ABSTRACTS

PLENARY ORAL SESSIONS

Plenary Session 1: Radiocarbon and Archaeology
#1

RADIOCARBON DATING THE DISPERSAL OF THE FIRST ANATOMICALLY MODERN

HUMANS INTO WESTERN EUROPE
Tom Higham1, Roger Jacobi2, Laura Basell1, Rachel Wood1, Katerina Douka1,
Christopher Bronk Ramsey1
1. Oxford Radiocarbon Accelerator Unit, RLAHA, University of Oxford, Oxford.
2. The British Museum, London.

A reliable chronology is one of the keys to understanding the nature of the transition from the Mid-
dle to Upper Paleolithic in western Eurasia. The transition describes the period during which ana-
tomically modern humans (AMHs) replaced Neanderthals—who ultimately became extinct. The
period over which this took place and the length of the temporal overlap between the two groups is
a central question concerning researchers in this field with wide-ranging implications in a number of
areas; for example, the cognitive abilities of Neanderthals and whether there was any genetic
exchange between the two populations.
Within a large project funded by the NERC in the UK, we have been dating over 300 samples of
bone, shell and charcoal from more than 50 key Paleolithic sites in over 10 countries. The main
focus has been on sites with a succession of contexts containing lithic industries attributed to the
Mousterian, Uluzzian, Châtelperronian (all associated with Neanderthals), Aurignacian and Gravet-
tian (associated with AMHs).
Work undertaken in Oxford over the last decade has been aimed at improving the dating of material
between ~25–55 kyr BP, which covers the Middle–Upper Paleolithic transition. We have developed
aspects of our pretreatment chemistry, particularly the purification of bone collagen using ultrafil-
tration. When comparing the ultrafiltered results with previously determined samples of the same
bone from our laboratory, and other laboratories, the results in many cases are quite different. When
ultrafiltration is used, the dates are often older, and we consider, more accurate.
We are also applying ABOx-SC methods to samples of charcoal, which shows similar improve-
ments in many cases. In tandem, work has been undertaken on improving the applicability of our
background correction for bone (Wood et al., this conference). In addition, we have been refining
the detection of trace calcite in samples of marine shell for radiocarbon dating (Douka et al., this
conference).
In this paper, we will discuss the emerging chronology for the dispersal of the earliest anatomically
modern humans into Europe by presenting results from some of the key sites in France, Germany,
Italy, Spain, and Belgium.

1
2 Abstracts

#2

RADIOCARBON DATING OF THE FIRST EUROPEAN MODERN HUMANS: NEW DIRECT AMS
DATES OF AURIGNACIAN SHELL ORNAMENTS
Katerina Douka1, Robert E M Hedges1, Thomas F G Higham2
1. Research Laboratory for Archaeology and the History of Art, University of Oxford, Dyson Perrins Building, South Parks
Road, Oxford OX1 3QY, UK.
2. Oxford Radiocarbon Accelerator Unit, Research Laboratory for Archaeology and the History of Art, University of Oxford,
Dyson Perrins Building, South Parks Road, Oxford OX1 3QY, UK.

