SEA SURFACE RADIOCARBON RESERVOIR AGE CHANGES IN THE AEGEAN SEA FROM ABOUT 11,200 BP TO PRESENT

SEA SURFACE RADIOCARBON RESERVOIR AGE CHANGES IN THE AEGEAN SEA FROM ABOUT 11,200 BP TO PRESENTYorgos FacorellisDepartment of Antiquities and Works of Art Conservation, Faculty of Fine Arts, Technological Educational Institute of Athens, Aghiou Spyridonos St., 12210 Egaleo, Athens, Greece. Corresponding author. Email: [email protected].

Evi Vardala-TheodorouHydrobiological Department, Goulandris Natural History Museum, 100 Othonos St., 14562 Kifissia, Athens, Greece. Email: [email protected].

ABSTRACT. Archaeological excavations in two coastal sites of Greece, Ftelia on Mykonos and Cyclops Cave on Youra, have provided suitable material (charcoal/marine mollusk shell paired samples deposited simultaneously in undisturbed anthropogenic layers) to estimate regional changes of the sea surface radiocarbon reservoir effect (ΔR) in the Aegean Sea. Moreover, pre-bomb 14C ages of marine mollusk shells of known collection date, from Piraeus and Nafplion in Greece and Smyrna in Turkey, also contributed to the marine reservoir calculation during recent years. In this article, these already published results, 10 in total, are considered and calibrated again using the latest issues of the calibration curves IntCal13 and Marine13. The same calibration data were applied to 11 more paired samples from the archaeological sites of Palamari on Skyros and Franchthi Cave in the Argolic Gulf, published here for the first time, in order to investigate the fluctuation of the reservoir ages R(t) and ΔR values in the Aegean Sea from ~11,200 BP (~13,000 cal BP) to present. Our data show that R(t) and ΔR values are not constant through time and may vary from 1220 ± 148 to –3 ± 53 yr and –451 ± 68 to 858 ± 154 14C yr, respectively. An attempt was also made to correlate these fluctuations with eastern Mediterranean paleoenvironmental proxies and other relevant paleoceanographic data found in the literature.

INTRODUCTION

It is well known that samples from species living contemporaneously in the atmosphere and sea surface waters (<100 m depth) show different conventional 14C ages. This difference between the 14C ages of such paired samples is called the marine reservoir age R(t), or apparent age. R(t) is not constant (t = cal age) due to the difference in sea surface reservoir and atmosphere 14C-specific ac-tivity changes caused by (1) oceanic circulation processes that tend to advect intermediate and deep 14C-depleted water masses to the surface, (2) atmospheric 14C production changes, and (3) air-sea CO2 exchange processes (Stuiver et al. 1986, 1998; Stuiver and Braziunas 1993; Siani et al. 2000, 2001).

A global box-diffusion carbon model (Oeschger et al. 1975), which reproduces depth-dependent marine 14C variations in response to atmospheric 14C production changes (Stuiver et al. 1986; Stu-iver and Braziunas 1993), is used in order to estimate the reservoir ages of the global mixed sea surface layer. These were found to be ~400 yr older in the well-ventilated gyres of the central North and South Pacific and the Atlantic Ocean (Bard 1988). Moreover, the magnitude of this effect may also differ from region to region due to e.g. long residence time of deep waters in upwelling areas, which is in turn affected by the coastline shape, climate, and sea bottom topography (Robinson and Thompson 1981; Domack 1992; Dye 1994; Heier-Nielsen et al. 1995; Berkman and Forman 1996) and the hardwater effect in marine organisms that have grown near estuaries (Ingram and Southon 1997; Goodfriend and Flessa 1997).

The local sea surface reservoir deviations (ΔR) from R(t) may be determined either by 14C dating of (1) live-collected pre-bomb marine mollusk shells (Siani et al. 2000; Reimer and McCormac 2002); (2) foraminifera or mollusk shells extracted from marine sediments and plant material from terrestri-al settings, such as peat profiles paired samples, assuming a coeval date of organism death for both sample types, based on volcanic ash onshore/offshore isochrones (Ascough et al. 2005); and (3) marine mollusk shell/charcoal paired samples, assumed to be deposited simultaneously, collected from undisturbed archaeological layers (Facorellis et al. 1998).

Radiocarbon, Vol 57, Nr 3, 2015, p 491–503 DOI: 10.2458/azu_rc.57.18363 © 2015 by the Arizona Board of Regents on behalf of the University of Arizona

Proceedings of the 1st International Radiocarbon in the Environment Conference 18–22 August 2014, Queen’s University Belfast, Belfast, Northern Ireland, UK Edited by Evelyn Keaveney and Paula Reimer

mailto:[email protected]

mailto:[email protected]

492 Y Facorellis & E Vardala-Theodorou

Consequently, calculation of ΔR allows (1) reliable calibration of the conventional 14C ages of ma-rine samples (e.g. marine mollusk shells), in the absence of other suitable material from an excavat-ed site (as they are usually short-lived samples, rarely transported over long distances), (2) reliable calibration of 14C ages of human bones in case of mixed (marine and terrestrial) diet (Sveinbjörns-dóttir et al. 2010), and (3) obtaining of paleoceanographic information (Siani et al. 2001).

The first ΔR value in the Mediterranean Sea corresponds to a pre-bomb marine mollusk shell sample collected in AD 1954 at the Kouali Point of Tipasa in Algeria (36°40′N, 2°30′E) (Olson and Bro-ecker 1959; Broecker and Olson 1961), which produced a ΔR value of –133 ± 83 14C yr (Stuiver et al. 1986). Facorellis et al. (1998) then published a mean R(t) of 515 ± 22 yr and a ΔR value of 149 ± 30 14C yr in the Aegean Sea for the time interval 8800–7400 BP (9900–8000 cal BP) based on three paired charcoal/marine mollusk shell samples collected from the Mesolithic anthropo-genic layers of Cyclops Cave on Youra Island. Siani et al. (2000) followed with the publication of 26 modern, pre-bomb marine mollusk shells live-collected between AD 1837 and 1950 throughout the Mediterranean Sea, which enabled estimation of the R(t)mean = 390 ± 85 yr and the ΔRmean = 35 ± 70 14C yr. They also calculated a pre-bomb R(t) value of 520 ± 40 yr in the Dardanelles Strait (collection year AD 1900), which is in a very good agreement with the several millennia older one from Youra Island. Siani et al. (2001) then published another seven tephra charcoal/planktonic for-aminifera paired samples from the South Adriatic Sea spanning from 16,130 to 3820 BP (19,790–4090 cal BP), which showed significant deviations from the Mediterranean R(t)mean. Additionally, Reimer and McCormac (2002) found that pre-bomb R(t) in at least two Aegean sites [Piraeus, collection year AD 1943, R(t) = 427 yr; and Nafplion, collection year AD 1940, R(t) = 324 yr] is statistically the same at a 95% level with the aforementioned values. A third pre-bomb marine mollusk shell from Smyrna, Turkey (collection year AD 1893/4) produced a much higher R(t) of 652 yr, which may be attributed to the estuarine environment that this marine mollusk species often inhabits (Reimer and McCormac 2002). Subsequently, Facorellis and Maniatis (2002), based on three paired charcoal/marine mollusk shell samples collected from the Middle Neolithic layers of the Ftelia site on Mykonos, calculated a regional mean R(t) of –32 ± 31 yr and a mean ΔR value of –393 ± 40 14C yr for the time interval 6080–5720 BP (7150–6400 cal BP), showing that during that time period there was no significant local sea surface reservoir effect.

