Many of Greek colonies were founded along the Black Sea coast in the Archaic period (7^th – 6^th century BC) at locations where natural conditions were conducive for thriving socio-economic activities (Avram et al., 2004). Some of them are at present partly submerged by the Black Sea (Fornasier and Böttger, 2002; Oppermann, 2004). Histria is one of the oldest colonies that was along this coast in 657 BC during the Archaic period and is the best preserved archeological site along the Romanian coast, with little anthropogenic alteration since its abandonment during the 7^th century AD. Desciphering the landscape changes during Histria’s lifetime and after its decline is an important challenge for archeologists and coastal scientists as most current scenarios relate the city’s demise to dramatic coastal transformations.

There is a commonly held archaeological assumption that during the long existence (13 centuries) of Histria, the regional topographic configuration was subject to multiple and dramatical natural changes which affected its socio-economic life. A range of archeological and geological evidence suggests Histria was originally built at an open coast location with a morphological configuration suited to high-capacity marine navigation and port activities. Epigraphical and archeological sources document the intensive trades of fish, cereals and other goods such as pottery (Alexandrescu and Schuller, 1990) which wouldnt have been possible without navigation facilities and harbors in the proximity of the ancient city. Nowadays, the acropolis of Histria lies inland from the current shoreline, within Razelm-Sinoe lagoon system (Fig. 1A). Three quarters of the last fortification (4^th c. AD) lie on a green-schist outcrop (former rocky island) while the rest overlies marine sands, landward of the rocky core. The Razelm-Sinoe lagoon system is clearly delineated from the northern part of the Danube delta by the Sf. Gheorghe tectonic fault line (Ivanovici et al., 1968; Săndulescu, 1984). Unlike the evolutionary models conceived for the deltaic lobes developed to the north of the fault line, which are commonly accepted by most of the scientists (Panin, 2003; Giosan et al., 2006), there is a general lack of consensus when refering to the sandy barriers and lagoon system developed to the south, although the archeological finds provide reliable chronological proxies for coastal landscape reconstruction. Precisely datable archeological elements (such as tumuli and defensive walls) were interpreted in relation to their specific location in order to provide the first chronological references in the attempt to reconstruct the approximate position of the paleo-shorelines at Histria (Bleahu, 1963; Coteţ, 1966; Ştefan, 1987; Canarache, 1956; Pippiddi, 1983; Alexandrescu, 1978; Höckmann et al., 1996). Therefore, the majority of the coastal evolutionary scenarios are limited to the time covered by the archaeological record.

The main questions apertaining to the regional environmental changes, as formulated by archeologists in the early 20^th century (Moisil, 1909; Pârvan, 1914; Lambrino, 1922) and which are  still unsolved, are: i) what was the geography of the site and coastline when the Millesians established the core of the settlement; in particular, the initial site founded upon an island, a peninsula or a part of the mainland?, ii)  when did the city become isolated from the Black Sea by progradation of the coastline, iii) what triggered the decline of the city, and iv) how did the surrounding ecology transform during and after the lifetime of the city?

Fig. 1A. Location and geography of the Histria coastal area (marked with black contour and detailed in inset B) within the Danube Delta. B. Morphology of the modern Histria region. The beach ridge orientation (black lines), topographical profiles, fault line position and the absolute ages are shown.

The first attempts to reconstruct the ancient landscape of in and around Histria focused on the city itself and the issue of its connection to the mainland. Pârvan (1915, 1916) and  Lambrino (1938), proposed that the city had been settled on an island, occupying a rocky promontory that had no direct connection with either the mainland or the necropolis. Archeological research conducted in the 1950’s in the area surrounding the city and in the tumular necropolis led to important discoveries that supported a new hypothesis, that of a rocky promontory as part of the mainland at the time of the Greeks arrival (Condurachi et al., 1954). The grounds of this hypothesis were broadened by Canarache (1956) with a re-evaluation of archaeological findings such as aqueducts, submerged tumuli and wall remains (Fig. 2A, B).

Coteţ (1962) and Bleahu (1963) placed the ancient city at the seaward edge of a sandy littoral plain connected to the mainland towards the north and west. Later, Coteţ (1966) refined his model in considering the island as connected to the mainland only towards the necropolis (northern-western part) via a narrow tombolo. Alexandrescu (1978) and Ştefan (1987) examining aerial photographs recognised  new antique roads which pointed to a supplementary  south-western terrestrial connection over the area presently occupied by Nuntasi Lake and western Saele. Independantly, two other works dealing with the Danube delta evolution provided the first, few chronometric ages from the region (Panin, 1983, 2003; Giosan et al., 2006).

The common element in most of the archeological theories is the formation of the Razelm-Sinoe lagoon system, and implicitly of the Histria region, as a result of the barrier systems development which confined the ancient marine gulf. However, the nature of the processes which transformed the ancient gulf into a mosaic of littoral plains, beach ridges and lakes is still under debate. Most of the hypotheses invoke eustatic sea level oscillations as the main controlling factor of morphological changes (Bleahu,1963; Coteţ, 1966; Alexandrescu, 1978, Pippiddi, 1983; Panin, 1983, 2003), whilst a few suggest the combined action of the neotectonics, erosion and coastal progradation as key-processes which contributed to the present day morphology (Canarache, 1956; Giosan et al., 2006). Currently, the paleogeographical evolution of the Histria region remains controversial, with numerous hypotheses and a lack of chronological synthesis of the evolutionary phases.

Fig. 2. A. The age and location of Histria walls and tumuli. Roman aqueducts paths and the coring sites detailed in Fig. 4 are shown. B. Cross-section showing the wall succesion and the archaic and post-archaic archeo-deposits (modified after Alexandrescu, 1978).

In this context, this paper addresses the first two questions mentioned above using new geomorphological, sedimentological and absolute age data. We also present a new interpretation of the Late Holocene evolution of the southern Danube Delta, focused on the Histria region and in the light of the newly-obtained absolute chronology to establish the key-events succession which lead to the actual configuration. The study also proposes a sea level curve for the Histria region taking into account the local tectonics.

