Preliminary Assessment of Submerged Beachrock and Tsunamigenic Deposit, Hasir Island, Marmara Archipelago, Turkey

and Alper Demirci

Faculty of Arts and Sciences Bitlis Eren University TR-13000 Bitlis, Turkey

††_{Department of Geophysical}

Engineering Faculty of Engineering

and Architecture Bitlis Eren University TR-13000 Bitlis, Turkey

ABSTRACT

Ertek, A.; Kılı¸c, E.; Erginal, A.E.; Ekinci, Y.L., and Demirci, A., 0000. Preliminary assessment of submerged beachrock and tsunamigenic deposit, Hasır Island, Marmara Archipelago, Turkey. Journal of Coastal Research, 00(0), 000–000. Coconut Creek (Florida), ISSN 0749-0208.

A preassessment of coexisted submerged beachrock and a fossil-laden near-shore deposit on the coast of Hasır Island, SW Marmara Sea, is presented based on depositional characteristics, two-dimensional electrical resistivity tomography (ERT) survey and accelerator mass spectrometry (AMS) radiocarbon ages. ERT-derived geophysical images clearly showed the subsurface position and contact relationship of submerged beachrock under the studied beach. Textural features of beachrock are typical solely of marine-phreatic cementation, including consecutively developed cement fabrics, i.e. micrite coatings, radial aggregates consisting of scalenohedral high-Mg calcites, and reticulated needles of aragonite. The intertidal cementation of beachrock took place between 2940 and 2470 YBP when the level of the Marmara Sea was about 1.5 m lower than that of the present. Its purely submerged nature is likely concerned with rise in sea level in pursuit of the cementation period. Though dated between 2340 and 1590 YBP, the fossil-rich near-shore deposit behind the studied beach could be of a tsunamigenic origin based on its sequence characteristics typical of such a high-energy event.

ADDITIONAL INDEX WORDS: Beachrock, near-shore deposit, tsunami, submergence, Marmara Sea, AMS radiocarbon ages, ERT, Turkey.

INTRODUCTION

The uplifted near-shore deposits and submerged beachrocks are among key indicators in determining sea-level changes and/or the rate of tectonic displacements on coastal areas in that they, in both cases, are found behind their original depositional environment. Considering that beachrock is mostly the result of intertidal cementation of beach materials during a period of sea-level stabilization through the precipi-tated connective cement comprised mostly of high-Mg calcite and aragonite, its totally submerged nature can be attributed to either tectonic subsiding or ensuing sea-level rise (Kindler and Bain, 1993; Stefanon, 1971) after cementation of its seaward dipping slabs. In the same manner, uplifted near-shore deposits laden with marine fossils also compose a notable marker as long as their absolute age and shoreline angles are determined. The facies characteristics and age of such raised deposits are mostly considered on a preferential basis. On the other hand, fossil-rich deposits reminiscent of the raised marine deposits may be of a distinctive origin based on the

presence of visible chaotic sequence characteristics, such as disorderly accumulated marine gastropods and shells.

In this preliminary note, we present new data on the coexistence of a fossil-rich near-shore deposit together with a submerged beachrock on the coast of the Hasır Island, Marmara Sea, Turkey, where these occurrences have not been documented so far. The Marmara Sea is an inland sea between the Aegean Sea and the Black Sea and comprises the central part of the Turkish Strait System, connecting to them through the Strait of ¸Canakkale (Dardanelles) to the south and the Strait of Istanbul (Bosphorus) to the north, respectively. Its key importance in understanding sea-level exchanges between the Mediterranean and the Black Sea has been a focus of interest since early investigations in the second half of 19th century. Late Pleistocene marine terrace deposits along the coast of this inland sea have been studied with regard to their rich fossil contents and stratigraphic characteristics (see Sakın¸c and Yaltırak [1997] for a detailed review). Absolute dating of these deposits is greatly restricted (Yaltırak et al., 2002). Herein, we aim to contribute to the understanding of the Late Holocene sea-level changes in that part of the Marmara Sea. Based on microanalytical data, electrical resistivity tomography (ERT), stable isotopes, and accelerator mass spectrometry (AMS) radiocarbon dating results, Holocene deposits are first reported from an island coast of the Marmara Archipelago.

