At Bronze Age Sidon, the northern and southern sistent age depth a coarse shelly sand unit typical of a shallow marine embayment. pocket beaches were ideally predisposed to serve as proto-harbours. inversions, are inter - preted as evidence of The northern cove afforded the best natural shelter for larger merchant harbour dredging. boats during storms. Conversely, the sandy beaches prevalent in the southern cove would have accommodated smaller vessels, drawn from northern harbour underwent significant changes with the edifice of an the water onto the beachface. inner artificial mole perpendicular to the sandstone ridge. This yielded an extremely well-protected inner basin, translated in the geological archive 2 Dating the beginnings of the artificial harbour by a silt facies containing lagoonal molluscs and microfossils. Towards the end of the Late Bronze Age and the Early Iron Age 144 145 (~1200-1000 BC), expanding international trade prompted coastal popu - Under these closed conditions sedimentation rates rose significantly (~1 lations into modifying these natural anchorages. For example, the mm/yr during the mid-Holocene compared to ~10 mm/yr for the Roman Phoenician mole at Atlit has been dated to the ninth century BC (Haggai, period), not least because of the overriding confinement, but also linked 2006). Another archetype derives from Tabbat el-Hammam in , to increasing human use and abuse of the surrounding watershed that where a quasi-identical mole has been dated to ca. the ninth/eighth cen - flushed sediment into harbour depocentres. turies BC (Braidwood, 1940). More speculative examples of harbour infra - structure are known from the Levantine coast (Frost, 1995). In Sidon’s In addition to artificial dredging, engineering solutions to the siltation northern harbour, transition from shelly to fine-grained sands appears to problem have been asserted, although many of these remain speculative be the first granulometric manifestation of human coastal modification. A (Blackman, 1982 a and b). In Sidon, Poidebard and Lauffray (1951) iden - single radiocarbon date constrains this alteration to the Middle Bronze tified a flushing channel, linking the northern basin with the open sea. The Age (~1700 cal. BC) and must be confirmed by further data. The bio-sedi - stratigraphy shows dredging and coeval desilting infrastructure to have mentological datasets show that it was the northern harbour, naturally pro - been insufficient in eradicating the problem; two thousand years later, tected from the open sea by a sandstone ridge, which became the city’s rapid silting up means that the majority of the ancient basin is now buried major port basin. Conversely, there is no sedimentary evidence for har - beneath the Modern city centre. bourworks in the southern cove. 5 Byzantine apogee Difficulty in dating this first phase of artificial confinement appears concur - While Bronze Age populations benefited from Sidon’s geological endow - rent with two complimentary dynamics (1) modest artificial harbourworks ment, Byzantine societies inherited the Romans’ rich maritime savoir faire . during the Middle and Late Bronze Age or (2) intense dredging during the The Byzantine period is marked by advanced reinforcement of the Roman and Byzantine occupations. antecedent port infrastructure, with a notable persistence of Roman tech - nology and its opulent legacy of engineering achievements (Hohlfelder, 3 Absence of Iron Age sedimentary archives 1997). This is corroborated by a plastic clay unit with diagnostic lagoonal Given the relative absence of pre-Hellenistic facies, reconstruction of the fossils, typical of pronounced confinement. These geological data sup - northern harbour’s phoenician history is problematic. In effect, advanced port archaeological evidence from Beirut’s Byzantine harbour suggesting harbour management techniques during the Roman and Byzantine periods that the Levantine coast was still an important trade zone at this time culminated in the repeated dredging of the northern harbour, removing (Saghieh Beydoun, 2005). Such a trade apex, coeval with advanced port this strata from the sedimentary record (fig. 4). infrastructure and management techniques, leads us to propose a harbour apogee for Sidon during the Byzantine period. Siltation, notably under deltaic and urban contexts, was a well-recog - nized problem in antiquity with four sedimentary sources of note: (1) 6 Harbour abandonment phase local watercourses; (2) regional longdrift currents; (3) erosion of adobe Radical coastal changes are witnessed during the Islamic period, with a constructions and urban runoff; and (4) use of the basin as a base-level gradual reopening of Sidon’s northern harbour. We advance three waste dump. Sidon’s gravels fraction from the Roman period comprises a hypotheses to explain these data, namely (i) cultural, (ii) tectonic and (iii) whole suite of discarded objects, trapped at the bottom of the basin, tsunamogenic: including ceramics, wood, seeds, leather artefacts etc. (i) Historians traditionally argue that the Byzantine crisis (sixth century AD) 4 The Roman revolution and ensuing Islamic conquest (seventh century AD) engendered pro - By Roman times, the discovery and use of hydraulic concrete greatly found changes in the eastern Mediterranean’s trade network (Pirenne, enhanced engineering possibilities, locally deforming coastal landscapes 1937). It was speculated that harbour infrastructure fell into a state of dis - (Brandon, 1996; Oleson et al ., 2004). In effect, at this time we observe repair. A priori our sedimentological data are consistent with these inter - pronounced transition from environmental to anthropogenic determinism. pretations. Nevertheless, recent archaeological and historical research The Romans were able to conceive long breakwaters or offshore harbour tends to moderate the premise of a general decline of ‘Syrian’ harbours basins, Maritima being an example par excellence (Blackman, (Borrut, 1999-2000). Historians believe the Levantine coast to have been 1996). All-weather basins could be constructed at locations where no the cradle of Islamic maritime development, control which opened the natural roadstead existed. During this technological revolution, Sidon’s gates of the Mediterranean (Borrut, 2001). Our data indicate rapid coastal Year (AD) progradation from In conclusion, we would like the sixth century AD to insist on three research 300 400 500 600 700 800 900 1000 1100 Phoenician m N 5 onwards, entraining seawall m advances: 0 12 the disloca - T5 N tion of Sidon’s 146 T1 T4 147 T2 T6 harbour. Although T9 a y 0 200 m (1) It has been demonstrated 10 u s e w smaller, it is difficult T y r e A n c i e n t c a that harbour history can be ) i s l a n d 1