All the aforementioned data, together with the Mediterranean Sea R(t) changes since the Last Gla-cial Maximum reported by Siani et al. (2001), indicate that R(t) is not constant through time. In or-der to estimate these changes in the Aegean Sea, we reconsidered all the available data found in the literature (Facorellis et al. 1998; Siani et al. 2000, 2001; Facorellis and Maniatis 2002; Reimer and McCormac 2002), as well as 11 new paired samples we dated from two other archaeological sites in that region, which are presented here for the first time. More specifically, two pairs from Palamari on Skyros and nine more from Franchthi Cave in the Argolic Gulf have been dated. This study aims to present all the up-to-date published regional R(t) values and recalculate the corresponding ΔR values using the latest IntCal13 and Marine13 calibration curves (Reimer et al. 2013), together with the 11 new R(t) and ΔR values from Palamari and Franchthi Cave. Finally, an attempt was made to correlate these values with published southeastern Mediterranean paleoproxies and regional paleoceanographic data in order to understand the underlying reasons for the variability observed.

MATERIALS AND METHODS

The new charcoal and marine mollusk shell samples were 14C dated in the laboratory of Archaeom-etry of NCSR “Demokritos” in Athens using the conventional gas counting technique, as described elsewhere (de Vries and Barendsen 1953; Münnich 1957; Kromer and Münnich 1992; Facorellis 1996; Facorellis et al. 1997; Maniatis and Papadopoulos 2011).

49314C Reservoir Age in the Aegean Sea, 11,200 BP to Present

The two charcoal samples from Palamari (DEM-1036 and DEM-1038) were thoroughly examined and manually cleaned under a stereoscope. The samples were then chemically treated using the standard acid-base-acid treatment to remove all carbon contaminants (Olsson 1979, 1986; Mook and Streurman 1983).

The 11 marine mollusk shells, two from Palamari (DEM-1074 and DEM-1076) and nine from Franchthi Cave (DEM-1048 to DEM-1056), were washed with deionized water in an ultrasonic bath and the depositions on the surface were removed with a dentist’s drill. Next, more than 35% of their remaining weight was removed with a solution of 2% HCl to eliminate any recrystallized carbonates during the time of burial (Goslar and Pazdur 1985; Facorellis et al. 1998). 14C ages of the nine marine mollusk shells samples from Franchthi Cave were paired with the corresponding published 14C ages of charcoal samples originating from the same undisturbed archaeological layers (W R Farrand, personal communication, 2000), which were dated in the University of Pennsylvania 14C laboratory (Lawn 1971, 1974, 1975; Fishman et al. 1977).

The shells of short-lived marine mollusks living in relatively shallow waters are ideal material to be used for the calculation of the sea surface regional reservoir effect R(t) and ΔR values, as they reflect the 14C content of their environment and they are directly related to their archaeological use. Such marine mollusk shells are appropriate only if they are food residues, originating from the same undisturbed anthropogenic layers with the paired charcoal samples. Table 1 summarizes all the available aforementioned information for the marine mollusk species used in this study, proving that they fulfill all the above requirements. From the table, one can see that most of the marine mollusks belong to short-lived (<20 yr), edible species living from just offshore in shallow waters and deeper, on the sea floor at depths less than 100 m (Shackleton and van Andel 1980; Forster 1981; Sabelli et al. 1990; Delamotte and Vardala-Theodorou 2001; Papathanassiou and Zenetos 2005; Papaconstan-tinou et al. 2007; Peharda et al. 2012).

Figure 1 shows the locations of the sites in the Aegean Sea region involved in this study. More spe-cifically, the samples originate from undisturbed anthropogenic layers of the following:

1. The Bronze Age site Palamari on Skyros (38°58′N, 24°30′E) [2 pairs] (Parlama 2007, 2009; Parlama et al. 2010; Pavlopoulos et al. 2010);

2. Neolithic, Mesolithic and Paleolithic layers of Franchthi Cave in Porto Cheli, Argolic Gulf (37°25′20.90″N, 23°7′52.73″E) [9 pairs] (Lawn 1971, 1974, 1975; Fishman et al. 1977; Jacob-sen 1981; Jacobsen and Farrand 1987; Farrand 2000, 2003; Perlès 2001, 2003; Facorellis 2003, 2011, 2013);

3. The Neolithic site Ftelia on Mykonos (37°27′N, 25°22′E) [3 pairs] (Sampson 2002; Facorellis and Maniatis 2002);

4. The Neolithic and Mesolithic layers of Cyclops Cave on Youra (37°27′N, 24°10′E) [4 pairs] (Facorellis et al. 1998; Facorellis 2003, 2013; Sampson 2008, 2011); and

5. Three pre-bomb 14C ages of marine mollusk shells from Piraeus (37°56′N, 23°38′E), Nafplion (37°32′N, 22°47′E), and Smyrna, Turkey (38°26′N, 27°9′E) (Reimer and McCormac 2002).

It is worth noting that the successive cultural phases in the Helladic area are roughly correlated to the Paleolithic, which ends at ~10,000 BC, the Mesolithic (10,000 to ~6500 BC), the Neolithic (~6500 to 3200/3000 BC), and the Bronze Age (3200/3000 to 1050/1025 BC).


Tabl

e 1

Sum

mar

y of

the

mar

ine

mol

lusk

shel

l spe

cies

info

rmat

ion

used

in th

is st

udy

for t

he e

stim

atio

n of

the

sea

surf

ace

rese

rvoi

r in

the

Aeg

ean

Sea.

