2012 LakeChokrak Kelterbaum Et Al. Geoarchaeology
2012 LakeChokrak Kelterbaum Et Al. Geoarchaeology
Correspondence                                  This project has reconstructed the palaeogeographic and environmental evo-
*Corresponding author;                          lution of Lake Chokrak on the Kerch Peninsula, Ukraine, during the mid- and
E-mail: [email protected];
                                                late-Holocene. This record has been evaluated in association with a regional
[email protected]
                                                archaeological data set to explore human–environment interactions over this
Received                                        period. The results show major changes in the palaeogeographic setting of Lake
6 July 2011                                     Chokrak since the 3rd millennium B.C., when the postglacial marine trans-
Accepted                                        gression had started to fill the study area. Microfaunal analyses reveal the long
13 January 2012                                 persistence of an open marine embayment that only became separated from
                                                the Sea of Azov when a sand barrier developed during the late 2nd millen-
Scientific editing by Jamie Woodward
                                                nium B.C. When colonizing the Black Sea region after the 8th century B.C.,
Published online in Wiley Online Library        the Greek settlers erected a fortification with a small settlement on a promon-
(wileyonlinelibrary.com).                       tory that was by then a peninsula-like headland extending into the lake. The
                                                colonists abandoned their settlement at the end of the 1st millennium B.C.
doi 10.1002/gea.21408
                                                when the depth of the surrounding lake decreased from 1.5 m to less than
                                                1 m. Today, Lake Chokrak dries up completely during summer. A detailed
                                                relative sea level (RSL) curve for the northern coast of Kerch has been estab-
                                                lished. Sea level reached its highest position at the present day. Since the 3rd
                                                millennium B.C., sea level continuously rose, without any of the previously
                                                postulated regression/transgression cycles. The RSL curve indicates differential
                                                subsidence rates within short distances in relatively stable areas, exceeding 40
                                                cm per 1000 years.  C 2012 Wiley Periodicals, Inc.
Figure 1 Coring sites and simplified geomorphological map of Lake Chokrak and its environs. (a) Overview of the area; (b) study site on the sand barrier;
(c) study site on the Chokrak Peninsula.
northern part of the Kerch Peninsula in Crimea, Ukraine.                             olithic and Mesolithic epochs. These small settlements are
A small peninsula stretches into the lake, which today is                            generally situated inland in the central area of the Kerch
totally disconnected from the open Azov Sea. At the tip                              Peninsula. However, further prehistoric sites may also be
of “Chokrak Peninsula,” an ancient Greek settlement was                              located on the Black Sea shelf, which was flooded by
founded during the 1st millennium B.C.                                               the postglacial sea level rise. It is noteworthy that dur-
   The results presented here are based on sedimento-                                ing the Bronze Age (4th–2nd millennium B.C.), hunt-
logical investigations of vibracorings carried out on the                            ing and gathering were supplemented by fishing: exca-
sand barrier separating Lake Chokrak from the Azov Sea,                              vations show the occurrence of sinkers, fish bones, and
and in the adjacent lagoon. This research was carried                                shells in the cultural layers of the settlements, which in-
out in cooperation with the archaeological research of                               dicate the important role of fishing and therefore the sea
the Demetra Foundation in Kerch and Professor Maslen-                                as a driving cultural factor (Danovskiy & Maslennikov,
nikov’s excavations in the northern part of the Kerch                                2006).
Peninsula.                                                                              The best historical record exists for the period of the
   This study has three principal aims: (1) to reconstruct                           Greek colonization, which started in the 6th century
the palaeogeography of Lake Chokrak for different time                               B.C. Originating from the center of classical Greece, the
periods; (2) to establish a relative sea level (RSL) curve                           Hellenic people settled along the coasts of the Mediter-
for the northern coast of Kerch and therefore for the Sea                            ranean, the Marmara Sea, and the Black Sea. One ma-
of Azov; and (3) to establish the influence of any envi-                             jor center of this period of colonization and conquest
ronmental changes on the ancient Greek colonization of                               was the area of the present Strait of Kerch, formerly
the Kerch Peninsula.                                                                 known as the Kimmerian Bosporus, the marine passage
                                                                                     separating the peninsulas of Taman (Russia) and Kerch
                                                                                     (Ukraine). Renowned settlements on Kerch Peninsula
                                                                                     are Pantikapaion, Myrmekion, Nymphaion, Tyritake, and
HISTORICAL BACKGROUND                                                                Kytaia. Most of them were easy to defend, had har-
Numerous ancient and prehistoric settlements on the                                  bor facilities, and were places of high strategic value
Kerch Peninsula show that the area has been occupied                                 for the control of sea and land trading routes. Pantika-
and strongly influenced by humans for millennia. The                                 paion was the most prominent city. Together with some
earliest human settlements date back to the late Palae-                              smaller settlements on Taman Peninsula, it formed the
Figure 2 View looking to the northeast on the Lake Chokrak sand barrier. Important geomorphological features and the coring sites are highlighted.
Areas of relative uplift (+) and areas of relative subsidence (–) are shown. Photograph taken by DK in spring 2009.
Bosporan Kingdom (Gajdukevic, 1971; Fornasier &                              were deposited. In many places, mud volcanoes are sit-
Böttger, 2002).                                                             uated in the depocenters of the Maykop series (Herbin
   This study focuses on the northern Kerch Penin-                           et al., 2008). Within synclinal structures, the bedrock is
sula, which was most likely the granary of Pantika-                          covered by colluvial material and loess with a thickness
paion (Saprykin, 2006). There, several impressive ar-                        of up to 10 m (Zubakov, 1988; Dodonov et al., 2000).
chaeological Greek sites are situated along the coastal                      The coastline around Lake Chokrak is characterized by
zone on rocky capes. In the south of Lake Chokrak, on                        steep cliffs and small embayments. The lake forms the
the Chokrak Peninsula, a strongly fortified administra-                      only large depression reaching inland. Its genesis is due
tive center was erected with a ditch and solid walls that                    to a large N–S striking fault line, which runs perpendicu-
served as a capital for governing the region. This place                     lar to the generally W–E striking Jurassic and Cretaceous
was inhabited from the late 5th to the 1st century B.C.                      ridges of the peninsula.