8 T o m b o l o 9 to accurately con - clearly chronicled by diagnos - 1

, Bronze y j t 8 Medieval m i o

Age s strain the exact tic litho- and biostratigraphies. g 0 n coastline castle 1 Submerged o e t K

urban quarters n dimensions of this Indeed, explicit use of the i d

Buried

n m

e m

a

5 ancient k

6 port as the post- 0 coastal sedimentary record a d basin r u a q 0 1000 m s h Byzantine sediment has the possibility to greatly s t r a l a P E record lies beneath enhance our understanding of r

e 4 t f

a the present basin human-environment interac - ( and is not accessi - tions; we therefore postulate 2 BH II ble at present. Ancient harbour's geoscience to be a powerful BH IV EARLY BYZANTINE TECTONIC maximum extension tool in expounding the spatial PAROXYSM RELATIVE DEMISE OF SIDON AND TYRE 0 (ii) Data of the opening organisation of harbour areas 300 400 500 600 700 800 900 1000 1100 of Sidon’s harbour Inner and their coastal evolution northern Year (AD) chronologically con - harbour through time. tradict the Early Earthquake (intensity > 8) Earthquake (intensity < 8) Tsunami BH IX BH V Byzantine Tectonic (2) For Sidon, a clear transition BH I BH VI Paroxysm dated to the fourth to sixth centuries AD (Pirazzoli, 1986, is manifest in the stratigraphy 5. AD 300 to BH XV 1000 earthquake and Pirazzoli et al., 1996). In effect, this opening appears to be later, after the BH III BH VII from environmental determin - events affecting sixth century AD. Fig. 5 plots earthquake and tsunami events on the ism during the Bronze Age, the Phoenician coast Levantine coast illustrating that the fourth to eleventh centuries were chara c- BH XII through various intermediary (data compiled from Sandstone Ambraseys, 1962; terised by repeated seismic shocks, possibly provoking partial harbour ridge BH XIV phases, to full anthropogenic Plassard and Kogoj, damage. BH XIII determinism by the Roman 1981; Russel, 1985; BH X period. Guoidoboni et al ., Murex Tell 1994; Soloviev et al ., (iii) This discrepancy leads us to moderate tectonic mobility in favour of 2000). tsunamogenic impacts. In effect, more than ten tsunami struck the (3) Emerging geological data Levantine coast between the fourth to eleventh centuries AD, severely Open southern delineates a largely analogous Dominant harbour damaging harbour infrastructure. Notwithstanding these catastrophist sce - swell BH XI chronostratigraphic pattern for narios, Sidon port is still in use today, 5000 years after its foundation. many of the Mediterranean’s BH VIII ancient ports. The most impor - Conclusion tant factors in explaining Sandstone coastal progradation are not Although Sidon had three port complexes during antiquity, the northern ridge small scale relative sea-level