Shel

l spe

cies

Life

span

Con

sum

ptio

n in

tere

stLi

ving

zon

eAr

ca n

oae

Linn

aeus

, 175

8 U

p to

16

yr, w

ith m

ax le

ngth

7 c

mEd

ible

; in

som

e lo

catio

ns th

e sp

ecie

s is c

omm

erci

ally

im

porta

nt a

s sea

food

reso

urce

; maj

or in

tere

stIn

terti

dal z

one,

just

off

-sh

ore

to d

epth

s of 1

00 m

Ac

anth

ocar

dia

tube

rcul

ata

Linn

ae-

us, 1

758

Age

s ran

ge fr

om 5

mon

ths t

o 11

yr

(Peh

arda

et a

l. 20

12)

Edib

le; C

roat

ian

cuis

ine

Inte

rtida

l zon

e to

de

pths

of 1

00 m

, dis

tribu

t-ed

in ~

1–40

mBa

rbat

ia b

arba

ta L

inna

eus,

1758

Edib

le; w

idel

y co

llect

ed fo

r hum

an c

onsu

mpt

ion;

m

ajor

inte

rest

Cal

lista

chi

one

Linn

aeus

, 175

8>4

0 yr

(For

ster

198

1), s

low

gro

w-

ing,

6 c

m in

15–

17 y

r (Fo

rste

r 198

1;

Papa

cons

tant

inou

et a

l. 20

07)

Edib

le; e

xcel

lent

seaf

ood;

maj

or in

tere

stFr

om ju

st o

ffsh

ore

to

130

m d

epth

Cer

asto

derm

a gl

aucu

m P

oire

t, 17

89Ed

ible

; col

lect

ed lo

cally

in N

esto

s and

Vis

toni

s la

goon

s for

hum

an c

onsu

mpt

ion

Upp

er su

blitt

oral

zon

e tra

nsiti

onal

eco

syst

em

lago

ons,

estu

arie

s 1–4

0 m

C

eras

tode

rma

edul

e Li

nnae

us,

1758

10–1

6 yr

(Peh

arda

et a

l. 20

12)

Edib

le; F

renc

h: c

oque

s; E

nglis

h: e

dibl

e co

ckle

or

com

mon

Eur

opea

n co

ckle

Upp

er su

blitt

oral

zon

e 1–

20 m

Myt

ilus e

dulis

Lin

naeu

s, 17

58

Myt

ilus g

allo

prov

inci

alis

Lm

ck,

1819

Ed

ible

; exc

elle

nt se

afoo

d; m

ajor

inte

rest

Inte

rtida

l 0.5

m to

40

m

dept

h

Ost

rea

edul

is L

inna

eus,

1758

grow

ver

y ol

d >2

0 yr

Edib

le; e

xcel

lent

seaf

ood;

maj

or in

tere

st1–

150

m su

blitt

oral

to

circ

alitt

oral

zon

e Pi

nna

nobi

lis L

inna

eus,

1758

20

+ yr

(Pap

acon

stan

tinou

et a

l. 20

07)

Edib

le; m

inor

com

mer

cial

inte

rest

Mid

litto

ral t

o su

blitt

oral

an

d m

ore

from

0.5

–60

m

Gas

trop

oda

Cer

ithiu

m v

ulga

tum

Bru

guiè

re,

1792

Ined

ible

, use

d as

fish

ing

bait

Mid

litto

ral t

o su

blitt

oral

1–

40 m

C

olum

bella

rust

ica

Linn

aeus

, 175

8In

edib

leM

idlit

tora

l to

subl

ittor

al

1–40

mC

onus

med

iterr

aneu

s HA

WSS

IN

BR

UG

UIR

E, 1

792

Ined

ible

Mid

litto

ral t

o su

blitt

oral

1–

40 m

Dio

dora

ital

ica

Def

ranc

e, 1

820

Edib

leM

idlit

tora

l to

subl

ittor

al

1–40

m


Tabl

e 1

Sum

mar

y of

the

mar

ine

mol

lusk

shel

l spe

cies

info

rmat

ion

used

in th

is st

udy

for t

he e

stim

atio

n of

the

sea

surf

ace

rese

rvoi

r in

the

Aeg

ean

Sea.

Shel

l spe

cies

Life

span

Con

sum

ptio

n in

tere

stLi

ving

zon

eH

exap

lex

trun

culu

s Lin

naeu

s, 17

58Ed

ible

; exc

elle

nt se

afoo

d; m

ajor

inte

rest

, fish

ing

bait

Subt

idal

to su

blitt

oral

5–

30 m

Pate

lla c

aeru

lea

Linn

aeus

, 175

8Ed

ible

; lim

pets

are

wid

ely

colle

cted

for h

uman

co

nsum

ptio

n an

d as

fish

ing

bait;

min

or c

omm

erci

al

inte

rest

Low

er M

idlit

tora

l

Pate

lla u

lyss

ipon

ensi

s Gm

elin

in

L., 1

791

Edib

le; l

impe

ts a

re w

idel

y co

llect

ed fo

r hum

an

cons

umpt

ion

and

as fi

shin

g ba

it; m

inor

com

mer

cial

in

tere

stM

idlit

tora

l

Figu

re 1

M

ap o

f th

e A

egea

n Se

a in

dica

ting

the

geog

raph

ic

loca

tion

of th

e st

udy

site

s (m

odifi

ed fr

om G

oogl

e Ea

rth®)


RESULTS AND DISCUSSION

The Appendix (online supplemental file) presents all the dating results sorted by age. A total of 21 pairs of samples are shown, including the three single pre-bomb marine mollusk samples of known collection date. The conventional 14C ages of the samples were calculated using δ13C values of –25‰ for charcoal and 1‰ for marine mollusk shells (Stuiver and Polach 1977), which are the typical values used for that type of material in the Aegean region (Facorellis et al. 1998). However, these δ13C values may have a possible ±2‰ error due to 13C variation (Stuiver and Polach 1977), which corresponds to an additional ±32 yr error in the calculated 14C ages. This was incorporated into the final 14C age error of the new samples by using the following formula for summing errors:

4

4) the Neolithic and Mesolithic layers of Cyclops cave on Youra (37° 27΄ N, 22´ N, 24° 10´ E) [4 pairs] (Sampson 2008, 2011; Facorellis et al. 1998; Facorellis 2003, 2013) and 5) three pre-bomb radiocarbon ages of marine mollusk shells from Piraeus (37° 56΄ N, 23° 38´ E), Nafplion (37° 32΄ N, 22° 47´ E) and Smyrna, Turkey (38° 26΄ N, 27° 9´ E) (Reimer and McCormac 2002). It is worth noticing that the successive cultural phases in the Helladic area are roughly defined in a consensual manner to the Palaeolithic, which ends at ca. 10,000 BC, the Mesolithic (10,000 to ca. 6,500 BC), the Neolithic (ca. 6,500 to 3,200/3,000 BC) and the Bronze Age period (3,200/3,000 to 1,050/1,025 BC). RESULTS AND DISCUSSION Table 2 presents all the dating results sorted by age. A total of 21 pairs of samples are shown including the three single pre-bomb marine mollusk samples of known collection date. The conventional 14C ages of the samples were calculated using δ13C values of -25‰ for charcoal and 1‰ for marine mollusk shells (Stuiver and Polach 1977), which are the typical values used for that type of material in the Aegean region (Facorellis et al. 1998). However, these δ13C values may have a possible ±2‰ error due to 13C variation (Stuiver and Polach 1977), which corresponds to an additional ±32 yr error in the calculated 14C ages. This was incorporated into the final 14C age error of the new samples by using the following formula for summing errors:

σ!"#$% = σ!"!! + σ!"#!