(Maslennikov, 1998). It was abandoned c. 500 years be-                          Figure 1b and Figure 2 show the sand barrier and the
fore the Bosporan Kingdom disappeared. The reason for                        core drilling sites. The surrounding highlands are mainly
this abandonment many centuries before the Kingdom’s                         formed in limestones and marls. The latter are strongly
doom was one of our research foci.                                           eroded—possibly as a result of the long and intensive an-
                                                                             thropogenic impact on the region’s ecosystems. The sand
GEOLOGICAL AND TOPOGRAPHICAL                                                 barrier, built up of sediments from adjacent cliffs, is visible
SETTING                                                                      in the center of Figure 2. According to Nikonov (1995a,
                                                                             1995b), the northern Kerch Peninsula is part of a neo-
In the northern Black Sea region, the Crimean Penin-                         tectonically active region, which is strongly influenced by
sula separates the Sea of Azov from the Black Sea. To-                       low magnitude earthquakes. This author also describes
day, both seas are connected only by a small opening,                        strong subsidence in the north of the Kerch Peninsula
the so-called Strait of Kerch (formerly the Kimmerian                        since this area is part of the Indolo-Kuban trough. How-
Bosporus), which is also the border between Asia and Eu-                     ever, in contrast to the subsidence of the Sea of Azov,
rope. The west coast of the marine passage belongs to the                    the Kerch Peninsula can be described as a quasi-stable or
Kerch Peninsula, which is the easternmost part of Crimea                     slowly rising area as shown by late Holocene marine ter-
(Figure 1).                                                                  races vertically uplifted up to 3 m above the present sea
   The Kerch Peninsula is mainly formed of carbonate                         level (asl) (Danovskiy et al., 2009). In Figure 2, the dif-
rocks of Jurassic and, in some parts, of Cretaceous age.                     ferential tectonics is indicated along presumed fault lines.
Both geological units are exposed due to the Alpine
orogeny that produced the Crimean Mountains and the
Caucasus. The area is situated near the Indolo-Kuban                         MATERIAL AND METHODS
trough, which is part of the Northern Caucasus mo-                           During fieldwork, several cores were collected using the
lasse basin (e.g., Ershov et al., 1999). During the Middle                   percussion coring device Cobra 248 (Atlas Copco Hold-
Tertiary, sandstones of the so-called Maykop Formation                       ing GmbH, Essen, Germany) with augerheads of 6, 5, and
Table I AMS radiocarbon dating results from Lake Chokrak. All ages are calibrated with the calibration software Calib. 6.0.1 (http://intcal.qub.ac.uk/calib/).
A marine reservoir effect of 408 years was applied (Hughen et al., 2004) with a delta R of 5 ± 35 years (Jones & Gagnon, 1994). The Libby half life of 5568
was used. M = articulated bivalve (C. glaucum, only), in situ; PR = plant remains. AMS 14 C dating was carried out by the Center for Applied Isotope Studies
(CAIS) of the University of Georgia at Athens, USA (Lab code: UGAMS).
                                              Conventional 14 C     13 C     Depth Upper limit Lower limit Palaeo mean sea Calibrated ages 2-sigma
Sample code Lab code Dated material                age              (ppm)     (m bsl) (cm)         (cm)       level (m bsl)        (B.C./A.D.)
3.6 cm in diameter. A maximum depth of 9 m below the                                  15 kV, Take-off Angle = 30◦ circ). Although quantitative
modern land surface (b.s.) was reached. The coring strat-                             EDX data may show high errors, especially concerning
egy was based on transects through nearshore sedimen-                                 light atoms, a general trend for the abundance of heavier
tary archives, chosen by field survey, supported by satel-                            atoms can be followed.
lite image interpretation and former research carried out                                For cores CHO 2 and CHO 4 facies determination
by Nevessky (1967). The position and elevation of each                                was done using microfossil occurrence as an index for
coring site was measured by DGPS with a vertical resolu-                              palaeoenvironments and sedimentation conditions (for
tion of less than 2 cm (Leica Geosystems Co.; Leica 530                               further details see Figures 2–4). The studied sediment vol-
SR). On site, the cores were studied for color (MUNSELL                               umes differ from 5 to 10 cm3 depending on the amount
Soil Color Charts), grain size, rounding and texture, as                              of available material. Each sample was washed in distilled
well as macrofaunal and macrofloral remains.                                          water and only the 50–200 μm fraction was analyzed un-
   Multiproxy geochemical analyses were carried out for                               der a binocular microscope. Microfossils were determined
selected samples in the laboratories of the Faculty of Ge-                            after Caraion (1967) and Murray (1986, 2006). The ana-
ography at the Philipps-Universität Marburg (Germany).                               lytical proxies represent the basis for a distinct facies dif-
Samples were air-dried and pestled by hand. CaCO3 was                                 ferentiation. Combined with the grid of cores, the verti-
gas-volumetrically measured (Scheibler method). AMS                                   cal and horizontal variations of the different facies (such
radiocarbon dating was used to develop a chronologi-                                  as marine, littoral, lagoonal, limnic, or fluvial) were used
cal framework and was applied mainly on articulated                                   for the reconstruction of the palaeogeographic evolution
bivalves at sedimentary boundaries. All ages were cali-                               of the research area.
brated with Calib 6.0.1 [Stuiver and Reimer (1993) and
Reimer et al. (2009); see Table I], taking also into account
a marine reservoir effect of 408 years (Hughen et al.,                                RESULTS
2004). It is important to note however that the real palae-
                                                                                      Results from the Chokrak Sand Barrier
oreservoir effect is still unknown and may have varied
widely in different marine and brackish environments,                                 Two cores were collected from the top of the Chokrak
such as lagoons, coastal swamps, or littoral zones. Ad-                               sand barrier, with core CHO 4 in its center and core CHO
ditionally, especially in semienclosed basins such as the                             5 in its western part (Figure 1b). Both sites are located
Black and Azov Seas, the discharge of rivers and their in-                            close to washover structures, which reach from the bar-
put of 12 C may have to be taken into account.                                        rier into the lake (Figure 1b & 2). In the following sec-
   Energy dispersive X-ray spectroscopy (EDX) measure-                                tions, the analytical results of core CHO 4 will be pre-
ments on crushed carbon coated samples were analyzed                                  sented in detail (Figure 5). The drilling depth reached
in the laboratories of the University of Marburg using                                down to 8 m below the present surface (b.s.), that is,
a raster electron microscope equipped with an energy                                  7.67 m below the present sea level (bsl). The core shows
dispersive X-ray spectrometer (EDX, PROZA correction,                                 the sand barrier’s sedimentary architecture and reflects its
Figure 3 View over Lake Chokrak toward the northwest. The archaeological site on the Chokrak Peninsula is indicated in the center close to the core
sites CHO 1 and 2. Photograph taken by DK in spring 2009.