 

The Histria region is a low-lying coastal plain developed at the contact between Central Dobrogea (Casimcea tableland) and the Black Sea, comprising beach ridge plains (Saele and Chituc), sandy barriers (Lupilor barrier) and shallow lakes (Sinoe, Histria and Nuntaşi) which together form the southernmost unit of the Danube delta (Fig. 1A, B). The Histria acropolis lies 8 km inland from the present Black Sea shoreline on a green schist outcrop which was chosen by the early Greek colonists to be the political and religious center of the city. This rocky promontory is a former island with an maximum altitude of 7 m asl, currently located on the landward shoreline of the Sinoe Lake which is the southern compartment of the Razelm-Sinoe lagoon complex, the largest lacustrine system (867 km²) formed during the Late Holocene in the western part of the Black Sea. It consists of four lakes (Razelm, Goloviţa, Zmeica and Sinoe) which are delineated by sandy barriers and beach ridge plains, inter-connected by natural and artificial channels. The main water and sediment supply comes from the Sfântu Gheorghe arm (the present southernmost Danube arm) through Dunavăţ and Dranov channels, which before channelisation works (1912) had been secondary natural branches of the Sf. Gheorghe arm (Antipa, 1914). The mean water discharged through the two channels is 48 m³/s  while the lacustrine water discharged into the sea is 41 m³/s (Breier, 1976).

Geologically, the Histria region extends over two main geotectonic units: (i) the North Dobroudjan Orogen (Babadag sedimentary basin composed of Cretaceous limestones), present in the northern part of Sinoe lake, which is delineated to the north by the Sf. Gheorghe fault and to the south by Peceneaga-Camena crustal fault, and ii) the Central Dobrogea (Casimcea Tableland made up of Precambrian green schists), present in the southern sector of Sinoe Lake, extended from Peceneaga – Camena crustal fault (N) to Capidava-Ovidiu fault (S). The northern unit of the Histria region is a smaller compartent of the Babadag sedimentary basin which represents an NW-SE elongated syncline which dips eastwards into the Black Sea Romanian continental shelf (Atanasiu, 1940; Ianovici et al., 1968). The southern morphotectonic unit of Histria is divided by several superimposed anticlines and synclines that influenced the present morphologic configuration, i.g. the area in that the green schist rocky outcrop over which lies Histria’s acropolis is coincident with the Histria – Saele anticline axis, while the Histria lake overlies Fantanele –Saele syncline axis (Dimitriu et al., 2004). The Histria region is located within a highly active neotectonic zone where the subsiding rates of up to 4mm/yr occur, which represent the greatest value recorded along the Romanian coastal zone (Polonic et al., 1999). Taking into consideration the neotectonic movements and the long term sea level rise recent investigations suggest that important archeo-cultural layers are burried at 4-5 m depth (Dimitriu, 2012).

Fig. 3. Shoreline position and coastal configuration of the Histria region (southern Danube Delta) at different times of evolution. The paleogeographical reconstuction is based on the new absolute dating, and on the morphological interpretation of the current research.

Southwards (downdrift) of the acropolis-host rocky promontory has formed the drift-aligned Saele beach ridge plain (3 km wide and 9.5 km long).  The distinctive morphostratigraphical feature of Saele beach ridge plain is the most prominent ridge (C-ridge), NE-SW aligned, which divides the Saele plain into two sets of ridges (Fig. 1B). The beach ridge plain is a morphological feature which generally forms in conditions of wave dominated coasts fronted by a wide, low-gradient continental shelf with an abundant sediment supply (Masselink and Hugues, 2003). At present the Saele beach ridge plain is fronted seaward by Sinoe Lake and Chituc barrier and backed by Istria and Nuntaşi Lakes. These latter two lakes are situated at the contact between Casimcea Tableland and the Histria lowland with lacustrine shorelines generally steep (2-10 m high), cut in loess, which overlies narrow green schists platform in places. Further northward and westward, beyond Istria and Nuntaşi lakes, the mainland consists of a green schist bedrock covered by a 2-15 m of loess. An important topographical characteristic of the region is represented by the gentle and constant dipping of the surface to the E and SE. At 2-5 km seaward of the Saele beach ridge plain lies Chituc beach ridge plain which encloses the Sinoe Lake. Southern Chituc has a similar morphology to Saele plain, with juxtaposed beach ridges of 0.7 – 1.5 m height with dunes on the highest of them; northern Chituc has the same basic morphology but at a lower elevation.

The Histria region currently lies within a dry-temperate climate with a mean annual rainfall of 380 mm and moderate to high seasonal thermal amplitudes (-0.5 °C / 21.8 °C). On the present coast (Chituc) the significant wave height is of 1.43 m (medium-wave energy environment, cf. Vespremeanu-Stroe and Tătui, 2011) while the prevalent northern winds impose a net southward longshore sediment transport of about 0.6 – 0.7 x 10⁶ m³ y^-1 (Dan et al., 2007).

The beach ridges, marshy flats and the lakes from Histria region have been investigated by means of geomorphological, sedimentological and geochronological methods. The coring strategy aimed to obtain basic lithostratigraphic information from all relevant sedimentary units with a focus on the Saele plain and the ancient city’s surroundings. The coring used a Cobra TT percussion device (corerheads of 8 and 5 cm diameter) and boreholes used a 6 cm diameter hand auger. The core database consists of: i) 14 cores of 4 to 7 m depth, within the ancient city area and surrounding marshy flats, ii) 6 cores of 1-3 m depth in the Istria, Nuntaşi and Sinoe Lakes, and iii) 35 boreholes up to 3 m depth distributed across a transect on Saele beach ridge plain. The cores and borehole sediments were described on site (colour, grain-size, texture, main minerals, macrofaunal and artefacts composition) The textural parameters have been determined by sieving. The mineralogic examination was undertaken on 0.03 – 0.50 mm siliciclastic fraction obtained after preliminary sieving and 16% HCl and 10% peroxide treatment for carbonates and organic matter removal. More than 1000 grains from each sample have been immersed in glycerine and inspected using a polarizing microscope.