DOI: 10.2112/JCOASTRES-D-12-00177.1 received 7 September 2012; accepted in revision 21 May 2014; corrected proofs received 30 July 2014; published pre-print online XX Month XXXX.

*Corresponding author: aerginal@gmail.com Ó Coastal Education & Research Foundation 2014

METHODS

The studied deposits are placed on the east coast of the Hasır Island (408290N, 278340E) at the western part of the Marmara Sea, NW Turkey (Figure 1a). The island has a total surface area of 0.016 km2and forms one of the small islands of Marmara Archipelago, composing 24 islands and cays, of which four (Marmara, Avsa, Pasalimanı, and Ekinlik islands) are inhab-ited because of their surface area over 1 km2. Having a total area of 165 km2, they rise abruptly from the wide SW shelf of the Marmara Sea and structurally form part of the Kapıda˘g Peninsula (Figures 1b and c). The islands generally consist of magmatic and metamorphic rocks, such as granite, granodio-rite, marble, and schist (Ardel and Kurter, 1973). The archipelago is mainly Mediterranean in climate, characterized by relatively hot and dry summers and cold and rainy winters. The area receives an average annual precipitation of about 704 mm. The average temperature is 14.1 8C. Slightly different than typical Mediterranean Climate, a Marmara Transitional Climate prevails in the islands (Tun¸cdilek et al., 1987).

The samples of three beachrocks and five fossil shells were collected for analyses and radiocarbon dating. Studied sites are shown in Figure 2a. Energy dispersive X-ray spectrometry and X-ray Diffractometry (Philips X’Pert Pro) analyses of beachrock cements were carried out to identify the nature of beachrock cements. Scanning electron microscopy (SEM) images were taken to examine the microfabric of connective cements. The total carbonate within beachrocks was measured using a calcimeter. Stable isotope (d13C and d18O) values of connective carbonates within beachrock and fossil shells in near-shore deposits were analyzed in the Environmental Isotope Labora-tory at the University of Arizona. AMS radiocarbon dating of

beachrock carbonates and fossil shells were carried out in the radiocarbon dating laboratory of BETA, Miami, USA.

Considering the successful applications of the ERT technique in coastal environs (Erginal et al., 2012, 2013a,b,c), 120-m- and 60-m-long transects, parallel to the coastline and placed about 8 m away from each other, were taken to reveal the subsurface structure. Taking transects perpendicular to the coastline along the beach comprised primarily of a mixture of sands and pebbles together with fragments of seashells were not possible due to its narrow width (maximum 4 m). Thus, measurements were carried out parallel to coastline both on the beach surface (Trs1) and on the slightly undulated surface underlain by fossil-rich near-shore deposits (Trs2). The lengths of transects were also arranged based on space limitation in the site. Apparent resistivity data were acquired using dipole-dipole electrode configuration via the GF ARES multielectrode resistivity-meter system, and the inversions of the measured apparent resistivity data sets were achieved by a two-dimensional tomographic inversion scheme using the RE-S2DINV software package (Loke and Barker, 1996). The inversion process, based on smoothness-constrained least squares (Sasaki, 1992) implemented by a quasi-Newton optimization technique (Loke and Barker, 1996), produced resistivity tomograms after five iterations with root mean square (RMS) errors of 2.4% and 6.5% for transects Trs1 and Trs2, respectively.

RESULTS

Following research findings in respect of the origin, composition and depositional environment of the studied beachrock and fossil-laden near-shore deposit were obtained on the basis of field notes, petrographic, micromorphologic, and paleontological examinations as well as a geophysical imaging survey.