evolution. Based on geochemical and microfaunal find-                         4/13 MS, 5.37 m bsl: 2035–1764 cal B.C.; CHO 4/9 MS,
ings, six different sedimentary units were distinguished.                     4.02 m bsl: 1195–916 cal B.C.; CHO 4/8 MS, 3.27 m bsl:
   Unit F: The lowermost sediments (8.00–7.40 m b.s.)                         791–547 cal B.C.). Based on the microfaunal assemblage,
consist of very homogeneous silty clay of a reddish-gray                      unit D can be divided into three subunits (unit D.1: 6.60–
color due to secondary oxidation processes. The unit is                       5.60 m b.s.; unit D.2: 5.60–4.00 m b.s.; unit D.3: 4.00–
void of micro- and macrofossils. The calcium carbonate                        2.75 m b.s.). The density of the microfauna decreases
content is low as a result of terrestrial conditions during                   from bottom to top, while the amount of reworked mi-
sedimentation.                                                                crofauna increases in the upper two subunits, reflecting
   Unit E: Then follows a layer of gray, silty fine sand                      changing environmental and depositional conditions.
(7.40–6.60 m b.s.). Its lowermost part is reworked ma-                           Unit C: At 2.75 m b.s. (2.43 m bsl), the sediment
terial of unit 1. Several fragments, single valves, and ar-                   changes entirely. Unit C (2.75–1.00 m b.s., 2.43–0.67
ticulated specimens of Cerastoderma glaucum occur. One                        m bsl) contains several coarse, partly graded layers with
specimen of the latter was dated to 3472–3162 cal B.C.                        plant remains and abundant broken macrofaunal re-
(CHO 4/18 MS; 7.27 m bsl). It has to be noted that this                       mains, the latter causing an increased CaCO3 content of
sample was located in the pretransgressive sediment, rep-                     up to 60%. Electrical conductivity decreases due to the
resenting the possible penetration depth of this species.                     coarser grain size. The microfaunal composition is re-
Another articulated specimen from 7.02 m bsl has an age                       stricted to a few and mostly reworked foraminifera spec-
of 2876–2631 cal B.C. (CHO 4/17 MS). The microfau-                            imens of A. tepida, A. parkinsoniana and H. germanica, and
nal analysis from this unit indicates a medium abundance                      the ostracod C. torosa. Microfaunal remains could only be
of foraminifera, mostly Ammonia tepida and A. parkinsoni-                     detected in the lowermost section of this unit.
ana. A few other species such as Cribroelphidium gunteri,                        Unit B: Unit B is covered by gray clayey silt (1.00–0.75
Haynesina germanica, Trochamina inflata, and Porosononion                     m b.s.; 0.67–0.42 m bsl) showing a low calcium carbonate
sp. were detected. The ostracod spectrum includes a high                      content, increased electrical conductivity, and a decrease
abundance of Cyprideis torosa, Xestoleberis sp., and Loxocon-                 in pH. The microfauna comprise more species than before
cha pontica as well as a medium abundance of Leptocythere                     and, in addition to the ones mentioned above, Quinquelo-
histriana, Pontocythere sp., and Palmoconcha sp.                              culina sp. and L. pontica. However, the specimens are few
   Unit D: While unit E is sand-dominated, unit D con-                        and not well preserved.
sists of gray clayey silt (6.60–2.75 m b.s.). The electrical                     Unit A: The uppermost unit (0.75–0.00 m b.s.; 0.42 m
conductivity is higher than in the units below, indicat-                      bsl to 0.33 m asl) consists of coarse sand and gravel with
ing a stronger marine influence. Below 4.55 m b.s., the                       abundant shell debris. In the lower part, the sediment
sediment is enriched by large mollusc fragments (mostly                       is of a gray color; it changes to a reddish yellow toward
C. glaucum and Abra ovata). Three AMS 14 C ages on ar-                        the top (see the photograph in Figure 5). No microfauna
ticulated specimens of C. glaucum were generated (CHO                         were detected. The geochemical fingerprint shows high
Figure 5 Core CHO 4 with a photograph of the sediments, geochemical results, microfossil content, and facies interpretation.
Figure 6 Core CHO 2 with a photograph of the sediments and example EDX spectra, grain size analysis, and facies interpretation.
kilometres) seawards due to the extremely shallow shelf                              level changes, the studied area is of special interest since
of the Sea of Azov. This has never been evidenced in ar-                             it is only 50 km away from the eponymous site of the
chaeological excavations or surveys nor has it been re-                              “Phanagorian Regression” on the Taman Peninsula (SW
ported in historical accounts (for a detailed discussion, see                        Russia).
Fouache et al., 2011 as well as Brückner et al., 2010).                                Phanagoria is an ancient Greek settlement founded in
   In addition, local tectonic processes have not been con-                          the 8th/7th century B.C. Its lower city is now partly sub-
sidered. Tectonic processes have recently been shown to                              merged by 1–2 m in the Gulf of Taman. This observation
contribute to local differences in sea level changes in                              was first described by Blavatsky (1961) who attributed
the Mediterranean (Sivan et al., 2004; Morhange, 2005;                               it to a temporary fluctuation in sea level. His opinion
Marriner et al., 2006; Vött, 2007) and the Black Sea                                still prevails in the literature. There are also other areas
(Giosan et al., 2006; Giosan, Filip, & Constantinescu,                               along the Black Sea coastline where temporary sea level
2009; Brückner et al., 2010; Fouache et al., 2011; Kelter-                          changes can be observed. However, they cannot be used
baum et al., 2011). Considering the postulated rapid sea                             to establish a general sea level curve for the Black Sea
Figure 7 Sea level curve in the vicinity of Lake Chokrak for the last six millennia. The length of the arrow equates to the vertical offset of the dated material
caused by the tectonic subsidence (∼0.4 m/ka) in the area of Lake Chokrak’s sand barrier.