basin, sheltered by an extensive aeolianite ridge, was naturally predis - Dakerman varia tions but anthro - posed to become the main seaport from the Bronze Age onwards. As pogenic sediment supply with many of the ancient seaports on the Levantine façade, Sidon’s northern 6. Sidon’s and harbour management. reconstructed harbour harbour is an example of a buried urban harbour par excellence . limits in antiquity. Diagnostic harbour facies have been elucidated around the fringes of the present fishing harbour, for a distance of 100 m inland (fig. 6). Coastal progradation of the port coastline after the Byzantine period, diminished the size of harbour to its present dimensions. As at Tyre, this land encroachment accommodated urban growth during the medieval and modern periods. C. Doumet-Serhal, 2003, “Sidon- the persistence of Roman technolo - J. Plassard, B. Kogoj, 1981, K. W. Russell, 1985, “The earth - British Museum Excavations 1998- gy”, Byzantinische Forschungen Catalogue des seismes ressentis quake chronology of and 2003”, Archaeology and History in Internationale Zeitschrift für au Liban , Ann. Sism. Observ. Northwest Arabia from the 2 nd Lebanon, 18, p. 2-19. Byzantinistik , 24, p. 367-389. Ksara, CNRS, Beyrouth. through the mid-8t h century A.D.”, Bulletin of the American School of REFERENCES L. Dubertret, 1955, Carte E. Marcus, 2002, in: E. C. 148 H. Pirenne, 1937, Mahomet et 149 Oriental Research 260, p. 37-59. géologique du Liban au 1/200,000, M. van der Brink, T. F. Levy Charlemagne , Presses Universi- N. N. Ambraseys, 1962, “Data for avec notice explicative, Ministère (Eds.), and the Levant, taires de France, Paris. th R. Saidah, 1979, “Fouilles de the investigation of seismic sea- des Travaux Publics, Beyrouth. Interrelations from the 4 through Sidon-Dakerman: l’agglomération waves in the Eastern Mediter- rd P. A. Pirazzoli, 1986, “The Early the Early 3 Millennium BCE, chalcholithique”, , 27, p. 29- ranean”, Bulletin of the L. Dubertret, 1975, “Introduction à Byzantine Tectonic Paroxysm”, Continuum International Publishing 55. Seismological Society of America , la carte géologique à 1:50 000 du Zeitschrift für Geomorphologie, 62, Group, London. 52, p. 895-913. Liban, notes”, Mémoire Moyen- p. 31-49. M. Saghieh Beydoun, 2005, “Urban Orient , 13, p. 345-403. N. Marriner, C. Morhange, 2005, planning in a seaport city: Beirut D. J. Blackman,1982a, “Ancient P. A. Pirazzoli, J. Laborel, S. C. “Under the city centre, the ancient from Hellenistic to Byzantine peri - harbours in the Mediterranean”, M. El Amouri, M. El Helou, M. Stiros, 1996, “Earthquake cluste- harbour. Tyre and Sidon: heritages ods”, Bulletin d’Archéologie et part 1, The International Journal of Marquet, B. Nourredine, I. Seco, M. ring in the Eastern Mediterranean to preserve”, Journal of Cultural d’Architecture Libanaises , HSII, Nautical Archaeology and Under- Alvarez, 2005, “Mission d’expertise during historical times”, Journal of Heritage, 6, p. 183-189. 147-184. water Exploration , 11, p. 79-104. archéologique du port sud de Tyr, Geophysical Research, 101, B3, p. sud Liban: résultats préliminaires”, 6083-6097. N. Marriner, C. Morhange, M. P. Sanlaville, 1977, Etude géomor - D. J. Blackman, 1982b, “Ancient Bulletin d’Archéologie et Boudagher-Fadel, M. Bourcier, P. phologique de la région littorale du harbours in the Mediterranean”, d’Architecture Libanaises , HSII, p. A. Poidebard, J. Lauffray, 1951, Carbonel, 2005, “Geoarchaeology Liban , Publications de l’Université part 2, The International Journal of 91-110. Sidon, aménagements antiques du of Tyre’s ancient northern harbour, Libanaise, section des études géo - Nautical Archaeology and port de Saïda . Etude aérienne au Phoenicia”, Journal of graphiques, Beirut. Underwater Exploration, 11, p. 185- K. Espic, C. Morhange, M. sol et sous-marine, Imprimerie Archaeological Science, 32, p. 211. Bourcier, C. Bruzzi, P. Carbonel, T. Catholique, Beyrouth. 1302-1327. S. L. Soloviev, O. N. Solovieva, C. Nammour, C. Doumet-Serhal, N. Go, K. S. Kim., N. A. D. J. Blackman, 1996, in: A. Raban, 2002, “Les ports antiques de Sidon: A. Raban, 1985, “The Ancient N. Marriner, C. Morhange, 2006, Shchetnikov, 2000, in the K. G. Hollum, : A nouvelles données paléo-environ - Harbours of in Biblical “Geoarchaeological evidence for Mediterranean Sea 2000 B.C.