Lithology and Diagenetic Features of Beachrock

Having a total thickness of about 1.5–2 m, very tightly indurated beachrock beds with a length of about 150 m run

Figure 1. Location map of the study area (a), and Southern Marmara Archipelago and studied sites (b) and (c) on Google earth images.

Figure 2. View of the submerged beachrock (a) and thin sections (b) and (c) showing algae oncoids, coral fragments, quartz grains, and various lithoclasts weakly amalgamated within a calcite cement.

parallel to the coastline on the western part of the Hasır Island (Figure 2a). Its most remarkable feature is that the beds are completely submerged. Underwater observations showed the extension of slightly (,58) inclined submerged beachrock beds up to about 30 m offshore, terminating at about2.5 m at most the seaward extent.

Petrographic thin section analyses displayed that beach-rock principally comprised algae oncoids, coral fragments, little polycrystalline quartz grains, and lithoclasts of micaschist and sandstone with tiny shell fragments (Figures 2b and c). The void ratio is high, and siliciclastic grains are cemented weakly by calcite cement. Plagioclase and biotite form primary mineral components that fill the void between sand-sized grains. These minerals, together with quartz with wavy extinction, are indicative of derivation from metamorphic and magmatic rocks that crop out in the island. Poorly developed roundness indicates either short-distance littoral drift of grains or proximity to the sediment sources.

SEM analyses displayed that grains comprised angular or poorly rounded medium to coarse sands, which mostly range in size between 250 lm and 1 mm (Figure 3a). As evidence of the early stage of sequential cementation (Neumeier, 1999) precipitated from supersaturated pore waters (Vieira and De Ros, 2007), grain surfaces are partially or wholly perfused by encrusting cement consisting of thin (,20 lm) micrite crystals (Figure 3b). Closer examination of these coatings reveals the predominance of scalenohedral high-Mg calcite crystals on micrite coatings (Figures 3c and d), indicative of precipitation from marine waters (Holail and Rashed, 1992). These equigranular micrite crystals with an average size of 4 lm entirely bind grain surfaces and constitute radial aggregates, which grow preferentially on siliciclastic substrates or micrite envelopes (Vieira and De Ros, 2007).

The third generation of cement that grows on scalenohe-dral crystals is represented by aragonite needles (Figure 3e),

which do not represent isopachous fringes or rims as usually encountered, but rather display randomly oriented inter-laced crystals thinner than 3 lm. The length of these crystals ranges from 5 lm to 10 lm. These reticulated needles, within which some ostracods and coccolith Emiliania huxleyi are embedded (Figure 3f), are typical for early marine-phreatic diagenesis (Beier, 1985).

Subsurface Structure of Beachrock

Figure 4 shows the ERT images obtained by a two-dimensional inversion process. The image in the upper panel illustrates the subsurface resistivity variation beneath tran-sect taken on the beach (Trs1). Nine data levels with an electrode spacing of 3 m yielded a depth range of about 6.5 m. The tomogram shows a clear resistivity contrast between beachrock and seawater-saturated beach material. The upper-most layer presents the beach material, marked by A, with low resistivity values less than about 5 ohm.m. Black dashed lines indicate beachrock, which is highlighted with B. It is seen that the buried beachrock slabs lie at the depths of about 0.5–2.5 m below the present sea level. Slightly tilted toward the NW, the cemented beds compose a blocky structure. A single deformed block at a closer position to the surface is also observed in the middle of the section. The thicknesses of beds reach more than 3 m in some parts of the tomographic image, exceeding that of the submerged beachrock. Nevertheless, this could be decep-tive because of the similar composition of the overlying and underlying unconsolidated beach sediments, as confirmed by resistivity values. The underlying unit is characterized by low resistivity values, such as the uppermost layer, and the low resistivity values result from the salinity because of the seawater input. Thus, it can be mentioned that the ERT image showed the tectonically tilted position of submerged beachrock under the beach sediments.