region since every single area shows different fluctuation                         unfortunately, no peat layers occur, which are much
patterns. One has to consider various local factors that                           more reliable sea level indicators (e.g., Allen, 1990; Pi-
influence sea level. Brückner et al. (2010) and Fouache                           razzoli, 1991, 1996, 2005; Morhange, 2005; Brückner
et al. (2011) strongly oppose the concept of the major                             et al., 2010; Fouache et al., 2011). However, we recently
Holocene regression/transgression cycles postulated for                            showed that molluscs at stratigraphic boundaries can also
the Black Sea with arguments based on sedimentology,                               be used for a sea level reconstruction (Kelterbaum et al.,
hydrodynamics, archaeology, and historical accounts. In                            2011). Considering the fact that the living habitat of C.
contrast to curves from the Black Sea given by Balabanov                           glaucum extends to 1 m bsl at sedimentary boundaries
(2009) and others, our studies could not verify any pro-                           (e.g., shallow marine to beach barrier sedimentation; cf.
longed period of significant regression (>1 m) in sea level                        Vött, 2007), a relative palaeo-sea level envelope of c. 1
for the last five millennia, which would have been repre-                          m has to be estimated. Yet the sea level curve presented
sented by erosional disconformities or palaeosols in the                           here should only be regarded as a first approximation for
sediment columns.                                                                  the study area.
                                                                                      As can be stated so far from the coring results, around
                                                                                   3000 B.C. sea level was approximately 7 m lower than
A sea level curve for the study area                                               today at the Chokrak sand barrier (based on the dating
In the study region, recent investigations on sea level                            results for CHO 4/18 and CHO 4/17). Cores CHO 4/13
changes have focused mainly on the Taman Peninsula,                                (2035–1764 cal B.C.) and CHO 5/14 (1907–1658 cal B.C.)
that is, the eastern side of the Strait of Kerch. Sea level                        document that the water table rose by 1.50 m to a level of
evolution around the Kerch Peninsula was investigated                              5.30 bsl during the following millennium. By then, shal-
by Porotov (2007) and Danovskiy et al. (2009), but only                            low marine conditions prevailed that lasted until the be-
to a limited extent. A RSL curve can now be recon-                                 ginning of the 1st millennium B.C. Thereafter, the sand
structed based on the results of the core sediment records                         barrier formation was initiated, most probably during the
from Lake Chokrak. The curve includes nine AMS 14 C                                8th century B.C. (sample CHO 4/8); it ended during the
ages for sea level markers (Figure 7 and Table I). Only                            5th century A.D. (sample CHO 5/5). Thus, the evolution
sample CHO 5/8, a single valve, had to be rejected due to                          of this coastal feature took 1500 or so years.
reworking. In most cases, in situ finds of articulated speci-                         The estimated sea level envelope for the Chokrak Lake
mens of C. glaucum, found at facies changes, were used;                            sand barrier—which can be refined as new data be-
come available—shows a quasi-linear progression with                                 of the tectonic subsidence of the area, which is part of
differing rates of sea level rise (Figure 7): for the period                         the Indolo-Kuban and Kerch-Taman troughs (cf. Figure
3000 B.C. to 500 A.D., sea level rose c. 1.5 m/ka, slowing                           1; Nikonov, 1995b; Ershov et al., 1999; Saintot & Ange-
down to a rate of 1 m/ka during the last 1500 years. These                           lier, 2000).
findings correlate well with a sea level rise of 0.7 m during                           The sea level curve shows a gradual postglacial sea
the last 1000 years in the area of the city of Kerch, which                          level rise over the last five millennia including a deceler-
Boguslavsky et al. (1998) postulated for this geosynclinal                           ation since c. 1000 B.C. for the Lake Chokrak region. The
structure. Since Lake Chokrak is also situated in such a                             general trend of this curve supports the sea level curves
structure, it may be influenced by the same tectonic ef-                             reconstructed for the Taman Peninsula by Brückner
fects.                                                                               et al. (2010), Fouache et al. (2011), and Kelterbaum et al.
   Kelterbaum et al. (2011) have demonstrated that in the                            (2011) and for the Kerch Peninsula by Porotov (2007).
synclines of the Taman Peninsula subsidence is the driv-                             The curve presented here is also similar to curves from
ing factor for the local sea level rise. Tectonics is also likely                    the Mediterranean (Morhange, 2005; Müllenhoff, 2005;
to be the important factor in the area of the Chokrak sand                           Brückner et al., 2006; Marriner et al., 2006; Vött, 2007;
barrier. Compaction can be neglected due to the sandy                                Engel et al., 2009) and the Black Sea region (Giosan et al.,
environment and the lack of organic layers. Therefore,                               2006; Giosan, Filip, & Constantinescu, 2009; Brückner et
tectonic movements, plus the glacioeustatic sea level rise                           al., 2010; Fouache et al., 2011; Kelterbaum et al., 2011).
due to the postglacial warming, are the key processes that
influenced the postglacial sea level rise in the study area.
                                                                                     Palaeogeographical Scenarios for Lake Chokrak
   As an indicator of differential tectonics, one can com-
pare the 14 C ages of the sand barrier with those around                             Summarizing the results from the sediment cores, we are
the Chokrak Peninsula (Figure 7). It is noteworthy that                              able to reconstruct palaeogeographic maps for four time
the Chokrak Peninsula samples (CHO 2/4 and 2/3) dif-                                 slices: 3000 cal B.C., 1500 cal B.C., 500 cal B.C., and 500
fer significantly in their vertical position from the sand                           cal A.D. (Figure 8).
barrier samples (CHO 4/8, 4/9, and 5/5), although they                                  Around 3000 cal B.C., the postglacial marine transgres-
date back to the same period. This might be due to the                               sion reached the area of Lake Chokrak. The former syn-
vertical habitat of the dated specimens (Vött, 2007) or to                          clinal structure with its small river valleys was inundated
vertical tectonic movements and compaction. However,                                 by the sea. Thus, pretransgressive colluvial deposits were
nearly stable tectonics can be estimated for the whole                               covered by marine sediments. The ancient coastline was
hard rock structures of the study area and the north-                                situated somewhere between the Chokrak Peninsula and
ern Kerch Peninsula as a whole, since no generations of                              the (later) sand barrier, adapted to the palaeotopography.