Then the 14C ages of the charcoal samples were calibrated with the latest issue IntCal13 of the international calibration curve (Reimer et al. 2013). The 14C ages of the marine mollusk shell samples were calibrated using the calculated ΔR value corresponding for each pair (Table 2) in conjunction with the Marine13 international calibration curve. All calibrations were performed with the OxCal v4.2.4 software (Bronk Ramsey 2009, 2010). All calibrated dates are rounded to the closest decade. The regional sea surface reservoir R(t) value of each pair is calculated by subtracting the 14C conventional age of the charcoal sample from that of the marine mollusk sample. The R(t) error is calculated by using a similar formula for summing errors as mentioned above. The local deviation ΔR values are calculated by combining the calibrated ages of the terrestrial and marine sample, each time using the latest issues of the atmospheric and the marine international curves (Reimer et al. 2013) respectively (Southon et al. 1995; Facorellis et al. 1998) as follows: the measured 14C age of the terrestrial sample gives a calendar date using the atmospheric calibration curve. Based on this calendar date, the marine calibration curve gives an equivalent 14C age. ΔR is the difference between this equivalent and the measured 14C age for the marine sample. In this study, this is performed with the OxCal v4.2.4 software using the following routine, as adapted for the Palamari-1 paired samples (Table 2, Figure 2). For each pair of samples the uncertainty U(x1,x2) was set accordingly, so that the ΔR probability distribution is at the center of the plot. Plot() { Curve("IntCal13","IntCal13.14c");

(1)

The 14C ages of the charcoal samples were then calibrated with the IntCal13 calibration curve (Re-imer et al. 2013). The 14C ages of the marine mollusk shell samples were calibrated using the cal-culated ΔR value corresponding for each pair (Appendix) in conjunction with the Marine13 inter-national calibration curve. All calibrations were performed with the OxCal v 4.2.4 software (Bronk Ramsey 2009). All calibrated dates are rounded to the nearest decade.

The regional sea surface reservoir R(t) value of each pair is calculated by subtracting the 14C con-ventional age of the charcoal sample from that of the marine mollusk sample. The R(t) error is calculated using a similar formula for summing errors as mentioned above.

The local deviation ΔR values are calculated by combining the calibrated ages of the terrestrial and marine samples, using the latest issues of the atmospheric and marine international curves (Reimer et al. 2013), respectively, as follows (Southon et al. 1995; Facorellis et al. 1998). The measured 14C age of the terrestrial sample gives a calendar date using the atmospheric calibration curve. Based on this calendar date, the marine calibration curve gives an equivalent 14C age. ΔR is the difference between this equivalent and the measured 14C age for the marine sample. In this study, this is per-formed with the OxCal v 4.2.4 software using the following routine, as adapted for the Palamari-1 paired samples (Appendix, Figure 2). For each pair of samples, the uncertainty U(x1,x2) was set accordingly, so that the ΔR probability distribution is at the center of the plot.

Plot() { Curve(“IntCal13”,”IntCal13.14c”); Curve(“Marine13”,”Marine13.14c”); Delta_R(“PALAMARI-1”,U(0,600)); Combine() { Curve(“=IntCal13”); R_Date(“DEM-1036”, 3545, 41); Curve(“=PALAMARI-1”); R_Date(“DEM-1076”, 4161, 51); }; };

https://journals.uair.arizona.edu/index.php/radiocarbon/rt/suppFiles/18363/0




Figure 3 shows the plot of R(t) values in the Aegean Sea versus 14C age from ~11,200 yr BP until present (black dash-dotted line), with respect to the pre-bomb mean R(t) value in the Mediterranean Sea [R(t)mean = 390 ± 85 yr, area between the red dotted lines], as estimated by Siani et al. (2000). This figure also includes, for comparison purposes, data (spanning from ~10,400 to 16,000 BP) from the South Adriatic Sea (Siani et al. 2001, supplementary online material). One can see that the R(t) values fluctuate significantly from 1220 ± 148 (Franchthi-9) to –3 ± 53 yr (Ftelia-2), resulting in regional ΔR values ranging from –451 ± 68 (Ftelia-3) to 858 ± 154 14C yr (Franchthi-9). Of the 21 pairs (Appendix), 13 pairs (Piraeus, Nafplion, Palamari-1 and -2, Franchthi-1, -2, -3, -6 and -7, Youra-1, -2, -3 and -4) produced R(t) values falling within two standard deviations (2σ, probability 95.4%) in the range of the Mediterranean pre-bomb mean value.

The prevailing question concerning our data is whether these sea surface reservoir ages R(t) are valid for all over the Aegean Sea within the same time period. Until further research adds new ev-idence, the black dash-dotted line connecting the points in Figure 3 is only tentative. By carefully looking at our results, one can see that there is an intrasite repeatability, overlapping within one standard deviation (1σ, probability 68.3%), in the following cases:

1. Ftelia site on Mykonos, three pairs (Ftelia-1, -2 and -3): mean R(t) = 26 ± 96 yr and mean ΔR = –400 ± 113 14C yr for the time interval 7150–6400 cal BP.

2. Franchthi Cave, two pairs (Franchthi-2 and -3): mean R(t) = 357 ± 120 yr and mean ΔR = –34 ± 123 14C yr and for two more pairs (Franchthi-4 and Franchthi-5): mean R(t) = 765 ± 161 yr and mean ΔR = 461 ± 140 14C yr for the time intervals 7690–7440 and 9140–8520 cal BP, respec-tively.

3. Cyclops Cave on Youra, three pairs (Youra-1, -2, and -3): mean R(t) = 529 ± 90 yr and mean ΔR = 167 ± 116 14C yr for the time interval 9530–8050 cal BP.