Composition of Near-shore Deposit

The studied near-shore deposit is placed at about 8 m behind the submerged beachrock. Overlain by an unconsol-idated colluvial layer with heterometric nature, it is a

0.75-m-Figure 3. SEM images from beachrock: (a) poorly rounded coarse sands; (b) closer view of grains encrusted by micrite; (c) and (d) closer images of (b) showing scalenohedral high-Mg calcite crystals on micrite coatings; (e) aragonite needles; and (f) ostracod and coccolith Emiliania huxleyi embedded within reticulated needles of aragonite.

Figure 4. Two-dimensional ERT images along transects Trs1 and Trs2 obtained by two-dimensional inversion process. (a) Beach material with high saturation level; (b) beachrock; (c) fossil-rich near-shore deposit; and (d) fractured metamorphic basement with high saturation level.

thick fossil-rich unit that includes small angular metamor-phic rock fragments and contains a considerable amount of Cerithium vulgatum, Gibbula albida, and lesser amounts of Patella vulgata, Pecten maximus, and Cardium edule fossils (Figures 5a and b). The whole unopened mollusc forms are common, indicating the original deposition area. The most spectacular characteristic of the sequence is the widespread existence of sea snail Cerithium vulgatum. This marine gastropod together with other species is rife in the Pleisto-cene-raised coastal deposits of the so-called Marmara Forma-tion (Sakın¸c and Yaltırak, 1997; Yaltırak et al., 2002). Albeit in limited amounts, the foraminiferal analyses showed the presence of various benthic foraminifera with broken forms, such as Elphidium aculeatum (d’Orbigny, 1846), Elphidium crispum (Linnaeus, 1758), Elphidium macellum (Fichtel and Moll, 1798), Elphidium sp., Rosalina sp., and Quinquelocu-lina spp, indicating similar marine conditions as such in the present Marmara Sea.

The lower panel in Figure 4 demonstrates the ERT image obtained from the transect taken from the surface underlain by the fossil-rich near-shore deposit (Trs2), which is covered by a blanket of colluvium. Because the transect was taken at about 4 m above sea level, electrode spacings were set to 4 m, and 13 data levels were measured to reach the depth levels, as obtained from the tomogram Trs1. This configuration deter-mined the electrical resistivity variation at a depth range of about 10.8 m. As seen from the tomographic image, landward extent of the beachrock under the beach sediments is not observed beneath transect Trs2. The 0.75-m-thick near-shore deposit as well as the underlying and overlying colluvium, jointly marked with C, was identified by its relatively high resistivity values in proportion to underlying low resistivity metamorphic basement (about 7–15 ohm.m) in the tomogram Trs2 (Figure 4). The thickness of C ranges from about 2.5–4.5 m along transect Trs2. The metamorphic bedrock, marked by D in the tomogram, shows lower resistivity values than expected. As also observed from the field observations, this finding can be explained as increasing porosity in response to decreasing compaction attributable to the fractured nature of the metamorphic basement with high saturation level, which contributes to the lowering of the resistivity values. Addition-ally, based on the integration of field observations with ERT-derived geophysical survey results, a cross-sectional view of the studied site is demonstrated in Figure 6.

DISCUSSION

Based on cementation characteristics of submerged beach-rock and the aforementioned compositional and depositional traits of both coexisting deposits, two important implications, namely a lower stage of sea level during the formation of beachrock and an ensuing tsunamigenic event evidenced by the depositional characteristics preserved within the nearshore fossiliferous deposit, are discussed.