notches are visible bsl and asl apart from the recent one.                           After 2000 cal B.C., the decelerating sea level rise led to
Therefore, the two radiocarbon dates obtained from in situ                           the establishment of strong coastal long-shore currents,
molluscs from coring site CHO 2 (CHO 2/4 and 2/3) may                                which became the dominant factor shaping the coast-
reflect the palaeo sea level very well. CHO 2/4 dates the                            line. While cliff erosion of the nearby soft rock (marls)
contact from terrestrial to marine environments, while                               provided the material, sand spits and bars formed to the
CHO 2/3 dates the last influence of lagoonal sedimenta-                              north of the marine embayment. Around 1000 cal B.C.,
tion.                                                                                the ongoing transgression reached the area of the later
   The vertical discrepancy between the peninsula and the                            Chokrak Peninsula. Cliff erosion of the Pleistocene soft
adjacent sand barrier may therefore be interpreted as the                            rock led to the formation of the peninsula. When the
rate of subsidence of the Chokrak sand barrier, which can                            Greeks erected their settlement in the middle of the 1st
be estimated to ∼0.4 m/ka. One has to bear in mind that                              millennium B.C., the peninsula was surrounded by the
the lake forms a large depression reaching inland, its gen-                          sea. The water depth did not exceed 1.5 m. The sand bar-
esis being due to a N–S striking fault line.                                         rier was still accessible by boat via small channels, like
   These results document the influence of local tecton-                             those visible today in the western part of the sand barrier
ics on the sea level evolution. Since earthquakes have                               after heavy rainfall.
taken place on the Kerch peninsula and in the whole re-                                 However, as today, the marine inundation of Lake
gion during historical times (e.g., Nikonov, 1994, 1995b;                            Chokrak was not permanent. The small navigable con-
Dotsenko & Ingerov, 2007; see also USGS marks in                                     nections to the Sea of Azov were closed rapidly by the
GoogleEARTH), they must also have influenced the re-                                 sediment supplied by the coastal long-shore drift. This is
gion of Lake Chokrak. There has been a continuous trans-                             in contrast to the Taman Peninsula, where the continu-
gression, which is first and foremost an effect of the gen-                          ous water supply of the Kuban River has kept these con-
eral glacioeustatic sea level rise, but also an expression                           nections open until today (Kelterbaum et al., 2011). In
Figure 8 Palaeogeographic reconstruction for the Lake Chokrak area at 3000 cal B.C., 1500 cal B.C., 500 cal B.C., and 500 cal A.D. For the last
reconstruction, the drying out of the lake during the summer has to be considered.
the following millennium, until the 5th century A.D., the                     i. A marine embayment evolved in the area of the
water depth became more and more shallow; during the                             present Lake Chokrak in the 5th millennium B.C. due
summer it probably dried out—as it does today. Torrential                        to the continued postglacial transgression.
rains triggered erosion that supplied the lake with sedi-                    ii. Fully marine conditions terminated in the 3rd mil-
ments, which led—and still does so today—to siltation.                           lennium B.C. when a sand spit started to form in
   This evolution explains the abandonment of the settle-                        the north of the study area. This was induced by the
ment on the Chokrak Peninsula. Due to the ongoing sil-                           slow down in sea level rise, allowing for long-shore
tation process, the site lost its initial strategic advantage.                   currents to become the driving factor in shaping the
Other settlements along the northern coastline (e.g., at                         coastal configuration.
Cape Zuk) took over its function.                                           iii. The enclosure of the former marine embayment took
                                                                                 place in the 2nd/1st millennium B.C., when the sand
                                                                                 barrier of Lake Chokrak had become fully developed.
                                                                            iv. During the time of the Greek colonization in the
                                                                                 Black Sea region, a fortified settlement was erected
CONCLUSIONS                                                                      on the Chokrak Peninsula in the middle of the 1st
For the first time, geoarchaeological and palaeogeograph-                        millennium B.C.—by this time the peninsula was sur-
ical research was carried out on the northern coastline of                       rounded by a lagoon.
the Kerch Peninsula on Crimea, Ukraine. The major find-                      v. A local sea level curve has been reconstructed for the
ings may be summarized as follows:                                               area of Lake Chokrak using 14 C-dated sea level indi-
    cators. It reflects the continued glacioeustatic sea level                          Locri-Epizephiri, Calabria, Italy. Geoarchaeology, 26,
    rise, plus a local subsidence trend at the sand barrier                             33–60.
    of several tens of cm per 1000 years. However, it also                           Blavatsky, V.D. (1961). Podvodnye raskopki Fanagorii v 1959
    reflects a stable situation near the previously settled                             g (Underwater Excavations of Phanagoria in 1959).
    Chokrak Peninsula. Thus, the study shows the im-                                    Sovietskaya arkheologia, 1, 277–279 (in Russian).
    portance of local tectonics.                                                     Boguslavsky, S.G., Kubryakov, A.I., & Ivaschenko, I.K.
vi. The RSL curve does not show any major or minor                                      (1998). Variations of the Black Sea level. Physical
    transgression/regression cycles for the last five mil-                              Oceanography, 9(3), 199–208.
    lennia. This is another piece of evidence that chal-                             Brückner, H., Müllenhoff, M., Gehrels, R., Herda, A.,
                                                                                        Knipping, M., & Vött, A. (2006). From archipelago to
    lenges the existence of an oscillating Black Sea level
                                                                                        floodplain — geographical and ecological changes in
    during mid- and late-Holocene times.
                                                                                        Miletus and its environs during the past six millennia
                                                                                        (Western Anatolia, Turkey). Zeitschrift für
  Finally, it should be stated that it would be worth tak-                              Geomorphologie N.F., 142(Suppl.), 63–83.
ing a closer look at the wash-over fans situated behind the                          Brückner, H., Kelterbaum, D., Marunchak, O., Porotov, A., &
recent beach barrier (Figure 2) as they can be good indi-                               Vött, A. (2010). The Holocene sea level story since 7500 BP
cators of high energy wave events, triggered by storms or                               — lessons from the Eastern Mediterranean, the Black and
tsunamis.                                                                               the Azov Seas. Quaternary International, 255, 160–179.