As the elevation and sedimentary architecture of the prograding beach ridges can be used to derive the past sea level (Goy et al., 2004; Clemmensen and Nielsen, 2010) we dated different sandy ridges and barriers for the reconstruction of both sea-level and coastline evolution. A total of 15 Optically Stimulated Luminescence (OSL) ages and one ¹⁴C age have been determined in order to construct a chronological framework of the regional sedimentary units and to track the shoreline position before, during and after Histria’s occupation. The OSL sampling strategy consisted of transversal profiles over the barriers and beach ridges, parallel to coastal progradation, to get an insight into the shoreline adjustments (migration rates) and to infer the stages of coastal evolution. Four sand samples were processed at IRCBNS (Babeş-Bolyai University from Cluj-Napoca, RO) and eleven at Gloucestershire University (UK), by applying multi-grain, single-aliquot regeneration (SAR) (Murray and Wintle, 2000, 2003) to fine sand sized quartz. The natural and regenerative-dose preheats ranged between 200 and 280°C for 10, with the elected thermal treatment driven by dose recovery tests. A test dose of 5 Gy followed by a preheat of 220°C for 10s was used to monitor and correct for sensitivity change. The analytical validity of equivalent dose estimation by SAR was further assessed through recycling and recuperation tests. Dose rate was determined by in situ NaI (where feasible) and ex situ Ge gamma spectrometry, accounting for the effects of grain size and present moisture content and overburden.

OSL samples were collected from beach foreshore deposits, assuring they were entirely below sand dunes. All cores and OSL sampling positions were levelled and altitudinally benchmarked relative to MSL with a Leica SR 530 RTK  DGPS ( ±3 mm vertical error and ± 1 cm lateral error).

The radiocarbon ages have been obtained using the Oxcal calibration software (Rethemeyer et al., 2012) to which the Marine 09 calibration curve (Reimer et al., 2009), a reservoir age of 440±40 yr BP and a standard deviation ΔR 75±60 (Siani et al., 2001) have been considered to obtain the final age.

Additionally, 5 transverse and longitudinal topographical profiles, summing 30 km, were surveyed over Saele and Chituc units in order to determine the beach-ridges altitude and spacing. Within the dated beach deposits the top of the foreshore-backshore facies and, if sand dunes where present, the wave-borne/wind-borne sediments contact were used as a sea-level indicator (van Heteren et al., 2000; Giosan et al., 2006). Seasonal topographical surveys of several modern Danube delta coast sectors show that subaerial beach altitudes, on which foredunes may develop, can vary between 0.7 and 1.1 m. Therefore, 0.9 m was subtracted from the absolute height of the sampled dune-free beach ridges and from each interface between dune sand and beach sand. The altitude error of the sea-level index points combines those errors from the assessment of the dune/beach interface (± 10 cm), or of the beach top (± 10 cm) with the indicative range of the beach altitudes (± 20 cm).

Fig. 4. Stratigraphy of the marshy flats and lakes (H – H3), acropolis (H4 – H7) and of the beach ridges (H8). For coring site location see Fig. 2A

Morphological evolution of the Histria region

Tracking the morphological changes, particularly of the shoreline position during and after the occupation of the city is challenging, as this is the site with the most complex coastal dynamics along the Romanian Black Sea coast. Its morphological evolution was successively controlled by: i) a sinuous shoreline which influenced the pattern of longshore sediments circulation, ii) the tectonic fragmentation of the region where the neotectonic activity generated differential subsiding and stable compartments, and iii) the progressive development of the deltaic lobes updrift of this site, which increased the coastal sedimentary budget.

Pre-Histria colonisation (4000 – 2650 BP) 

Irrespective of the differing scenarios for the Early Holocene oscillations of the Black Sea’s level – catastrophic flood proposed by Ryan et al., (1997) and Ballard et al., (2000) or gradual rise as proposed by Aksu et al., (1999) and Hiscott et al., (2002) – its waters were reconnected with the Mediterranean Sea by at least 7500 BP rising concomitantly to the modern sea-level by ca. 5000 BP (Pirazzoli, 1991; Giosan et al., 2006). From this moment on the region of Histria evolved from a moderately-indented coast (coastline sinuosity index: 1.7), composed of a succession of bays and rocky promontories which limited the longshore sediment circulation (Fig. 3A), to a smooth sandy shoreline conforming to the tendency of such systems towards equilibrium (Johnson, 1919). As the region was subject to an intense coastal progradation there is a net tendency for the sedimentary units to become younger eastwards; this was later complicated by neotectonics that caused the submergence of some of these units.

According to our chronostratigraphic results, the present Istria and Nuntaşi lakes, representing the westernmost territories, achieved the pre-colonisation configuration in two distinct phases: i) the marine gulf phase, which started slightly before the stabilization of the postglacial sea-level rise, and ii) the beach ridge plain formation phase during which the gulfs have been gradually filled up by drift-aligned beach barriers, which later (ca. 5000 BP; Fig. 1B) were juxtaposed laterally and permitted the development of a continuous beach ridge plain in front of them – Saele.

The position of the sandy coastline (Saele) at the dawn of Greek colonization was determined by correlating the mean progradation rate of 1.3 m/yr, recorded during 3330 – 2730 BP interval with the geographical coordinates defining the beach ridges dated to 850 BC ± 287 yrs and 730 BC ± 270 yrs. This algorithm projected the shoreline position by that time at a distance of 540 ± 350 m landward of the island. Presently, for more than 500 m in the lee of the island, the sandy ridges cannot be seen due to the 2 – 4 m thick archeological deposits, but outside of this area, the naturally curved beach ridges suggest that cuspate forelands, salients, and even tombolo features developed in the 1st millennium BC. Sunamura and Mizuno (1987) found that tombolos develop when the J/I ratio of the distance between sandy coast and an island (J) and the alongshore length of an island (I) is lower than 1.5, while salients initiate for ratios of 1.5 – 3.5. At Histria, the J/I ratio, computed at the time of its foundation is 1.4 (0.5 – 2.3, considering the age resolution) and suggests the existence of a wide tombolo as the most probable scenario (Fig. 3B).