Sea-level Lowstand Confirmed by Beachrock

Composing most of the seaward extent of beachrock slabs, three samples of beachrock were collected from a depth of about 2–2.5 m for appraisal of radiocarbon ages. The AMS radiocar-bon ages from these fossils are given in Table 1. The samples were dated radiometrically, e.g., two connective bulk carbon-ates and one fossil residue of Ostrea edulis, which was found embedded within an upper level of beachrock. The two-sigma calibrated AMS radiocarbon ages fall in the range of about 3 ka and 2.5 ka BP. Based on results, the oldest strata formed between 2940 and 2730 YBP, which is followed by the upper layer with ages between 2730 and 2470 YBP. The oyster shell has also yielded the same age with the latter, suggesting synchronous deposition with the precipitation of authigenic aragonite cements. Stable isotope values from the oldest sample reveal that it is slightly enriched with13_{C that ranges}

between 0.91 and 1.2; however,18_{O values vary between0.96}

and1.44, showing slightly depletion of heavy isotopes. The aforementioned microfabric characteristics and stable isotope values of beachrock confirm the existence of cements precipitated solely from marine waters, indicating that beachrock took place under marine phreatic conditions. The present submerged nature of beachrock slabs can thus be explained by pondering the rise in sea level after the formation of beachrock with the lack of any evidence for tectonic displacement. Previous data regarding Late Holocene sea levels along the Northern Aegean Sea coastline of Turkey indicated that the rising trend of sea level ceased at 6000 YBP based on both radiocarbon-dated mollusks and archaeological evidence from Karamenderes delta and was followed by a 2-m sea-level fall (Kayan, 1999). This is also confirmed by new data from beachrock on Bozcaada Island (Erginal, Kiyak, and Ozturk, 2010) and by several previous studies (Lambeck et al., 2004; Mourtzas, 2012; Poulos, Ghionis, and Maroukian, 2009; Scicchitanoa et al., 2011).

If we consider sea-level curve by Kayan (1999), the level of the Marmara Sea that rose to the present level at 6 ka (McHugh et al., 2008) during formation of beachrock was about

Figure 5. Views from the fossil-rich near-shore deposit (a) and (b).

Figure 6. Cross-section of the studied site produced based on the integration of field observations with ERT-derived geophysical survey results.

1.5 m lower than the present, which is supported by the overlapped curves for the Aegean Sea (Poulos, Ghionis, and Maroukian, 2009), showing an agreement in 2-m lowering of the sea level during that period (see also Kambouroglou et al., 1988; Lambeck, 1996; Van Andel, 1990; Vouvalidis, Syrides, and Albanakis, 2005). Similar results were obtained by Desruelles et al. (2009) from Kemer beachrock on the Mediterranean coast of Turkey. Desruelles et al. (2009) demonstrated, based on submerged beachrock dated 2785 6 30 YBP, that sea level was between1.5 and 2.2 m during the sixth and seventh centuries BP. Other recent data obtained from beachrocks submerged at depths between1 and 1.5 m on the north coast of the antique city of Parion, SW Marmara Sea coast, suggested a lower sea level than that of the present sea level during the classical period (Erginal, 2012). Thus, it can be stated that beachrock beds on the studied coast occurred during a lowstand when sea level was about 1.5 m lower than at present and is in good agreement with the aforementioned previous data.

Possible Tsunamigenic Event

Results obtained from fossil mollusks and gastropods collected from the near-shore deposit, however, indicate that ages fall in the range between 2340 and 1590 years cal BP, pointing to a period spanning about 750 years. It is worth mentioning that this fossil-rich deposit yielded younger ages than submerged beachrock. Based on the fact that beachrock formed during a lower position of sea-level and then became submerged because of sea-level rise, this younger fossil-bearing sequence could either be attributed to a 2-m rise in sea level above the present level or ensuing tectonic uplift of the island. Besides the lack of sedimentological evidence in the studied sequence to infer such a consistent rise of sea level, there is no previous record with regard to such a Late Holocene sea-level highstand in the Marmara Sea. Furthermore, there is no tectonic evidence to link its position to a postdepositional uplift. Thus, the present level of this fossil-rich sequence is likely of different origin.