                                                                                     Caraion, F.E. (1967). Fauna Republicii Socialiste Romania. —
Funding of this research by the Fritz Thyssen Foundation (Ger-                          Crustacea (Ostracoda) 4. Fascicula, 10. Editura Academiei
many) is gratefully acknowledged (AZ 20.08.0.010). We thank                             Republicii Socialiste Romania, Bucuresti.
the Demetra Foundation (Kerch) for logistic support on Kerch                         Chepalyga, A.L. (1984). Inlandseabasins. In A.A. Velichko,
peninsula. Special thanks go to the journal reviewers Ron Mar-                          H.E. Wright, Jr., & C.W. Barnowsky (Eds.), Late quaternary
tin (University of Delaware), an anonymous reviewer, and Jamie                          environments of the Soviet Union (English edition; pp.
Woodward (The University of Manchester) for helpful com-
                                                                                        229–247). Minnesota, MN: University of Minnesota Press.
ments and further suggestions to improve the paper. The SEM
images of the microfauna were provided by Peter Frenzel (Uni-                        Danovskiy, A.A., & Maslennikov, A.A. (2006). Poselenie
versity of Jena). Kirstin Jacobson is thanked for polishing the                         epokhi bronzi Generalskoe-Zapadnoe (yugo-zapadny
English.                                                                                sklon) (Bronze epoch settlement “Generalskoe-zapadnoe”
                                                                                        (south western slope) — preliminary study results) Ancient
                                                                                        Bosporus, T. 9, Moscow, 77–89 (in Russian).
REFERENCES                                                                           Danovskiy, A., Dikarev, V., Kovalchyk, A., & Maslennikov,
Allen, J.R.L. (1990). The formation of coastal peat marshes                             A.A. (2009). Complex archaeological and
   under an upward tendency of relative sea-level. Journal of                           geomorphological studies on the northern coast of
   the Geological Society of London, 147, 743–                                          Kerchenskiy peninsula (Crimea, Ukraine). — In Extended
   745.                                                                                 Abstracts of the Fifth Plenary Meeting and Field Trip of
Artyukhin, I.V., Grudinova, L.Y., & Obukhovskaya, I.N.                                  IGCP 521 “Black Sea-Mediterranean corridor during the
   (1989). Ravnovesnye beregovye formy kak osnova dlia                                  last 30 ky: sea level change and human adaptation,” and
   paleogeograficheskikh rekonstruktsii. (Similarity coastal                            INQUA 0501 “Caspian-Black Sea-Mediterranean Corridor
   forms as a basis for paleogeographic reconstructions).                               during the last 30 ky: sea level change and human adaptive
   Izvestiia AN SSSR, Seriia Geograficheskaiia, 3, 91–96 (in                            strategies” (Çanakkale, Turkey, August 22–31, 2009) (pp.
   Russian).                                                                            55–57). Izmir: DEU Publishing House.
Balabanov, I.P. (2007). Holocene sea-level changes of the                            Degens, E.T., & Ross, D.A. (1972). Chronology of the Black
   Black Sea. In V. Yanko-Hombach, A.S. Gilbert, N. Panin, &                            Sea over the last 25000 yrs. Chemical Geology, 10,
   P.M. Dolukhanov (Eds.), The Black Sea flood question (pp.                            1–16.
   711–730). Dordrecht: Springer                                                     Dodonov, A.E., Tchepalyga, A.L., Mihailescu, C.D., Zhou,
Balabanov, I.P. (2009). Paleogeographical background to                                 L.P., Markova, A.K., Trubikhin, V.M., Simakova, A.N., &
   formation of modern natural conditions of the Caucasus                               Konikov, E.G. (2000). Last-interglacial records from central
   littoral Holocene terraces and their long-term development                           Asia to the northern Black Sea shoreline: stratigraphy and
   forecast. Moscow, Vladivostok: Dalnauka.                                             correlation. Netherlands Journal of Geoscience, 79(2/3),
Balabanov, I.P., & Izmailov, Y.A. (1988). Izmenenie                                     303–311.
   urovennogo i gidrokhimicheskogo rezhima Chernogo i                                Dotsenko, S.F., & Ingerov, A.V. (2007). Analysis of
   Azovskogo morei za poslednie 20000 let. Vodnye resursi, 6,                           observations and methods for calculating hydrophysical
   54–62 (in Russian).                                                                  fields in the ocean — characteristics of tsunami waves in
Bernasconi, M.P., & Stanley, J.-D. (2011). Coastal margin                               the Black Sea according to the data of the measurements.
   evolution and postulated “Basin-Shipyard” area at ancient                            Physical Oceanography, 17(1), 17–28.
Dzhanelidze, C.P. (1980). Paleogeografiia Gruzii v golotsene           Herbin, J.P., Saint-Germès, M., Maslakov, N., Shnyukov, E.F.,
   (Palaeogeography of Georgia in the Holocene). Tbilisi:                & Vially, R. (2008). Oil seeps from the ”Boulganack” mud
   Metsniereba (in Russian).                                             volcano in the Kerch Peninsula (Ukraine –Crimea), study
Dzhanelidze, Z.C. (2007). Amplituda Fanagori’skoi regressii              of the mud and gas: inferences fort he petroleum potential.
   Chernogo morya ([The amplitude of Phanagorian                         Oil & Gas Science and Technology — Rev. IFP, 63(5),
   regression of Black sea). In Problems of the coastal zone             609–628.
   management and sustainable development. Abstracts XXII              Hughen, K.A., Baillie, M.G.L., Bard, E., Bayliss, A., Beck,
   International Coastal conference (Gelendjik, Russia, May              J.W., Bertrand, C., Blackwell, P.G., Buck, C.E., Burr, G.,
   16–20) (pp. 327–328).                                                 Cutler, K.B., Damon, P.E., Edwards, R.L., Fairbanks, R.G.,
Engel, M., Knipping, M., Brückner, H., Kiderlen, M., & Kraft,           Friedrich, M., Guilderson, T.P., Kromer, B., McCormac,
   J.C. (2009). Reconstructing middle to late Holocene                   F.G., Manning, S., Bronk Ramsey, C., Reimer, P.J., Reimer,
   palaeogeographies of the lower Messenian plain                        R.W., Remmele, S., Southon, J.R., Stuiver, M., Talamo, S.,
   (southwestern Peloponnese, Greece): coastline migration,              Taylor, F.W., Plicht van der, J., & Weyhenmeyer, C.E.
   vegetation history, and sea level change. Palaeogeography,            (2004). Marine04. Marine radiocarbon age calibration,
   Palaeoclimatology, Palaeoecology, 284, 257–270.                       0-26 cal kyr BP. Radiocarbon, 46(3), 1059–1086.