There are two principal hypotheses for the dispersal of the first anatomically modern humans
(AMH) into Europe. The first posits an initial movement along the Danubian corridor into north and
central Europe, while the second concerns a “southern dispersal” along the Mediterranean Rim.
Resolving which of the two is more likely is dependent upon a reliable radiocarbon chronology,
which, at present, is lacking. We have investigated over 20 sites in southern Europe containing the
Aurignacian, an Upper Paleolithic culture closely associated with the first AMH in Europe. Unfor-
tunately, the survival of bone collagen, our preferred material for analysis, is severely compromised,
because of low survival of proteins remaining. In the light of this, we have sought to date perforated
shell ornaments from these sites. In doing so, the main problems to overcome are: 1) the removal of
contaminants from the carbonate; 2) the calculation of the marine reservoir effect; and 3) uncertainty
regarding the possibility of old, fossil shell being used for manufacturing ornaments. The first prob-
lem has been addressed (see Douka et al., this conference) by developing a novel density separation
method for removing calcite contamination and the application of high-precision XRD for improved
detection of calcite in the shell samples. The second is difficult to quantify, but is probably less sig-
nificant owing to the proportionally lower effect of the marine offset when analyzing material of this
age. The final problem has been addressed by others examining the time-averaging of naturally
forming shell beds along coastal margins, which shows that “fossil” shells although present tend to
be rare thus unlikely to be selected by the AMH.
Initial results from several of the key sites that date to this period will be presented. In some
instances, we are able to gauge the success of the approach by comparing the results against the
known-age Campanian Ignimbrite tephra of 39.3 kyr BP. There are important implications for our
understanding of the dynamics of the initial dispersal of modern humans and the concomitant
decline of Neanderthal populations, which occurred during the same period.

#3

IS MORE PRECISE DATING OF PALEOINDIAN EXPANSION FEASIBLE?

Stuart J Fiedel1 and Yaroslav V Kuzmin2
1. Louis Berger Group, Richmond, VA 23219, USA.
2. Institute of Geology & Mineralogy SB RAS, Novosibirsk 630090, Russia.

Recent efforts to precisely date the florescence of the Clovis culture have been hampered by both
practical and theoretical problems. These include: 1) The era of Clovis expansion (ca. 11,200–
10,700 BP, or 13,200–12,700 cal BP) coincides with the gap between the anchored Central Euro-
pean tree-ring sequence (back to 12,400 cal BP) and the floating Bølling–Allerød sequence; 2) Clo-
vis seems to immediately precede the onset of the Younger Dryas (YD) stadial. The “black mats” of
 Abstracts 3

the US Southwest appear to mark the regional occurrence of this climatic downturn. However, the
timing and means of long-distance propagation of this climatic event are not yet well understood.
Greenland ice cores (GISP2, GRIP, and NGRIP) remain poorly synchronized, with a discrepancy of
100 to 250 years for the date of onset (as late as 12,700 cal BP, or as early as 12,950 cal BP); 3) The
YD onset was accompanied by a rapid drop of 14C ages from 11,000 to 10,600 BP in less than a cen-
tury. The mechanism causing this was probably a change in overturning circulation in the North
Atlantic. Do variable Clovis ages, often from what appear to be single-occupation contexts, reflect
this “cliff” effect, slightly earlier minor reversals during the late Allerød, or simply the practical lim-
itations of precision of the radiocarbon method? 4) Dates for Fishtail or Fell I sites (with fluted,
stemmed points) in southern South America are statistically indistinguishable from Clovis dates in
North America. Does this imply very rapid population expansion, diffusion of toolmaking tech-
niques through long-established local populations (as argued by Waters and Stafford 2007), or
abnormally large inter-hemispheric radiocarbon offsets? 5) Are recent ostensibly high-precision col-
lagen-derived dates for Paleoindian-associated fauna (e.g. horse, mammoth) reliable? Are inter-lab
blind tests of the new filtration processes necessary? These issues will be considered at the meeting.
Reference: Waters MR, Stafford Jr. TW. 2007. Science 315:1122–6.

#4

THE IMPACT OF NEW RADIOCARBON PRETREATMENT TECHNIQUES FOR UNDERSTAND-

ING THE IBERIAN MIDDLE TO UPPER PALEOLITHIC TRANSITION
R E Wood and T F G Higham
ORAU, University of Oxford.