Histria colonisation (2650 – 1350 BP)

The tombolo scenario is supported by the limited area of archaic deposits, which although spread over the entire Western Plateau are missing for about 300 m between the acropolis and the western residential districts. Here, it is likely that the fragile coastal features, exposed to regular coastal storms and floods, were inappropriate for continuous accommodation (Fig. 2B). The new OSL chronology together with the archeological testimonies support the hypothesis of the ancient city’s foundation on a peninsular shaped littoral complex composed of a rocky promontory fronting a beach ridge plain, a connecting sandy barrier (tombolo) and a small inlet channel draining the loess-covered bedrock hinterland to the east. The direct succession of the five OSL ages (S1 – S5) over the first 1800 m of the western Saele, between 5000 – 2730 BP, demonstrates the pre-existence of at least 2 km wide strandplain (the Saele) joining the rocky promontory, at the time of the Greeks arrival. Consequently, the colony was founded both on the green-schist outcrop, on which the city center was built, and over the adjacent sandy units where residential districts were established (Zirra, 1953; Alexandrescu, 1978). The oldest tumuli dated to 6th – 5th century BC are grouped in the central and southern parts of the hilly necropolis, while the second generation (4th – 1st c. BC) is spread from the hilly necropolis to the western Saele strandplain, over the actual submerged area between the acropolis and necropolis (Fig. 2A), presently occupied by Istria Lake (Alexandrescu, 1966,  1978).

Archeological excavations from the 1950’s found submerged ruins of a residential area in the central part of Istria Lake on a supposed loess island (Zirra, 1953). They were interpreted as having been built on a terrestrial area similar to that which currently surrounds the lake. The artifacts submerged position has been explained by their emplacement during the Phanagorian Regression when the sea level was 2 – 4 m lower (Bleahu, 1963), or by the combined action of the recent sea level rise and mainland drowning (Canarache, 1956). The stratigraphy from Istria and Sinoe lakes (including the marshy area between them) revealed marine sands below 0.8-1.5 m lake deposits of silts and clays. Textural and mineralogical analyses undertaken on these sands within the deepest cores (H1, H2) demonstrated distinct horizons for the upper and lower parts (Fig. 5A). The grain size frequency curve for the deeper layers shows very well sorted, fine and very fine sands. The upper layer sands are poorly-sorted with a wide bimodal curve indicating the presence of both fine and coarse sands. The mineralogical elements which compose the two sedimentary layers are mainly represented by: 78% quartz, 0.5% mica, 14% feldspars in the upper layers and 55% quartz, 18 % mica and  8% feldspar in the lower part (Fig. 5B); it is also noticeable the concentrations of chlorite, sericite and green schist fragments, derived from local bedrock, which are up to 10 times higher in the lower sands in comparison with the surface horizons. The two sediment fractions, with marked differences in texture and mineralogy, seem to reflect different sources, most probably Danubian sands and reworked loess in the upper layer while the bottom layer comprises higher percentages of older sands originating from local sources: e.g. proximal green schist sea cliff erosion and Istria river-born sediments.

In the Histrian plateau, H4, H5 and H6 cores (Fig. 2A) display marine sands which underly the archeological deposits. Below that, in core H4 the green-schist bedrock was intercepted, whereas core H5 presented a clayey deposit capped by a thin layer of detrital peat which could not be dated. The basal H5 layers seem to indicate an early stage of the region and a fluvial feature (e.g. floodplain lake).

Fig. 5. A. Mean grain size distribution along H2 core indicating different populations. B. Mineralogical content of the upper and lower sandy layers from H1 and H2 cores

The detailed bathymetric survey of the Razelm-Sinoe lagoon complex, carried by the National Institute of Marine Geology, mapped five submerged fossil sandy ridges representing the extensions of the most consolidated beach ridges from Lupilor barrier (Dimitriu et al., 2008). Their curved aspect, from NW-SE to W-E in a seaward direction, indicates the present-day Lupilor barrier as being the distal part of the sandy ridges composing the downdrift wing of a former deltaic lobe, corresponding to the Dunavăţ branch (Fig. 3B). The overall aspect of the Lupilor barrier, with the downdrift segments of the sandy ridges, is similar to that of the Crasnicol barrier which also supports the downdrift wing of a contemporary deltaic lobe (the Sf. Gheorghe 2, Panin, 2003). This barrier type expresses an evolutionary signature of the downdrift sides of the asymmetric wave-dominated deltaic lobes from the Danube delta, where significant river-borne muds impose a fast extension of the delta plain with the encasing of sub-parallel sandy ridges that originated as barrier islands or beach barriers (Bhattacharya and Giosan, 2003).

The existence of the Dunavăţ lobe was first proposed by Panin (1983, 2003), which called it the Coşna – Sinoe Delta, who estimated its age at 3500 – 1500 BP. The new OSL ages derived from early (L1, L2 samples) to late (I1, C1 samples) evolutionary stages of the Dunavăţ lobe sets its age to 1900 – 1300 BP, making it younger and more active than previously considered. It is likely that this rapid deltaic lobe development led to shallower waters in the offshore zone of the city and hindered boat circulation on the northward and eastward navigation routes. When tracking the southern shoreline position (the Saele) since the time of the colonization, the maintenance of the former progradation rate (1 – 1.5 m/yr) can be deduced from beach ridge succession for at least the first century. This trend was interrupted as reflected by the C-ridge position that cut the western set of the Saele beach ridges. The interruption of the beach ridges cannot be ascribed to any change in the local marine factors which could induce coastal erosion, as the coast extended south of the city functioned as a sediment trap for the longshore circulation (littoral cell end). In this case, the C-ridge discordant incision may reflect significant tectonic deformation produced sometime between 2700 – 980 BP, which corresponds with a marked hiatus, for more than 1500 years, within beach ridge succession. The abrupt disappearance of the distal end of the older ridges under the younger ones, formed in the last millennium, suggests that the former have been drowned. Seemingly, this process was decisive for the maintenance of an open coast to the east of the city, which otherwise would have been incorporated much earlier within the marine strandplains (Saele). However, the younger eastern set of the ridges is currently subaerial, without clear decreasing trends in transversal profile (Fig. 6), indicating the cessation or the slowing of the subsidence, previously recorded east of the C-ridge.