Ershov, A.E., Brunet, M.-F., Korotaev, M.V., Nikishin, A.M.,           Izmailov, Y.A. (2005). Evolutsionnaya geografiya poberejiy
   & Bolotov, S.N. (1999). Late Cenozoic burial history and              Tchernogo I Azovskogo morey (Evolutionary geography of
   dynamics of the Northern Caucasus molasse basin:                      the coastlines of Black and Azov Sea’s). Sochi:
   implications for foreland basin modelling. Tectonophysics,            Lazarevskaya Polygrafiya, 174 p. (in Russian).
   313, 219–241.                                                       Izmailov, Y.A., Arslanov, K.A., Tertychnaya, T.V., & Chernov,
Fedorov, P.V. (1977). Pozdnechetvertichnaia istoriia                     S.B. (1989). Rekonstruktsiia i datirovanie golocenovykh
   Chernogo moria i razvitie iuzhnykh morei Evropy (Late                 beregovykh linii moria v del’te Kubani Vostochnoe
   quaternary history of the Black Sea and evolution of the              Azovo-Chernomor’e. (Reconstruction and dating of
   southern seas of Europe). In P.A. Kaplin & F.A.                       Holocene shorelines in the delta of the Kuban River,
   Shcherbakov (Eds.), Paleogeografiia i Otlozheniia                     eastern coast of the Sea of Azov and Black Sea). Vestnik
   Pleistotsena Iuzhnykh Morei SSSR (Pleistocene                         LGU 321(suppl. 7), 61–69 (in Russian).
   paleogeography and sediments of the southern seas of the            Jones, G.X., & Gagnon, A.R. (1994): Radiocarbon chronology
   USSR) (pp. 25–32). Moscow: Nauka (in Russian).                        of Black Sea sediments — deep sea research part I. Oceanic
Fedorov, P.V. (1982). Poslelednikovaia transgressiia Chernogo            research papers, 41(3), 531–557.
   moria i problema izmeneniia urovnia okeana za poslednie             Kelterbaum, D., Brückner, H., Porotov, A., Schlotzhauer, U.,
   15000 let (The post-glacial transgression of the Black Sea            & Zhuravlev, D. (2011). Geoarchaeology of Taman
   and the problem of ocean level change during the last                 Peninsula (SW Russia) — the example of the ancient
   15000 years). In P.A. Kaplin, R.K. Klige, & A.L. Chepalyga            Greek settlement of Golubitskaya 2. Die Erde 142(3),
   (Eds.), Kolebaniia urovnia morei i okeanov za 15000 let               235–258.
   [Sea and oceanic level fluctuations over the last 15000             Kraft, J.C., Bückner, H., Kayan, I., & Engelmann, H. (2007).
   years] (pp. 151–156). Moscow: Nauka (in Russian).                     The geographies of ancient Ephesus and the Artemision in
Fornasier, J., & Böttger, B. (Eds.) (2002). Das Bosporanische           Anatolia. Geoarchaeology, 22(1), 121–149.
   Reich: Der Nordosten des Schwarzen Meeres in der Antike.            Marriner, N., Morhange, C., Doumet-Serhal, C., & Carbonel,
   Zaberns Bildbände zur Archäologie, von Zabern, Mainz.               P. (2006). Geoscience rediscovers Phoenicia’s buried
Fouache, E., Kelterbaum, D., Brückner, H., Lericolais, G.,              harbors. Geology, 34(1), 1–4.
   Porotov, A., & Dikarev, V. (2011). The Late Holocene                Martin, R.E., & Yanko-Hombach, V. (2011). Rapid Holocene
   evolution of the Black Sea — a critical view on the                   sea-level and climate change in the Black Sea: an
   so-called Phanagorian regression. Quaternary                          evaluation of the Balabanov sea-level curve. In I.
   International, doi: 10.1016/j.quaint.2011.04.008                      Buynevich (Ed.), Geology and geoarchaeology of the Black
Gajdukevic, V.F. (1971). Das bosporanische Reich. Berlin,                Sea region: beyond the flood hypothesis (pp. 51–58).
   Amsterdam.                                                            Boulder, Colorado: Geological Society of America.
Giosan, L., Donnelly, J.P., Constantinescu, S., Filip, F.,             Maslennikov, A.A. (1998). Ellinskaia khora na kraiu
   Ovejanu, I., Vespremeanu-Stroe, A., Vespremeanu, E., &                Oikumeny: Selskaia territoriia evropeiskogo Bospora v
   Duller, G.A.T. (2006). Young Danube delta documents                   antichnuiu epokhu (Hellenistic settlements at the border of
   stable Black Sea level since the middle Holocene:                     the “populated universe”: rural territories of the European
   morphodynamic, paleogeographic, and archaeological                    Bosporus in the ancient epoch). Moscow: Izd-vo “Indrik”
   implications. Geology, 34(9), 757–760.                                (in Russian).
Giosan, L., Filip, F., & Constantinescu, S. (2009). Was the            Morhange, C. (2005). Relative sea-level changes in Marseille’s
   Black Sea catastrophically flooded in the early Holocene?             ancient harbors (France) during the Late Holocene.
   Quaternary Science Reviews, 28, 1–6.                                  Zeitschrift für Geomorphologie, 137(Suppl.), 23–28.
Müllenhoff, M. (2005). Geoarchäologische,                                             dune system evolution in the Danube Delta, NW Black Sea
   sedimentologische und morphodynamische                                               basin. Journal of Coastal Research, SI 56, Proceedings of
   Untersuchungen im Mündungsgebiet des Büyük Menderes                               the 10th International Coastal Symposium, 347–351.