-2000 Retrospective after Two Millennia , nementales”, Archaeology and Times”, in: A. Raban (Ed.), Harbour dredging in Tyre’s ancient harbour, A.D., Kluwer Academic Publishers, Brill, Leiden, New York, Köln. History in Lebanon , 15, p. 28-36. Archaeology . Proceedings of the Levant”, Quaternary Research, 65, Dordrecht. First International Workshop on p. 164-171. A. Borrut, 1999-2000, “L’espace D. Fabre, 2004/2005, Seafaring in Ancient Mediterranean Harbours. S. Wachsmann, 1998, Seagoing maritime syrien au cours des pre - , Periplus, London. Caesarea Maritima, BAR e e N. Marriner, C. Morhange, C. ships and seamanship in the miers siècles de l’Islam (VII -X siè - International Series 257, , p. Doumet-Serhal, P. Carbonel, 2006, Bronze Age Levant , A&M University cle): le cas de la région entre Acre H. Frost, 1973, “The Offshore 11-41. “Geoscience rediscovers Phoeni- Press, Texas. et , Tempora”, Annales d’his - Island Harbour at Sidon and Other cia’s buried harbours”, Geology, 34, toire et d’archéologie, 10-11, p. 1- Phoenician Sites in the Light of A. Raban, 1987a, “Alternated river p. 1-4. 33. New Dating Evidence”, The courses during the Bronze Age International Journal of Nautical along the Israeli coastline”, in: C. Morhange, K. Espic, C. Doumet- A. Borrut, 2001, “Architecture des Archaeology and Underwater Déplacements des lignes de rivage Serhal, M. Bourcier, P. Carbonel, espaces portuaires et réseaux Exploration , 2, p. 75-94. en Méditerranée d’après les don - 2003 a “The ancient harbours of défensifs du littoral syro-palestinien nées de l’archéologie, CNRS, Sidon, attempt at a synthesis e e H. Frost, 1995, “Harbours and Paris. dans les sources arabes (VII -XI (1998-2002) ”, Archaeology and proto-harbours; early Levantine s.)”, Archéologie Islamique, 11, p. History in Lebanon, 18, p. 71-81. 21-46. engineering”, in: V. Karageorghis, A. Raban, 1987b, “The Harbor of D. Michaelides (Eds.), Proceedings the Sea People at Dor”, Biblical C. Morhange, F. Blanc, M. Bourcier, of the International Symposium Archaeologist , 50, p. 118-126. R. J. Braidwood, 1940, “Report on P. Carbonel, A. Prone, S. Schmitt- ‘Cyprus and the Sea’ , University of two sondages on the coast of Syria, Mercury, D. Vivent, A. Hesnard, Cyprus, Nicosia, p. 1-21. A. Raban, 1990, “Man instigated south of Tartous”, Syria , 21, p. 183- 2003 b, “Bio-sedimento-logy of the coastal changes along the Israeli 221. late Holocene deposits of the E. Guidoboni, A. Comastri, G. shore of the Mediterranean in ancient harbor of Marseilles Traina, 1994, Catalogue of Ancient ancient times”, in: G. Entjes- C. Brandon, 1996, in: A. Raban, K. (Southern France, Mediterranean G. Hollum, Caesarea Maritima: A Earthquakes in the Mediterranean Nieborg and W. van Zeist (Eds.), th sea)”, The Holocene, 13, p. 593- Man’s role in the shaping of the Retrospective after Two Millennia, Area up to the 10 Century , Istituto 604. Brill, Leiden, New York. Nazionale di Geofisica, Bologna. Eastern Mediterranean landscape , Balkema, Rotterdam, p. 101-112. J. P. Oleson, C. Brandon, S. M. N. Carayon, 2003, “L’île de Ziré à A. Haggai, 2006, “Excavation of the Cramer, R. Cucitore, E. Gotti, R. L. A. Raban, 1995, “The heritage of Saïda: nouvelles données Phoenician harbour at Atlit: findings Hohlfelder, 2004, “The ROMA - ancient harbour engineering in archéologiques”, Archaeology and and reassessment of its foundation CONS Project: a Contribution to the Cyprus and the Levant”, in: V. History in Lebanon , 18, p. 95-114. and purpose”, Journal of the Tel Historical and Engineering Analysis Karageorghis, D. Michaelides Aviv Institute of Archaeology , 33, p. of Hydraulic Concrete in Roman (Eds.), Proceedings of the Third J. Curtis, 2000, “New excavations 43-60. Maritime Structures”, International International Symposium ‘ Cyprus at Sidon”, National Museum News , Journal of Nautical Archaeology , and the Sea’ , University of Cyprus, 10, p. 27-40. R. L. Hohlfelder, 1997, “Building 33, p. 199-229. harbours in the early Byzantine era: Nicosia, p. 139-189.