The Iberian Middle to Upper Paleolithic transition—the period in which anatomically modern
human populations replace Neanderthal populations—has long held intense interest within the
archaeological community. In the south of the peninsula, radiometric and archaeological evidence
currently suggests that Neanderthals may have survived until the Last Glacial Maximum, that is,
around 10,000 years after the arrival of modern humans in the northern Cantabrian and Catalonian
regions. This makes southern Iberia the largest and most securely identified Neanderthal refugium
in Europe. This has profound effects on understanding the cognition of Neanderthals, the possibility
and nature of the interaction between the two species, and their response to environmental change.
However, given that the desert-like southern regions suffer from particularly poor organic preserva-
tion and that contamination is most likely to produce erroneously young ages, numerous authors
have rejected all late Neanderthal radiocarbon dates. They suggest instead that the southern part of
the peninsula was abandoned and empty of any hominins while modern humans were denizens of
the northern regions. The difficulty in assessing whether many of the radiocarbon dates are reliable
from the published information, both from a chemical and an archaeological perspective, means that
the debate cannot proceed without the construction of new chronologies. This paper presents initial
results from a study aiming to re-date key Iberian sites and fossils by dating bone (using ultrafiltra-
tion preparative methods) and charcoal (using the ABOx-SC protocol). Initial results will be pre-
sented from several key sites.
4 Abstracts

#5

ON THE ACCURACY OF RADIOCARBON DATING FOR THE NEOLITHIC

A Shukurov1, P M Dolukhanov2, D D Sokoloff3
1. School of Mathematics and Statistics, Newcastle University, Newcastle upon Tyne NE1 7RU, UK.
2. School of Historical Studies, Newcastle University, Newcastle upon Tyne NE1 7RU, UK.
3. Department of Physics, Moscow State University, Moscow 119992, Russia.

The accuracy of radiocarbon dating for Neolithic sites in Europe and the Near East is assessed using
selected, well-explored sites with a statistically large number of 14C age determinations. We select
those sites where we have reasons to believe that each represents a relatively short-lived archaeolog-
ical event. Among the sites discussed are Razdorskoe 2 (Mesolithic-Neolithic transition, ca. 7000
cal BC, southern Russia), Brunn am Gebirge (LBK, ca. 5273 cal BC, Austria), Serteya 2 (Zhizhit-
sian, late Neolithic–Early Bronze Age, ca. 2300 cal BC, northwestern Russia), together with 7 other
sites. The dates from Serteya 2 are tree-ring age determinations from a single pile-dwelling struc-
ture; those from Razdorskoe belong to a single hearth. The selected date sequences are used to cal-
culate the empirical dispersion of the calibrated dates, which is surprisingly similar in all cases,
ranging from 83 years for Serteya 2 and 99 years for Brunn am Gebirge to 214 years for Razdoskoe
2. This scatter can be attributed to various effects affecting the 14C content of the sample, e.g. con-
tamination by younger and older carbon, various systematic errors, etc. We suggest that the intrinsic
accuracy of 14C dating for the period is 100–200 years regardless of the dating technique (AMS ver-
sus conventional). It is natural to expect that the intrinsic accuracy varies with epoch and location,
but more detailed studies are required to identify such trends.

Plenary Session 2: Calibration

#6

14CCALIBRATION IN THE 2ND AND 1ST MILLENNIUM BC – EASTERN MEDITERRANEAN

RADIOCARBON COMPARISON PROJECT
B Kromer1, S Manning, M Friedrich, S Talamo
1. Heidelberg Academy of Sciences, INF 229, Heidelberg, Germany, D-69120, Germany.

Radiocarbon calibration in the 2nd and 1st millennium BC is of crucial importance to address ongo-
ing disputes between archaeology and 14C dating, such as the date of the Santorini eruption and the
dating of key sites in the Aegean and Near East regions in the Late Bronze Age. In the Eastern Med-
iterranean Radiocarbon Comparison Project (EMRCP), we remeasured German oak samples in 5-
and 10-year increments for several intervals in the 2nd and 1st millennium BC. As noted before,
some of the observed variance of the 14C dates appears to result from fluctuations of the atmospheric
14C level and it is suppressed to some extent in IntCal04. We report on the wiggle-match of the float-

ing Bronze Age juniper chronology of Gordion to our new German oak data and to the calibration
data sets of IntCal04 and IntCal98.
 Abstracts 5