Figure 2 Example of probability distribution plot for the Palamari-1 pair ΔR value (Appendix)




Intraregional repeatability is observed between the Franchthi-6 (10,000–9540 cal BP) and Youra-4 (9910–9700 cal BP) pairs, which happen to be the only pairs of our study that chronologically overlap: mean R(t) = 267 ± 88 yr and mean ΔR = –37 ± 99 14C yr. Although this specific sea surface reservoir effect is now confirmed by the measurement of two independent pairs from two different sites, one mainland (Franchthi Cave) and one island (Cyclops Cave on Youra), the Youra-4 pair was initially rejected by Facorellis et al. (1998), as it was considered an outlier at the time due to lack of confirming evidence.

There are two sites that produced outlying R(t) values, Franchthi Cave in the Argolic Gulf (Franchthi-4, -5, -8, and -9 pairs) and Ftelia on Mykonos Island (Ftelia-1, -2, and -3 pairs). The R(t) values of the Franchthi-8 and -9 pairs and those of Franchthi-4 and -5 are higher than the Mediterranean R(t)mean value by a factor of 3 and 2, respectively (Appendix, Figure 3). On the other hand, the Ftelia pairs R(t) values are lower by a factor of 1.

There is only one data set of paired samples from the eastern Mediterranean Sea found in the litera-ture (Siani et al. 2001, supplemental online data) that coincides chronologically with our data. More specifically, this is the charcoal/planktonic foraminifera pair (OCF97129PP & R-704/GIFA96208), whose charcoal (from the Agnano Pomici Principali eruption) age (10,450 ± 40 BP) overlaps within 2σ with the charcoal age (10,260 ± 110 BP) of the Franchthi-8 pair (Figure 3, dashed rectangle).

Figure 3 Changes of the sea surface reservoir age R(t) in the Aegean Sea versus 14C age (yr BP) (black dash-dotted line), with respect to the pre-bomb R(t) in the Mediterranean Sea calculated by Siani et al. (2000) (within 2σ, area between the red dotted lines), in comparison with the sea surface temperature fluctuation in the South Adriatic Sea (modified from Siani et al. 2000). Data from the South Adriatic Sea are from Siani et al. (2001, supplemental online material).



Their corresponding R(t) values are 380 ± 100 and 988 ± 120 yr, respectively. This important R(t) difference shows that there is a different prevailing mechanism providing 14C-depleted water in the Argolic Gulf, than that of the Mediterranean Sea.

The subsurface hydraulic communication between the Argolic plain (Argon field) and the Argolic Gulf was known since ancient times. Pausanias (Arcadica, §7–8) mentions that the word argon means “slow cultivation”; thus, “Argon field” means a field of slow cultivation. This is due to the occasional flooding during wintertime, caused by water coming from the surrounding mountains and the karstic springs flowing from the margins, consisting mainly of multifolded cretaceous lime-stones (Pindos geotectonic unit). The Argon field would have probably turned into a lake if the wa-ter had not been drained through a ground rift. In this case, the rift is apparently the cave-sinkhole of Nestani, which is located at the lowest topographic point of the whole field. After a few days (based on modern tracing experiments), the water reappears from this sinkhole in a submarine system of karstic springs in the Argolic Gulf, that in ancient times were known as Dini, located near Kiveri village (Mariolakos and Mariolakos 2004). In addition, there are also three rivers, named Tanos, Kephalari (Erasinos), and Inachos, emanating from the mountains Lirkio and Trachi that flow into the Argolic Gulf. Consequently, the aforementioned sources of freshwater near Franchthi Cave indicate that the significantly higher R(t) values of the regional sea surface reservoir compared to Mediterranean mean, may be a phenomenon caused by a combination of a predominant local hard-water effect and coastal retreat events during deglaciation (Shackleton and van Andel 1980; van Andel and Lianos 1983; Lambeck 1996). The higher R(t) value of the Franchthi-9 pair, of 1220 ± 148 yr, is observed during the Younger Dryas stadial (11,240 ± 140 BP or 13,380–12,790 cal BP).

Mykonos is an island of the Cyclades located almost at the center of the Aegean Sea. The three pairs from the Ftelia site gave R(t) values that are overlapped within 1 standard deviation, thus leaving no room for doubt about their validity. However, the calculated mean R(t) value (26 ± 96 yr) is consid-erably lower than the Mediterranean mean. The low regional R(t) value could only be explained by reduced mixing rates of the surface waters with 14C-depleted waters from greater depths, during the time period studied (about 6100–5700 BP or 7150–6400 cal BP).

In an attempt to propose a mechanism involved, we discuss the following environmental events and paleoproxies: (1) the sea surface temperature fluctuation and (2) the sapropel S1 formation in the Aegean Sea.

(1) Sea Surface Temperature Fluctuation

The Aegean Intermediate Water is derived from the Levantine Intermediate Water, which travels north along the Turkish coast. Prevailing offshore winds allow upwelling of the intermediate water to the surface, thus forming a single uniform water mass from the surface to the seafloor (Lascaratos 1989; Yüce 1995). As the upwelled Aegean Intermediate Water moves northward, its salinity con-tinues to increase due to evaporation and this together with winter cooling increases its density. This buoyancy loss drives the formation of Aegean Deep Water. The modern Aegean Sea is an important source of deep water for the eastern Mediterranean (Casford et al. 2002). Several Holocene cool-ing events reconstructed for the southern Adriatic Sea, lasting some hundred years, are apparently synchronous with those in the Aegean Sea, in which sea surface temperatures declined by approx-imately 2°C around 6000 and 3000 yr BP (Sangiorgi et al. 2003). Figure 3 shows the R(t) changes in the Aegean Sea with respect to the sea surface temperature fluctuation in the South Adriatic Sea (modified from Siani et al. 2001). Our data show that the 6000 yr BP cooling event coincides with an important decrease of R(t) values of the three paired samples from the Neolithic site Ftelia on Mykonos. However, cooling of the surface waters is expected to increase their density and make


them move deeper into the sea, causing high mixing rates with 14C-depleted bottom waters, thus resulting in rather higher than lower R(t) values, as in the case of the Ftelia pairs.

(2) Sapropel S1 Formation in the Aegean Sea

Sapropels develop during episodes of reduced oxygen availability and increase the nutrient supply in bottom waters as a result of reduced deep seawater circulation. The most recent sapropel S1 in the eastern Mediterranean Sea formed from 9500 until 6000 yr BP (Mercone et al. 2000; Casford et al. 2002). The low R(t) values of the Ftelia pairs (Appendix, Figure 3) were likely caused by the low mixing rates of the sea surface waters due to the reduced deep-sea circulation of 14C-depleted waters during the sapropel S1 formation.