Post – Histria colonisation (1350 BP – present)

A transversal topographical survey on the northern Saele strandplain reveals the dominant position of the C-ridge, corresponding to the contact between the Old (western) and Young (eastern) Saele (Fig. 6). The cross-plain profiles within each of the Saele units (western / eastern) display a subhorizontal lateral development of the ridges conformable with their OSL age estimates, indicating an un-fractured terrestrial unit. However, the increase in the height of aeolian deposition on the beach ridges from the extreme west Saele could indicate a slow subsidence (≤ 0.5 mm/yr) of this peripheral area. The C-ridge rises up to 2-3 m above the mean level of the plain in its northern part and gradually decreases in altitude and width towards the south where it diminishes, but remains within the general architecture of the Saele plain. The C-ridge is made up primarily of juvenile shells and fragments of shells (e.g. Mytilus sp., Cardium sp.) disposed in layers of different degrees of compaction. The C-ridge seems to follow a fault alignment, eastward of which the former sandy ridges were drowned, and the new set of Young Saele developed between 980 –720 BP owing to Danubian sediments reaching the area by longshore drift. Based on volumetric reconstruction, the location of OSL age estimates and the elevation of sandy ridges and closure depth (-10 m), the longshore sediment transport specific to non-restricted sediment conditions (post 980 BP) on the eastern Saele coast was computed as 1.0 – 1.2 x 10⁶ m3/yr. The Young Saele was built much faster than the Old Saele, concomitantly with a clockwise rotation of coastline progradation direction from W-E to NW-SE. This is likely related to the gradual infilling of this downdrift bay, representing the end of the littoral cell. The progradation rate of the Young Saele shoreline, reflected by the OSL ages, was ten times higher than for the western set of the ridges (Old Saele), measuring 10-15 m/yr with greater values recorded downdrift, in the south. We interpret this high progradation rate as reflecting a sudden increase of sediment input into a low wave-energy environment. The newly obtained ages indicate that abundant sediment influx reached the Histria region around 1400-1300 BP (L3, C1 samples). Taking into consideration the beach ridge orientation pattern, we assume this scenario occurred when the prominent Dunavăţ deltaic lobe situated updrift started to erode, supplying sediments to the longshore currents which moved them downdrift, to a sheltered location.

Some 2-5 km east of the Saele, at approximately the same latitude, lies the Chituc beach ridge plain. The present study proposes an absolute chronology that indicates inception of this ridge plain 1.3 ka BP as a drift-aligned barrier orientated ca. E-W (Fig. 3B). Further downdrift, the beach ridges align at an ever decreasing angle, becoming NE-SW orientated achieving a plan-view equilibrium when the shoreline becomes more perpendicular to the predominant swell (Hsu et al., 2010). The Chituc strandplain is considered the youngest marine feature, due to its extreme seaward position, with an inception age estimated sometime from the 2nd century BC (Pippidi, 1983) to the 3rd century AD (Bleahu, 1963) or the 6thcentury AD (Coteţ, 1966; Panin, 1983, 2003). In this case, the 1300 BP age of its oldest ridge (northern sector) is in agreement with the youngest estimates previously suggested. Magnetometric investigations undertaken over bottom sediments of the Sinoe Lake suggest the continuity of the eastern Saele beach ridges with those from Chituc barrier (Mihăilescu et al., 1983; Dimitriu et al., 2004). Although most authors noticed the similarity between the eastern Saele and Chituc beach ridges, they still consider them to be of different age. They attribute the present Sinoe lagoon configuration to the formation of the Chituc barrier, as a result of a marine transgressive phase (Bleahu, 1963; Coteţ, 1966; Panin, 1983, 2003). Recently, Giosan et al. (2006) proposed two possible scenarios: either they formed simultaneously, probably on each side of an inlet, or the Chituc formed after the Saele. The present OSL age estimates (S5, S6, S7, S8, C2, C3), as well as the magnetometric signal of the lacustrine sediments and sandy ridge geometry, reveal the coevolution of the eastern Saele and Chituc which formed a continuous beach ridge plain for an interval spanning 980 – 720 BP at least. The ages recorded on eastern Saele are the same as those from central Chituc, while the last interval of formation (320 BP – present) corresponds to the southern Chituc formation.

Fig. 6. Topography of the Chituc and Saele beach ridge plains. The y-axis (elevation) is identical for both profiles to comparatively assess the depressed profile of the northern Chituc and the subhorizontal profile of Saele, interrupted by C-ridge morphology.

The massive Saele-Chituc strandplain divided in two when the Sinoe Lake appeared. As the youngest beach ridges currently submerged by the Sinoe Lake waters are 780 – 720 BP, we have to assume a maximum age of lake inception in its southern part of ca. 700 BP. Sinoe banks are rectilinear, cut NNE-SSW, and parallel to the others lake banks from the region (Istria and Nuntaşi) which are delineated in the same direction, revealing they all have a tectonically-induced origin (Fig. 1B). Their common geometry indicates the presence of the NNE-SSW active faults, perpendicular to the major WNW-ESE tectonic faults; collectively they appear consistent with the regionally subsiding blocks.

Over the last 1300 years the evolution of the Dunavăţ lobe was marked by strong erosion along its updrift and central parts after the Dunavăţ branch abandonment (Panin, 2003). Also, the locally active subsidence forced the submergence of the downdrift part of the lobe, enabling the inception of the Sinoe northern basin and ultimately destroying most of the Dunavăţ lobe whose remnant is the Lupilor barrier. Estimates of the timing of the Nuntaşi and Istria lakes formation (mean depth of -1 m), whose surficial bottom sediments have been wave-reworked during regional storms, were made both by ¹⁴C dating of an articulated shell (710 ± 90 BP) placed within lacustrine muds, close to the marine-lacustrine interface, and from the extrapolation of the sedimentation rates.