When the composition and depositional characteristics of the deposit are considered, the fossil-rich zone is indicative of deposition during a high-energy event. The fact that this layer with a thickness up to 0.75 m not only overlies an unconsolidated colluvial unit but is also overlain by similar weathered material comprises an unusual deposition dissim-ilar to normal sedimentation with transition levels. The underlying unsorted unit has an erosive base on green schists and contains angular rock fragments and unsorted sands and granules with poor internal arrangement. The fossil-rich unit

in question with similar grain size and internal characteristics is comprised largely of Cerithium vulgatum and lesser amounts of Gibbula albida, Patella vulgata, Pecten maximus, and Cardium edule, and, albeit in limited amounts, various benthic foraminifera such as Elphidium aculeatum (d’Or-bigny, 1846), Elphidium crispum (Linnaeus, 1758), Elphidium macellum (Fichtel and Moll, 1798), Elphidium sp., Rosalina sp., and Quinqueloculina spp.

CONCLUSIONS

In this paper, composition, depositional characteristics, AMS radiocarbon ages, and ERT-based subsurface nature of the coexisting beachrock and fossil-rich nearshore deposit that crop out on the coast of Hasır Island, SW Marmara Sea, are presented to discuss their implications from the point of sea-level changes and an ensuing high-energy event possibly of tsunamigenic origin. The submerged beachrock is made up of a mixture of algae oncoids, coral fragments, shell debris, different mineral fragments, and poorly rounded sands and lithoclasts derived from metamorphic rocks. Intertidal cements of beachrock precipitated in order of micrite coatings, radial aggregates of scalenohedral high-Mg calcites, and reticulated needles of aragonite are indicative solely of a marine-phreatic environment. Based on AMS 14_{C ages obtained from bulk}

carbonates and fossil residue of O. edulis yielded ages of 2940 and 2730 YBP for the oldest (lower) and 2730 and 2470 YBP for the youngest (upper) beds, suggesting that sea level was lower than that of the present during the precipitation of connective carbonates. From a depositional viewpoint, the 0.75-m-thick fossil-laden near-shore deposit behind the submerged beach-rock represents a high-energy event. The fact that it overlies an unconsolidated colluvial unit with unsorted sands and granules resting on erosive base on green schists and it is overlain by the same deposit with poor internal arrangement indicates the lack of normal stratigraphic transition either upward or downward. Furthermore, the abnormal and disordered accumulation of marine gastropods and shells within a sandy and muddy matrix including benthic forami-nifera is likely typical of a tsunamigenic event because this level does not show any stratigraphic transition either upward or downward. The AMS 14_{C ages fall in the range between}

2340 and 1590 years cal BP and are suggestive of the period for the proposed Late Holocene tsunamigenic event.

ACKNOWLEDGMENTS

We thank Dr. Aydın B ¨uy ¨uksara¸c for allowing us to use the resistivity meter for this research. We also thank Elmas

Kırcı-near-shore deposit F. glaber þ2-2.5 1810þ/- 30 1840 - 1590 1.1

C.glaucum þ2-2.5 2060þ/- 30 2150 - 1920 þ0.9

C. vulgatum þ2-2.5 2010þ/- 30 2120 - 1870 þ1.0

M. brandaris þ2-2.5 2150þ/- 30 2290 - 2000 0.7

Elmas for microfossil analyses and Graham Lee for improving the English of the text. Mustafa Bozcu helped with petro-graphic thin section analyses. We are also indebted to the editorial board and two anonymous referees for their constructive comments that have greatly improved our paper. This study was supported by the Research Foundation of Istanbul University (project number: 9105).

LITERATURE CITED

Ardel, A. and Kurter, A., 1973. Marmara Sea (Physical Etude). Istanbul University Bulletin of the Institute of Geography, 18–19, 57–70.

Beier, J.A., 1985. Diagenesis of Quaternary Bahamian beachrock: petrographic and isotopic evidence. Journal of Sedimentary Petrology, 55, 755–761.