CONCLUSIONS

Three live-taken pre-bomb marine mollusk shells from Piraeus and Nafplion in Greece and Smyr-na in Turkey, as well as 18 charcoal/marine mollusk shells pairs from the archaeological sites of Franchthi Cave in the Argolic Gulf, Cyclops Cave on Youra, Ftelia on Mykonos, and Palamari on Skyros allowed calculation of the sea surface reservoir age changes R(t) and regional deviations ΔR in the Aegean Sea from ~11,200 BP (~13,000 cal BP) to present. Our data show that the R(t) and ΔR values in the Aegean Sea are not constant through time and may vary from 1220 ± 148 to –3 ± 53 yr and –451 ± 68 to 858 ± 154 14C yr, respectively.

Thirteen pairs produced R(t) values falling within 2σ of the Mediterranean pre-bomb mean value R(t)mean = 390 ± 85 yr. Two sites produced outlying R(t) values, Franchthi Cave and Ftelia. The mean R(t) and ΔR values for the Ftelia site is R(t) = 26 ± 96 yr and ΔR = –400 ± 113 14C yr for the time interval 7150–6400 cal BP. This very low R(t) value is probably due to the reduced deep-sea circulation of 14C-depleted waters during the sapropel S1 formation. The Franchthi Cave mean R(t) and ΔR values exhibit significant fluctuation probably due to the hardwater effect in the Argolic Gulf. During the time intervals 9140–8520 and 7690–7440 cal BP, the following mean values were calculated, respectively: R(t) = 765 ± 161 yr and ΔR = 461 ± 140 14C yr and R(t) = 357 ± 120 yr and ΔR = –34 ± 123 14C yr. The Cyclops Cave paired samples resulted in mean R(t) = 529 ± 90 yr and ΔR = 167 ± 116 14C yr for the time interval 9530–8050 cal BP. Combined data from Franchthi and Cyclops Cave produced mean R(t) = 267 ± 88 yr and ΔR = –37 ± 99 14C yr for the time interval 10,000–9540 cal BP.

The final goal of this ongoing project is to establish a regional detailed curve for short-lived marine mollusk shells from various coastal sites in the Aegean covering all the time range of the 14C dating method. This will allow accurate calibration of marine samples from that region in conjunction with the marine international calibration curve, and to obtain more paleoceanographic information for the region and time periods studied.

ACKNOWLEDGMENTS

The authors would like to thank Dr Yannis Maniatis for allowing the use of the 14C dating facility of the Laboratory of Archaeometry, NCSR “Demokritos” in Athens. Many thanks are also due to the archaeologists: Prof Adamantios Sampson, University of the Aegean, director of the Cyclops Cave excavation on Youra and Ftelia site on Mykonos; Prof Karen D Vitelli, Indiana University, director of the Franchthi Cave excavation; Drs Liana Parlama and Maria Theochari of the Greek Ministry of Culture, co-directors of the excavation at Palamari on Skyros, for providing the necessary sam-ples and archaeological information. We also thank the late Prof William R Farrand, University of Michigan, who indicated the most prominent undisturbed anthropogenic layers of Franchthi Cave



from which it was advisable to date marine mollusk shells, and Prof Catherine Perlès, University Paris Ouest Nanterre, for helping to locate them at Leonardo’s warehouse of the D’Ephoreia of Prehistoric and Classical Antiquities in Nafplion. We also thank Dr Brian Damiata, research asso-ciate at UCLA; Prof Christopher Bronk Ramsey, Oxford University; Prof Paula Reimer, Queen’s University of Belfast; and Dr Stella Katsarou-Tzouveleki, archaeologist at the Greek Ministry of Culture for helpful comments and discussion in the field of their expertise; as well as ΙΚΥ (Greek State Scholarships Foundation) and INSTAP (Institute for Aegean Prehistory) for funding.

REFERENCESAscough P, Cook G, Dugmore A. 2005. Methodological

approaches to determining the marine radiocarbon reservoir effect. Progress in Physical Geography 29(4):532–47.

Bard E. 1988. Correction of AMS 14C ages measured in planktonic foraminifera: paleoceanographic impli-cations. Paleoceanography 3(6):635–45.

Berkman PA, Forman SL. 1996. Pre-bomb radiocarbon and reservoir correction for calcareous marine spe-cies in the Southern Ocean. Geophysical Research Letters 23(4):363–6.

Broecker WS, Olson EA. 1961. Lamont radiocarbon measurements VII. Radiocarbon 3:176–204.

Bronk Ramsey C. 2009. Bayesian analysis of radiocar-bon dates. Radiocarbon 51(1):337–60.

Casford JSL, Rohling EJ, Abu-Zied R, Cooke S, Fontain-er C, Leng M, Lykousis V. 2002. Circulation chang-es and nutrient concentrations in the late Quaternary Aegean Sea: a nonsteady state concept for sapropel formation. Paleoceanography 17(2):1024–34.

de Vries H, Barendsen GW. 1953. Radiocarbon dating by a proportional counter filled with carbon dioxide. Physica 19:987–1003.

Delamotte M, Vardala-Theodorou E. 2001. Shells from the Greek Seas. Athens: The Goulandris Natural History Museum.

Domack EW. 1992. Modern carbon-14 ages and res-ervoir corrections for the Antarctic Peninsula and Gerlache Strait area. Antarctic Journal of the United States 27:63–4.

Dye T. 1994. Apparent ages of marine shells: implica-tions for archaeological dating in Hawai‘i. Radio-carbon 36(1):51–7.

Facorellis Y. 1996. Study of the conditions and the pa-rameters for high precision dating with 14C [PhD thesis]. University of Patras, Department of Physics, Greece. ISBN: 960-90516-0-X.

Facorellis Y. 2003. Radiocarbon dating the Greek Me-solithic. In: Galanidou N, Perlès C, editors. Pro-ceedings of the Round Table Meeting on “The Greek Mesolithic: Problems and Perspectives.” London: British School at Athens Studies Series 10. p 51–67.

Facorellis Y. 2011. Sequential radiocarbon dating and calculation of the marine reservoir effect. In: The Cave of the Cyclops: Mesolithic and Neolithic Net-works in the Northern Aegean, Greece. Volume II: Bone Tool Industries, Dietary Resources and the Paleoenvironment, and Archeometrical Studies. Prehistory Monographs 31. Philadelphia: INSTAP

Academic Press. p 361–72.Facorellis Y. 2013. Radiocarbon dates from archaeo-

logical sites in caves and rockshelters in Greece. In: Mavridis F, Jensen J, editors. Stable Places and Changing Perceptions: Cave Archaeology in Greece and Adjacent Areas. BAR International Series 2558 Oxford: Archaeopress. p 19–72.