Analysis of the lacustrine shorelines extracted from topographical maps and satellite images covering the last 120 years (1884, 1961, 1979 and 2008) points to a moderate erosion acting on 42 % and 35 % of the total length of Nuntaşi and Istria rocky shores, respectively, mainly cut into loess. The mean retreat rates (1 m/yr Nuntaşi, 0.8 m/yr Istria) multiplied by the length and height of the erosional sectors gives an annual overall respective wave-displaced sediment volume of 7200 m³ and 1700 m³, which relative to their present surface corresponds to a sedimentation rate of ca. 1.5 mm/yr, respectively 1.2 mm/yr.

In order to estimate the maximum ages of lake formation, the mean width of bottom sediments was reported to the modern sedimentation rates (assumed to be equal or smaller than the previous ones) producing age estimates of ca. 1200 BP and 900 BP for the Nuntaşi and Istria lakes, respectively.

 

Late Holocene sea level curve for Histria 

The discovery of numerous submerged archeological elements (tumuli and walls) within the Istria and Sinoe Lakes, led to the common interpretation that they were built during a regressive phase contemporaneous with the date of first colonization. Moreover, the submergence of some structures shown to have been built on subaerial marine sands, as well as Sinoe Lake formation, was related to a marine transgression phase locally named the “Histrian transgression” (Bleahu, 1963,Vespremeanu, 2005).  Consequently, when considering the landscape changes of Histria’s low-lying hinterland, most studies invoke eustatic sea level changes (Coteţ, 1962, 1966; Bleahu, 1963; Panin, 1983, 2003; Vespremeanu, 2005) while just a few appreciate neotetectonics or lake bank morphodynamics among the main controlling factors (Canarache, 1956; Giosan et al., 2006). The eustatic theory of the Histria landscape evolution was linked by the Phanagorian regression concept of Fedorov (1957) which first explained the submerged position of the Phanagoria lower city as being built during a Black Sea low-stand. Since then, most of the studies on the Mid and Late Holocene evolution of the Black Sea’s level commonly presumed a high-stand of 2–5 m at the beginning of the 2nd millennium BC, followed by a major drop in the 1st millennium BC of 5–10 m relative to the present, corresponding to the Phanagorian regression (Fedorov, 1957, 1977; Shilik, 1975; Chepalyga, 1984; Vespremeanu, 2005; Balabanov, 2009).

In order to derive a valid sea level curve for the region of Histria all the dates from the relatively stable beach ridge plains, covering the last four millennia were used (Fig. 7A). Surprisingly, a stable relative sea-level characterized the 1000 BC – 1000 AD, including the interval of Histria’s flourishing occupation, and suggests the absence of large-amplitude eustatic oscillations (Fig. 7B). During this period, the mean sea-level position was about -0.7 m, within the range of -0.2 m and -1.2 m of the current Black Sea’s level, followed by a slow, gradual rise over the last millennium. The only available sea-level indicators older than 3000 BP found within the western part of the Old Saele which seems to be affected by slow subsidence. Although the subsidence could explain the lowering of the oldest sea-level indicator (5000 BP) it is difficult to assess if the sea level rose or was relatively stable.  The different positions occupied on the sea-level curve by the contemporaneous sea-level indicators show active neotectonics. This is the case of the lowered position of the beach ridges and dune/beach interface from the Chituc northern half, compared to the eastern Saele, which points to a moderate subsidence rate (~ 0.5 mm/yr) that affects this unit, promoting lake inception and strong coastal erosion with 4-8 m/yr in shoreline retreating rate (Vespremeanu-Stroe et al., 2007). Further, the topographical surveys carried across Chituc highlight the northwardly depressed profile of this unit which also suggests its subsidence.

Fig. 7. A. Black Sea-level curve derived for the southern Danube Delta based on the age and altitude of the aeolian / marine deposits interface within the beach ridge plains. B. Different Late-Holocene sea level curves produced for the Black-Sea (adapted after Fouache et al., 2011). Note the lack of correspondence between the current research and the previous sea-level curves postulating the presence of significant highstands (Old Black Sea) and lowstands (Phanagorian).

 

References

Adamiec, G. and Aitken, M.J., 1998. Dose-rate conversion factors: new data. Ancient TL 16, 37-50.

Alexandrescu, P., 1966. Necropola tumulară. Săpături 1955-1961, in: Histria 2, Bucharest.

Alexandrescu, P., 1970. Peisajul histrian în Antichitate. Pontica 3, 77-85.

Alexandrescu, P., 1978. Notes de topographie histrienne, Dacia, 22, 331-342.

Alexandrescu, P. and Schuller, W.  (Eds.), 1990. Histria. Eine Griechenstadt an der rumänischen Schwarzmeerküste, Konstanz

Antipa, G., 1914. Câteva probleme știinţifice şi economice privitoare la Delta Dunării, An. Acad. Rom. Mem. Sect. St. 2 (36).

Atanasiu, I., 1940. Privire generală asupra geologiei Dobrogei, Lucr. Soc. Geogr. Dimitre Cantemir, Iaşi, 89.

Avram, A.,  Hind, J., Tsetskhladze, G., 2004. The Black Sea Area, in: Hansen, M. H. , Nielsen, T. H. An Inventory of Archaic and Classical Poleis, Oxford, 924-973.

Aksu, A.E., Hiscott, R.N., Yasar, D., Isler, F.I., and Marsh, S., 2002, Seismic stratigraphy of late Quaternary deposits from the southwestern Black Sea shelf: Evidence for non catastrophic variations in sea-level during the last 10,000 years: Marine Geology 190, 61–94

Balabanov, I.P., 2009. Paleogeographic Background to Modern Natural Condition of the Caucasus Littoral Holocene Terraces and their long-term development forecast.Dalnauka, Moscow, 352

Ballard, R.D., Coleman, D.F., Rosenberg, G.D., 2000. Further evidence of abrupt Holocene drowning of the Black Sea shelf. Marine Geology 170, 253-261

Bhattacharya, J.P., and Giosan, L., 2003, Wave-influenced deltas: Geomorphologic implications for facies reconstruction: Sedimentology 50, 187–210

Bleahu, M.,1962. Observaţii asupra evoluţiei zonei Histria în ultimele trei milenii, Probleme de Geografie 9, 45-56

Breier, A., 1976. Lacurile de pe litoralul românesc al Mării Negre. Studiu hidrogeografic. Editura Academiei Republicii Socialiste România, Bucharest

Brückner, H., Kelterbaum, D., Marunchak, O., Porotov, A., Vött, A., 2010. The Holocene sea level history since 7500 BP- lessons from the eastern Mediterranean, the Black and Azov seas. Quaternary International 225 (2), 160-179.