Desruelles, S.; Fouache, E.; ¸Ciner, A.; Dalongeville, R.; Pavlopoulos, K.; Kosun, E.; Coquinot, Y., and Potdevin, J.L., 2009. Beachrocks and sea level changes since Mid-Holocene: comparison between the insular group of Mykonos-Delos-Rhenia (Cyclades, Greece) and southern coast of Turkey. Global and Planetary Change Letters, 66, 19–33.

Erginal, A.E., 2012. Beachrock as evidence of sea-level lowstand during the classical period, Parion antique city, Marmara Sea, Turkey. Geodinamica Acta, 25, 96–103.

Erginal, A.E.; Ekinci, Y.L.; Demirci, A.; Avcıo ˘glu, M.; ¨Ozt ¨urk, M.Z.; T ¨urkes, M., and Yi ˘gitbas, E., 2013a. Depositional characteristics of carbonate-cemented fossil eolian sand dunes: Bozcaada Island, Turkey. Journal of Coastal Research, 29, 78–85.

Erginal, A.E.; Ekinci, Y.L.; Demirci, A.; Bozcu, M.; ¨Ozt ¨urk, M.Z.; Avcıo ˘glu, M., and ¨Oztura, E., 2013b. First record of beachrock on Black Sea coast of Turkey: implications for Late Holocene sea-level fluctuations. Sedimentary Geology, 294, 294–302.

Erginal, A.E.; Ekinci, Y.L.; Demirci, A.; Kırcı-Elmas, E., and Kaya H., 2012. First note on Holocene Coquinite on Thrace (Black Sea) coast of Turkey. Sedimentary Geology, 267–268, 55–62.

Erginal, A.E.; Kiyak, N.G., and Ozturk, B., 2010. Investigation of beachrock using microanalyses and OSL dating: a case study from Bozcaada Island, Turkey. Journal of Coastal Research, 26, 350–358. Erginal, A.E.; Kıyak, N.G.; Ekinci, Y.L.; Demirci, A.; Ertek, A., and Timur, C., 2013c. Age, composition and paleoenvironmental significance of a Late Pleistocene eolianite from the western Black Sea coast of Turkey. Quaternary International, 296, 168–175. Holail, H. and Rashed, M., 1992. Stable isotopic composition of

carbonate-cemented recent beachrock along the Mediterranean and Red Sea coasts of Egypt. Marine Geology, 106, 141–148. Kambouroglou, E.; Maroukian, H., and Sampson, A., 1988. Coastal

evolution and archaeology north and south of Khalkis (Euboea) in the last 5000 years. In: Raban, A. (ed.), Archaeology of Coastal Changes. Oxford: BAR International Series, 404, pp. 71–79. Kayan, ˙I., 1999. Holocene stratigraphy and geomorphological

evolu-tion of the Aegean coastal plains of Anatolia. Quaternary Science Reviews, 18, 541–548.

Kindler, P. and Bain, R.J., 1993. Submerged upper Holocene beachrock on San Salvador Island, Bahamas: implications for recent sea-level history. Geologische Rundschau, 82, 241–247. Lambeck, K., 1996. Sea-level changes and shoreline evolution in

Aegean Greece since Upper Paleolithic time. Antiquity, 70, 588–611. Lambeck, K.; Anzidei, M.; Antonioli, F.; Benini, A., and Esposito, A., 2004. Sea level in roman time in the Central Mediterranean and

implications for recent change. Earth and Planetary Science Letters, 224, 463–575.

Loke, M.H. and Barker, R.D., 1996. Rapid least-squares inversion of apparent resistivity pseudosections using a quasi-Newton method. Geophysical Prospecting, 44, 131–152.