Facorellis Y, Maniatis Y. 2002. Radiocarbon dating of the Neolithic settlement of Ftelia on Mykonos—cal-culation of the marine reservoir effect in the Aegean Sea. In: Sampson A, editor. The Neolithic Settlement at Ftelia, Mykonos. Rhodes: University of the Ae-gean. p 309–15.

Facorellis Y, Maniatis Y, Kromer B. 1997. Study of the parameters affecting the corre1ation of background versus cosmic radiation in CO2 counters: reliability of dating results. Radiocarbon 39(3):225–38.

Facorellis Y, Maniatis Y, Kromer B. 1998. Apparent 14C ages of marine mollusks shells from a Greek is-land—calculation of the marine reservoir effect in the Aegean Sea. Radiocarbon 40(2):963–74.

Farrand WR. 2000. Depositional history of Franchthi Cave-sediments, stratigraphy, and chronology. Ex-cavations at Franchthi Cave, Greece, Fascicle 12. Bloomington: Indiana University Press.

Farrand WR. 2003. Depositional environments and site formation during the Mesolithic occupations of Franchthi Cave, Peloponnesos, Greece. In: Gala-nidou N, Perlès C, editors. The Greek Mesolithic: Problems and Perspectives. London: British School at Athens Studies Series 10. p 69–78.

Fishman B, Forbes H, Lawn B. 1977. University of Pennsylvania radiocarbon dates XIX. Radiocarbon 19(2):188–228.

Forster GR. 1981. The age and growth of Callista chi-one. Journal of the Marine Biological Association of the UK 61:881–3.

Goodfriend GA, Flessa KW. 1997. Radiocarbon reser-voir ages in the Gulf of California: roles of upwell-ing and flow from the Colorado River. Radiocarbon 39(2):139–48.

Goslar T, Pazdur MF. 1985. Contamination studies on mollusk shell samples. Radiocarbon 27(1):33–42.

Heier-Nielsen S, Heinemeier J, Nielsen HL, Rud N. 1995. Recent reservoir ages for Danish fjords and marine waters. Radiocarbon 37(3):875–82.

Ingram BL, Southon JR. 1997. Reservoir ages in eastern Pacific coastal and estuarine waters. Radiocarbon 38(3):573–82.


Jacobsen TW. 1981. Franchthi Cave and the begin-ning of settled village life in Greece. Hesperia 50(4):303–19.

Jacobsen TW, Farrand WR. 1987. Franchthi Cave and Paralia: Maps, Plans and Sections. Excavations at Franchthi Cave, Greece 1. Bloomington: Indiana University Press.

Kromer B, Münnich KO. 1992. CO2 gas proportional counting in radiocarbon dating - review and per-spective. In: Taylor RE, Long A, Kra RS, editors. Radiocarbon after Four Decades: An Interdisci-plinary Perspective. New York: Springer-Verlag. p 184–97.

Lambeck K. 1996. Sea-level change and shoreline evo-lution in Aegean Greece since Upper Palaeolithic time. Antiquity 70(269):588–611.

Lascaratos A. 1989. Hydrology of the Aegean Sea. In Charnock H, editor. Winds and Currents of the Mediterranean Basin. Reports in Meteorology and Oceanography 40. Cambridge: Harvard University Press. p 313–34.

Lawn B. 1971. University of Pennsylvania radiocarbon dates XIV. Radiocarbon 13(2):363–77.

Lawn B. 1974. University of Pennsylvania radiocarbon dates XVII. Radiocarbon 16(2):219–37.

Lawn B. 1975. University of Pennsylvania radiocarbon dates XVIII. Radiocarbon 17(2):196–215.

Maniatis Y, Papadopoulos S. 2011. 14C dating of a final Neolithic-Early Bronze age transition period settle-ment at Aghios Ioannis on Thassos (North Aegean). Radiocarbon 53(1):21–37.

Mariolakos ID, Mariolakos DI. 2004. The Argon Field in Arcadia, the sinkhole of Nestani village, god Pose-idon and the submarine Dini Springs in the Argolic Gulf (Peloponnisos, Greece): a geomythological ap-proach of the Poseidon’s birth. Bulletin of the Geo-logical Society of Greece 36:1146–53.

Mercone D, Thomson J, Croudace IW, Siani G, Paterne M, Troelstra S. 2000. Duration of S1, the most re-cent sapropel in the eastern Mediterranean Sea, as indicated by accelerator mass spectrometry radio-carbon and geochemical evidence. Paleoceanogra-phy 15(3):336–47.

Mook WG, Streurman HJ. 1983. Physical and chemical aspects of radiocarbon dating. In: Proceedings of the First International Symposium on 14C and Archaeol-ogy. Groningen 1981. PACT. p 31–55.

Münnich KO. 1957. Messung naturlichen radiokohlen-stoffs mit einem CO2 proportional zahlrohr. Einige anwendungen der methode. Inaugural-disserta-tion zur erlangung der doktorwurde hohen natur-wissenschaftlich-mathematischen fakultat der Ruprecht-Karl-Universitat zu Heidelberg, Germany.

Oeschger H, Siegenthaler U, Schotterer U, Gugelmann A. 1975. A box-diffusion model to study the carbon dioxide exchange in nature. Tellus 27(2):168–92.

Olson EA, Broecker WS. 1959. Lamont natural radio-carbon measurements V. American Journal of Sci-ence Radiocarbon Supplement 1:1–28.

Olsson IU. 1979. The importance of the pre-treatment of

wood and charcoal samples in radiocarbon dating. In: Berger R, Suess HE, editors. In: Proceedings of the 9th International Radiocarbon Conference. Berkeley: University California Press. p 135–46.

Olsson IU. 1986. Radiometric methods. In: Berglund B, editor. Handbook of Holocene Palaeoecology and Palaeohydrology. Chichester: John Wiley & Sons. p 273–312.

Papaconstantinou C, Zenetos A, Vassilopoulou V, Tser-pes G, editors. 2007. State of the Hellenic Fisheries. Athens: National Centre for Marine Research.

Papathanassiou E, Zenetos A, editors. 2005. State of the Hellenic Marine Environment. Athens: National Centre for Marine Research.

Parlama L. 2007. Palamari Skyros. Observations on the evolution of the settlement during 3rd millennium BC and urbanization problems. In: Simantoni-Bour-nia E, Laimos AA, Mendoni LG, Κourou N, editors. Αμύμονα Έργα. Volume in Honour of V. Lambrin-oudakis. Athens. p 25–48. In Greek.