Canarache, V., 1956, Observaţii noi cu privire la topografia Histriei 7, Studii și Cercetări de istorie veche (2-3), 289-315.

Chepalyga, A., 1984. Inlands sea basins. In: Velichko,A.A., Wright, H. E., Barnosky,C.W. (Eds.), Late Quaternary Environments of the Soviet Union. University of Minnesota Press, Minneapolis, 237-240.

Clemmensen, L.B. and Nielsen, L., 2010. Internal architecture of a raised beach ridge system (Anholt, Denmerk) resolved by ground-penetrating radar investigations. Sedimentary Geology 223, 281-290.

Condurachi, E., 1954, Histria, Monografie arheologică 1, Bucharest.

Coteţ, P.V., 1960. Şantierul Histria. Materiale şi cercetări arheologice, 8.

Coteţ, P.V., 1962. Câteva date asupra evoluţiei paleogeografice cuaternare a regiunii Istria. Materiale şi Cercetări Arheologice 8, 424-431.

Coteţ, P. V., 1966, Ţărmul Mării Negre şi evoluţia lui în timpuri istorice (cu privire specială asupra regiunii Histria), Histria 2, 337-352.

Dan, S., Stive, M.J., Walstra, D.J.R., Panin, N., 2009. Wave climate, coastal sediment budget and shoreline changes for the Danube Delta. Marine Geology 262 (1-4), 39-49.

Dimitriu, R.G., Oaie, Gh., Szobotka, Şt., Gomoiu, M.T., Sava, C.S., Secrieru, D., Fulga, C., Sosnovschi, E., Anghel, S., 2004. Cartografierea geologică şi geofizică a sectorului Sinoe. CERES, Scientific Report.

Dimitriu, R.D., Oaie, Gh., Gomoiu, M.J., Begun, T., Szobotka, St., Radan, S.C., Fulga, C., 2008. O caracterizare interdisciplinară a stării geoecologice a Complexului Lagunar Razelm-Sinoe. Geo-Eco-Marina 14 (1), 69-74.

Dimitriu, R, G., 2012. Geodynamic and hydro-geological constraints regarding the extension of the prospective archaeo-cultural area within the northern Romanian coastal zone. Quaternary International, 261, 32-42.

Duller, G.A.T., 2003. Distinguishing quartz and feldspar in single grain luminescence measurements. Radiation Measurements 37, 161-165.

Fedorov, P.V. and Skira, I.A., 1957. Epoca actuală în istoria geologică a Mării Negre. Analele Rom. Sov. Seria Geol.Geogr., 3.

Fedorov, P.V.,1977. Late Quaternary history of the Back Sea and evolution of the southern seas of Europe. In: Kaplin, P.A., Shcherbakov, F. A (Eds.) Pleistocene paleogeography and sediments of the Southern seas of the USSR, Nauka, Moscow, 25-32 (in Russian).

Fornasier, J. , Böttger, B.,(Eds.). 2002. Das bosporanische Reich der Nordosten in der Antike, Mainz.

Fouche, E., Kelterbaum, D., Brückner, H., Lericolais, G, Porotov, A., Dikarev, V., 2011. The Late-Holocene evolution of the Black Sea – a ritical review on the so-called Phanagorian regression, Quaternary International, doi:10.1016/j.quaint.2011.04.008

Giosan, L., Donnelly, J.P., Constantinescu, Şt., Filip, F., Ovejanu, I., Vespremeanu – Stroe, A., Vespremeanu, E., Duller, G.A.T., 2006. Young Danube delta documents stable Black Sea level since the middle Holocene: Morphodynamic, paleogeographic, and archaeological implications. Geology.v.34, 9, 757-760.

Goy, J.Z., Zazo, C., Dabrio, C.J., 2003. A beach ridge complex reflecting periodical sea-level and climate variability during the Holocene (Gulf of Almeria, western Mediterranean). Geomorphology 50, 251-268.

Hiscott, R.N., Aksu, A.E., Yaşar, D., Mudie, P.J., Kostylev, V.E., MacDonald, J.C., Isler, F.I., Lord, A.R., 2002. Deltas south of the Bosphorus Strait record persistent Black Sea outflow to the Marmara Sea since10ka. Marine Geology 190, 95-118.

Hsu, J.R.C., Yu, M.-J., Lee, F. –C. & Benedet, L. 2010. Static bay beach concept for scientists and engineers: A review, Coastal Engineering 57, 76-91.

Ianovici, V., Giuşcă, D., Mirăuţă, O., 1968. Harta geologică Tulcea (L-35-XXIX). Institutul Geologic, Bucureşti, 32 p.

Jonhson, J.W., 1919. Shore processes and shoeline development, Wiley, New York. [facsimile edition: Hafner, New York (1965).]

Lambrino, M. F, 1938, Les vases archaique d’Histria, Bucarest, pp. 375

Masselink, G. and Hughes, M.G., 2003. Introduction to coastal processes and geomorphology. Hodder education.

Mejdahl, V., 1979. Thermoluminescence dating: beta-dose attenuation in quartz grains. Archaeometry 21, 61-72.

Mihăilescu, N., Rădan, S., Artin, L., Rădan, S., Rădan, M., Vanghelie, I., 1983. Modern sedimentation in the Razelm-Sinoe Lacustrine Complex, Anuarul Instit. de Geologie si Geofizica, LXII, 297-304.

Moisil, C., 1909. Antichităţi creştine din Istros. Buletinul Comisiunii Monumentelor Istorice 2, 165-170.

Murray, A.S., Wintle, A.G., 2000. Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol.Radiation Measurements 32, 57–73.