McHugh, C.M.G.; Gurung, D.; Giosan, L.; Ryan, W.B.F.; Mart, Y.; Sancar, U.; Burckle, L., and Cagatay, M.N., 2008. The last reconnection of the Marmara Sea (Turkey) to the World Ocean: a paleoceanographic and paleoclimatic perspective. Marine Geology, 255, 64–82.

Mourtzas, N.D., 2012. Fish tanks of eastern Crete (Greece) as indicators of the Roman sea level. Quaternary International, 250, 3–18.

Neumeier, U., 1999. Experimental modelling of beachrock cementa-tion under microbial influence. Sedimentary Geology, 126, 35–46. Poulos, S.E.; Ghionis, G., and Maroukian, H., 2009. Sea-level rise

trends in the Attico-Cycladic region (Aegean Sea) during the last 5000 years. Geomorphology, 107, 10–17.

Reimer, P.J.; Baillie, M.G.L.; Bard, E.; Bayliss, A.; Beck, J.W.; Blackwell, P.G.; Bronk Ramsey, C.; Buck, C.E.; Burr, G.S.; Edwards, R.L.; Friedrich, M.; Grootes, P.M.; Guilderson, T.P.; Hajdas, I.; Heaton, T.J.; Hogg, A.G.; Hughen, K.A.; Kaiser, K.F.; Kromer, B.; McCormac, F.G.; Manning, S.W.; Reimer, R.W.; Richards, D.A.; Southon, J.R.; Talamo, S.; Turney, C.S.M.; van der Plicht, J., and Weyhenmeyer, C.E., 2009. IntCal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon, 51, 1111–1150.

Sakın¸c, M. and Yaltırak, C., 1997. G ¨uney Trakya Sahillerinin Denizel Pleyistosen ¸C¨okelleri ve Paleoco ˘grafyası. Bulletin of the Mineral Research and Exploration of Turkey, 119, 43–62.

Sasaki, Y., 1992. Resolution of resistivity tomography inferred from numerical simulation. Geophysical Prospecting, 40, 453–464. Scicchitanoa, G.; Lo Presti, V.; Spampinato, C.R.; Gasparo Morticelli,

M.; Antonioli, F.; Auriemma, R.; Ferranti, L., and Monaco, C., 2011. Millstones as indicators of relative sea-level changes in northern Sicily and southern Calabria coastlines, Italy. Quaternary International, 232, 92–104.

Stefanon, A., 1971. Submerged beachrock in the Gulf of venice (Italy): key to the knowledge of the local coastline evolution in the last few thousand years. Quaternaria, 14, 191–193.

Stuiver, M. and Reimer, P.J., 1993. Extended 14C data base and revised CALIB 3.0 14C age calibration program. Radiocarbon, 35, 215–230.

Tun¸cdilek, N.; Kurter, A.; Sungur, K.A.; G¨ozen¸c, S.; Mater, B.; Narlı, F., and G ¨um ¨us, E., 1987. Marmara Takım Adaları (Bug ¨unk ¨u Arazi Kullanım Potansiyeli). Istanbul: Publication of the Istanbul University Institute of Marine Science and Geography, 146p. Van Andel, T.H., 1990. Addendum to ‘late quaternary sea level

changes and archaeology.’ Antiquity, 64, 151–152.

Vieira, M.M. and De Ros, L.F., 2007. Cementation patterns and genetic implications of Holocene beachrocks from northeastern Brazil. Sedimentary Geology, 192, 207–230.

Vouvalidis, K.G.; Syrides, G.E., and Albanakis, K.S., 2005. Holocene morphology of the Thessaloniki Bay: impact of sea level rise. Zeitschrift f ¨ur Geomorphologie N.F. Supplement, 137, 147–158. Yaltırak, C.; Sakın¸c, M.; Aksu, A.E.; Hiscott, R.N.; Galleb, B., and

Ulgen, U.B., 2002. Late Pleistocene uplift history along the southwestern Marmara Sea determined from raised coastal deposits and global sea-level variations. Marine Geology, 190, 283–305.