Parlama L. 2009. Palamari Skyros. The excavation of the fortifications, 2000–2007. In: Κερμάτια Φιλίας. Volume in Honour of I. Touratsoglou. Athens. p 227–41. In Greek.

Parlama L, Theochari M, Bonatsos S, Romanou Ch, Manos I. 2010. Palamari Skyros: the middle Bronze Age town. Proceedings of the International Confer-ence MESOHELLADIKA, Athens 8–12 March 2006. p 281–9.

Pavlopoulos K, Triantaphyllou Μ, Karkanas P, Kouli K, Syrides G, Vouvalidis K, Palyvos N, Tsourou T. 2010. Paleoenvironmental evolution and prehistoric human environment, in the embayment of Palamari (Skyros Island, Greece) during Middle-Late Holo-cene. Quaternary International 216(1–2):41–53.

Peharda M, Ezgeta-Balić D, Radman M, Sinjkević N, Vrgoč N, Isajlović I. 2012. Age, growth and popula-tion structure of Acanthocardia tuberculata (Bival-via: Cardiidae) in the eastern Adriatic Sea. Scientia Marina 76(1):59–66.

Perlès C. 2001. The Early Neolithic in Greece. Cam-bridge: Cambridge University Press.

Perlès C. 2003. The Mesolithic at Franchthi: an over-view of the data and problems. In: Galanidou N, Perlès C, editors. The Greek Mesolithic: Problems and Perspectives. London: British School at Athens Studies Series 10. p 79–87.

Reimer PJ, McCormac FG. 2002. Marine radiocarbon reservoir corrections for the Mediterranean and Ae-gean Seas. Radiocarbon 44(1):159–66.

Reimer PJ, Bard E, Bayliss A, Beck JW, Blackwell PG, Bronk Ramsey C, Buck CE, Cheng H, Edwards RL, Friedrich M, Grootes PM, Guilderson TP, Haflida-son H, Hajdas I, Hatté C, Heaton TJ, Hoffmann DL, Hogg AG, Hughen KA, Kaiser KF, Kromer B, Man-ning SW, Niu M, Reimer RW, Richards DA, Scott EM, Southon JR, Staff RA, Turney CSM, der Plicht J. 2013. IntCal13 and Marine13 radiocarbon age cal-ibration curves 0–50,000 years cal BP. Radiocarbon 55(4):1869–87.


Robinson SW, Thompson G. 1981. Radiocarbon cor-rections for marine shell dates with application to southern Pacific Northwest Coast prehistory. Syesis 14:45–57.

Sabelli Β, Giannuzzi-Savelli R, Bedulli D. 1990. An-notated Check-List of Mediterranean Marine Mol-lusks. Volumes I, II, III. Bologna: Societá Italiana di Malacologia.

Sampson A, editor. 2002. The Neolithic Settlement at Ftelia, Mykonos. University of the Aegean, Depart-ment of Mediterranean Studies, Rhodes.

Sampson A, editor. 2008. The Cave of the Cyclops: Me-solithic and Neolithic Networks in the Northern Ae-gean, Greece. Volume I: Intra-Site Analysis, Local Industries, and Regional Site Distribution. Prehisto-ry Monographs 21. Philadelphia: INSTAP Academic Press.

Sampson A, editor. 2011. The Cave of the Cyclops: Mesolithic and Neolithic Networks in the Northern Aegean, Greece. Volume II: Bone Tool Industries, Dietary Resources and the Paleoenvironment, and Archeometrical Studies. Prehistory Monographs 31. Philadelphia: INSTAP Academic Press.

Sangiorgi F, Capotondi L, Combourieu N, Vigliotti L, Brinkhuis H, Giunta S, Lotter AF, Morigi C, Negri A, Reichart G-J. 2003. Holocene seasonal sea-sur-face temperature variations in the southern Adriatic Sea inferred from a multiproxy approach. Journal of Quaternary Science 18(8):723–32.

Shackleton JC, van Andel TH. 1980. Prehistoric shell assemblages from Franchthi Cave and evolution of the adjacent coastal zone. Nature 288(5789):357–9.

Siani G, Paterne M, Arnold M, Bard E, Métivier B, Tis-nerat N, Bassinot F. 2000. Radiocarbon reservoir

ages in the Mediterranean Sea and Black Sea. Ra-diocarbon 42(2):271–80.

Siani G, Paterne M, Michel E, Sulpizio R, Sbrana A, Ar-nold M, Haddad G. 2001. Mediterranean Sea surface radiocarbon reservoir age changes since the Last Glacial Maximum. Science 294(5548):1917–20; supplemental online files at http://www.sciencemag.org/cgi/content/full/294/5548/1917/DC1.

Southon JR, Rodman AO, True D. 1995. A comparison of marine and terrestrial radiocarbon ages from northern Chile. Radiocarbon 37(2):389–93.

Stuiver M, Braziunas TF. 1993. Modeling atmospheric 14C influences and 14C ages of marine samples to 10,000 BC. Radiocarbon 35(1):137–89.

Stuiver M, Polach HA. 1977. Discussion: reporting of 14C data. Radiocarbon 19(3):355–63.

Stuiver M, Pearson GW, Braziunas T. 1986. Radiocar-bon age calibration of marine samples back to 9,000 cal yr BP. Radiocarbon 28(2B):980–1021.

Stuiver M, Reimer PJ, Braziunas TF. 1998. High-preci-sion radiocarbon age calibration for terrestrial and marine samples. Radiocarbon 40(3):1127–51.

Sveinbjörnsdóttir ÁE, Heinemeier J, Arneborg J, Lyn-nerup N, Ólafsson G, Zoëga G. 2010. Dietary re-construction and reservoir correction of 14C dates on bones from pagan and early Christian graves in Iceland. Radiocarbon 52(2–3):682–96.

van Andel TH, Lianos N. 1983. Prehistoric and histor-ic shorelines of the Southern Argolid Peninsula: a subbottom profiler study. Nautical Archaeology and Underwater Exploration 12(4):303–24.

Yüce H. 1995. North Aegean water masses. Estuarine, Coastal and Shelf Science 41(3):325–43.

SEA SURFACE RADIOCARBON RESERVOIR AGE CHANGES IN THE AEGEAN SEA FROM ABOUT 11,200 BP TO PRESENT

Documents

Transcript of SEA SURFACE RADIOCARBON RESERVOIR AGE CHANGES IN THE AEGEAN SEA FROM ABOUT 11,200 BP TO PRESENT