Murray, A.S., Wintle, A.G., 2003. The single aliquot regenerative dose protocol: potential for improvements in reliability. Radiat. Meas. 37, 377–381.

Oppermann, M. 2004. Die westpontischen Poleis und ihr indigenes Umfeld in vorrömischer Zeit, Langenweißbach

Panin, N., 1989. Danube delta. Genesis, evolution and sedimentology. Revue Roum. de Geology, Geophiz, Geogr. 33, 25-36.

Panin, N., Panin, S, Herz, N., Noakes, J.E., 1983. Radiocarbon dating of Danube Delta deposits. Quaternary Research 19, 249-255.

Panin, N., 1983. Black Sea coast line changes in the last 10,000 years: a new attempt at identifying the Danubes mouths as described by the Ancients. Dacia 27, 175-184.

Panin, N. 2003. The Danube Delta geomorphology and Holocene evolution: a syntheis. Geomorphologie: relief, processus, environment 4, 247-262.

Pârvan, V., 1915, Rumanien (Archaologische Funde im Jahre 1914), Archaologischer Anzeiger 4, 253- 270.

Pârvan,V., 1916, Campania a II-a de săpături la Histria, Raport asupra activităţii Muzeului Naţional de Antichităţi în cursul anului 1915, Bucureşti, 18-20.

Pippidi, D.M., 1953. Histria şi Callatis în sec. III-II î.e.n, Studii şi Cercetări de Istorie Veche 4 (3-4), 32-45.

Pippidi, D.M., 1983. Inscripţiile din Scythia Minor greceşti şi latine. 1, Histria şi împrejurimile. Editura Academiei Republicii Socialiste România, Bucureşti, 14-37

Pirazzoli, P., 1991. World Atlas of Holocene Sea-Level Changes. Elsevier Oceanography series.

Polonic, G., Zugravescu, D., Horomnea, M., Dragomir, V., 1999. Crustal Vertical Recent

Movements and the Geodynamic Compartments of Romanian Territory, Istanbul,

Turkey. In: 2nd Balkan Geophysical Congress, Book of Abstracts, p. 300-301.

Prescott, J.R. and Hutton, J.T., 1994. Cosmic ray contributions to dose rates for luminescence and ESR dating: large depths and long-term time variations. Radiation Measurements 23, 497-500.

Reimer, Paula J.; Baillie, Mike G. L.; Bard, Edouard; Bayliss, Alex; Beck, J. Warren; Blackwell, Paul G.; Ramsey, C. Bronk; Buck, Caitlin E.; Burr, George S.; Edwards, R. Lawrence; Friedrich, Michael; Grootes, P. M.; Guilderson, Thomas P.; Hajdas, Irka; Heaton, T. J.; Hogg, Alan G.; Hughen, Konrad A.; Kaiser, K. F.; Kromer, Bernd; McCormac, F. G.; Manning, S. W.; Reimer, Ron W.; Richards, D. A.; Southon, John R.; Talamo, Sahra; Turney, C. S. M.; van der Plicht, Johannes; Weyhenmeyer, Constanze E.,2009, IntCal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP, Radiocarbon, 51(4), 1111-1150.

Rethemeyer, J., Fülöp, R.-H., Höfle, S., Wacker, L., Heinze, S., Hajdas, I., Patt, U., König, S., Stapper, B., Dewald, A., 2012, Status report on sample preparation facilities for 14C analysis at the new CologneAMS center, Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atom, http://dx.doi.org/10.1016/j.nimb.2012.02.012

Ryan, W.B.F., Pitman,W.C., Major, C.O., Shimkus, K., Moskalenko,V., Jones, G.A., Dimitrov, P., Görür, N., Sakinç, M., Seyir, H.I., Yüce, H., 1997. An abrupt drowning of the Black Sea shelf. Marine Geology, 138, 119-126.

Săndulescu, M., 1984. Geotectonica României. Editura Tehnică, Bucureşti, 336 p.

Shilik, K.K., 1997, Oscillations of the Black Sea and ancient landscapes: Colloquia Pontica, 3, 115–130.

Siani, G., Paterne, M., Arnold, M., Bard, E., Métivier B., Tisnerat N., Bassinot, F., 2000, Radiocarbon reservoir ages in the Mediterranean Sea and Black Sea: Radiocarbon, 42, p. 271-280.

Solovyov, S. L. 1999. Ancient Berezan. The Architecture, History and Culture of the First Greek Colony in the Northern Black Sea, Leiden-Boston-Köln, 10-13

Sunamura, T and Mizuno, O., 1987. A study on depositional shoreline forms behind an island. Annual Report institute Geoscieces, University of Tsukuba, 13, 63-78.

Ştefan, A. S., 1987, Évolution de la côte dans la zone des bouches du danube durant l’antiquite in Déplacements des Lignes de Rivage en Méditerranée d’aprés les Données de l’Archéologie, Editions du CNRS, Colloques Internationaux, Paris, 191-209.

van Heteren, S., Huntley, D.J., van de Plassche, O., and Lubberts, R.K., 2000, Optical dating of dune sand for the study of sea-level change: Geology 28, 411–414.

Theodorescu, D., 1970. Notes Histriennes, Revue Archeologique 1, 29-48.

Zirra, V and Alexandrescu, P., 1957. Şantierul arheologic Histria: sectorul necropolei tumulare. Materiale şi Cercetări Arheologice 4, 22-31.

Vespemeanu, E., 2005. Geografia Mării Negre. Editura Universitară. Bucureşti, 255.

Vespremeanu-Stroe, A., Constantinescu, Ş., Tătui, F. and Giosan, L. 2007. Multi-decadal Evolution and North Atlantic Oscillation Influences on the Dynamics of the Danube Delta Shoreline, Journal of Coastal Research 50, 175-162.

Vespremeanu-Stroe, A. and Tătui, F, . 2011. North-Atlantic osscilations signature on coastal dynamics and climate variability of the Romanian Black Sea coast, Carpathian Journal of Earth and Environmental Sciences 6 (1), 135-144.

Zimmerman, D. W., 1971. Thermoluminescent dating using fine grains from pottery. Archaeometry 13, 29-52.