   (Mäander), Westtürkei. Marburg/Lahn: Marburger                                  Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W.,
   Geographische Schriften, 141.                                                        Blackwell, P.G., Bronk Ramsey, C., Buck, C.E., Burr, G.S.,
Murray, J.W. (1986). The living foraminiferida of                                       Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson,
   Christchurch harbour, England. Micropaleontology, 14,                                T.P., Hajdas, I., Heaton, T.J., Hogg, A.G., Hughen, K.A.,
   83–96.                                                                               Kaiser, K.F., Kromer, B., McCormac, F.G., Manning, S.W.,
Murray, J.W. (2006). Ecology and applications of benthic                                Reimer, R.W., Richards, D.A., Southon, J.R., Talamo, S.,
   foraminifera. New York: Cambridge University Press.                                  Turney, C.S.M., van der Plicht, J., & Weyhenmeyer, C.E.
Nevessky, E.N. (1967). Processe Osadkonakopleniya V                                     (2009). IntCal09 and marine09 radiocarbon age calibration
   Proibreznoi Zone Moria (The processes of sedimentation in                            curves, 0-50000 years cal BP. Radiocarbon, 51(4),
   the marine near shore zone). Moscow: Nauka, p. 255 (in                               1111–1150.
   Russian).                                                                         Ross, D.A., Degens, E.T., & MacIlvaine, J. (1970). Black Sea:
Nikonov, A.A. (1994). Severe earthquakes in the Krasnodar                               recent sedimentary history. Science, 170(3954), 163–165.
   reservoir region – new evaluations. Hydrotechnical                                Ryan, W.B.F. (2007). Status of the Black Sea flood
   Construction, 28(5), 281–284.                                                        hypothesis. In V. Yanko-Hombach, A.S. Gilbert, N. Panin,
Nikonov, A.A. (1995a). Manifestations of young tectonic                                 & P.M. Dolukhanov (Eds.), The Black Sea flood question
   activity in the southern Azov and Kerch fault zones                                  (pp. 63–88). Dordrecht: Springer.
   (Crimea). Geotectonics, 28 (5), 380–390 (english                                  Ryan, W.B.F., Pitman, W.C., Major, C.O., Shimkus, K.,
   translation from Russian original).                                                  Moskalenko, V., Jones, G.A., Dimitrov, P., Görür, N.,
Nikonov, A.A. (1995b). The stratigraphic method in the study                            Sakınç, M., Seyir, H.I., & Yüce, H. (1997). An abrupt
   of large earthquakes. Quaternary International, 25, 47–55.                           drowning of the Black Sea shelf. Marine Geology, 138,
Panin, N., & Popescu, I. (2007). The northwestern Black Sea:                            119–126.
   climatic and sea level changes in the Upper Quaternary. In                        Saintot, A., & Angelier, J. (2000). Plio-Quaternary paleostress
   V. Yanko-Hombach, A.S. Gilbert, N. Panin, & P.M.                                     regimes and relation to structural development in the
   Dolukhanov (Eds.), The Black Sea flood question: changes                             Kerch–Taman peninsulas (Ukraine and Russia). Journal of
   in coastline, climate, and human settlement (pp. 387–404).                           Structural Geology, 22, 1049–1064.
   New-York: Springer.                                                               Saprykin, S.J. (2006). The Chora in the Bosporan kingdom. In
Pirazzoli, P.A. (1991). World atlas of sea-level changes.                               P.G. Bilde, & V.F. Stolba, (Eds.), (Black Sea studies, 4,
   Oceonography Series, 58. Amsterdam, London: Elsevier.                                (pp.273–288). Aarhus, Denmark: Aarhus University Press.
Pirazzoli, P.A. (1996). Sea-level changes. the last 20000 years.                     Sivan, D., Lambeck, K., Toueg, R., Raban, A., Porath, Y., &
   Chichester, UK: Wiley and Sons.                                                      Shirman, B. (2004). Ancient coastal wells of Caesarea
Pirazzoli, P.A. (2005). Sea-level indicators, geomorphic. In                            Maritima, Israel, an indicator for relative sea level changes
   M.L. Schwartz, (Ed.), Encyclopedia of coastal science.                               during the last 2000 years. Earth and Planetary Science
   Dordrecht: Kluwer.                                                                   Letters, 222, 315–330.
Porotov, A. (2007). Relative sea-level changes and                                   Soter, S. & Katsonopoulou, D. (2011). Submergence and
   submersion of archaeological sites along the northern                                uplift of settlements in the area of Helike, Greece, from the
   shoreline of the Black Sea. Méditerranée, 108, 29–36.                              Early Bronze Age to late antiquity. Geoarchaeology, 26,
Porotov, A.V., Gorlov, Y.V., & Kaitamba, M.D.K. (2003). On                              584–610.
   the paleogeography of the eastern Black Sea region in                             Stuiver, M., & Reimer, P.J. (1993). Extended 14 C database
   classic time (O Paleogeografii vostochnogo                                           and revised CALIB 3.0 14 C age calibration program. In M.
   Prichernomor’ya v Antichnuyu Epohu). PFIK. Vyp, 13,                                  Stuiver, A. Long, & R.S. Kra (Eds.), Calibration 1993.
   207–216 (in Russian).                                                                Radiocarbon, 35, 215–230.
Porotov, A.V., Gorlov, Y.V., Yanina, T.A., & Fouache, E.                             Vött, A. (2007). Relative sea level changes and regional
   (2004). The evolution of the Black sea’s coasal zone of                              tectonic evolution of seven coastal areas in NW Greece
   Taman peninsula in the late Holocene (Osobennosti                                    since the mid-Holocene. Quaternary Science Reviews, 26,
   razvitiya Chernomorskogo poberezh’ya Tamanskogo                                      894–919.
   poluostrova v pozdnem golotsene). Geomorfologiya, 4,                              Zubakov, V.A. (1988). Climatostratigraphic scheme of the
   63–77 (in Russian).                                                                  Black Sea Pleistocene and its correlation with the
Preoteasa, L., Roberts, H.M., Duller, G.A.T., &                                         oxygen-isotope scale and glacial events. Quaternary
   Vespremeanu-Stroe, A. (2009). Late-Holocene coastal                                  Research, 29, 1–24.