#7

COMPARISON OF TERRESTRIAL AND MARINE ARCHIVE Δ14C VARIATION DURING MARINE

ISOTOPE STAGE 3

Warren Beck1, David Richards2, Dirk Hoffmann2, Peter Smart2, Joy Singarayer2, George Burr1,
Tricia Kretchmark1, Michelle Felton2, Paul Valdes2, Andy Ridgewell2
1. NSF-Arizona AMS Facility, PAS Building 81, University of Arizona, Tucson, AZ 85721, USA.
2. BRIDGE, School of Geographical Sciences, University of Bristol, University Rd., Clifton, Bristol, BS3 1DD, UK.

Since 1990 a great deal of effort has been dedicated to expanding the record of atmospheric Δ14C
variations beyond the limit of the tree-ring calibration using hybrid records based on corals, stalag-
mites, forams, and terrestrial macrofossils. Large discrepancies observed between some of these
records has prevented their combined use to formulate a unified calibration curve, though various
subsets of these data are now being used as the basis of several independent, and by some measures,
inconsistent radiocarbon calibrations. In the absence of convincing evidence from the terrestrial
realm, nearly all of these rely heavily on marine data to constrain atmospheric Δ14C by adopting
model assumptions about past changes in the marine reservoir age at specific site locations. Here we
investigate the extent to which Δ14C records based on speleothems from the terrestrial realm can be
compared with Δ14C records from the marine realm to inform our understanding of carbon exchange
between the ocean and atmosphere during marine isotope stage 3. Using an earth-system model
(GENIE) to elucidate likely past changes in marine reservoir effect, we find that the Atlantic and
Pacific basins significantly differ in their response for periods of rapid ocean circulation change
such as the Younger Dryas. These findings have implications for the reconstructed atmospheric
Δ14C record. Secondly, while we recognize that estimates of Δ14Catm using speleothems suffer from
a source of uncertainty analogous to the marine reservoir age for corals and forams, the Δ14C pattern
revealed by the Bahamas speleothems and Cariaco Basin records[1] (the later adjusted to the most
recent Greenland ice core chronology[2]) are now in broad agreement. GENIE model results for
marine isotope stage 3 using the GLOPIS-75[3] geomagnetic intensity record, are also substantially
in agreement with these records, indicating that production rate changes dominate the temporal pat-
tern of 14C concentration during this time.
[1] Hughen et al. (2006) Quaternary Science Reviews 25:3216–3227.
[2] Andersen et al. Clim. Past 3, 1235–1260.
[3] Laj et al. (2004) In: Timescales of the Paleomagnetic Field.

#8

SUIGETSU 2006: A WHOLLY TERRESTRIAL RADIOCARBON CALIBRATION CURVE

R A Staff1, C Bronk Ramsey1, C Bryant2, F Brock1, H Lamb3, M Marshall3, A Brauer4,
G Schlolaut4, P Tarasov5, R Payne6, E Pearson6, Y Yokoyama7, J Tyler7, T Haraguchi8,
K Gotanda9, H Yonenobu10, T Nakagawa6, Suigetsu 2006 Project Members
1. Research Laboratory for Archaeology and the History of Art, University of Oxford, UK.
2. NERC Radiocarbon Laboratory, SUERC, East Kilbride, UK.
3. Institute of Geography and Earth Sciences, University of Wales Aberystwyth, UK.
4. GeoForschungsZentrum, Potsdam, Germany.
5. Department of Earth Sciences, Freie Universität, Berlin, Germany.
6 Abstracts

6. Department of Geography, University of Newcastle-upon-Tyne, UK.

7. Department of Earth and Planetary Sciences, University of Tokyo, Japan.
8. Department of Geosciences, Osaka City University, Japan.
9. Faculty of Policy Informatics, Chiba University of Commerce, Japan.
10. College of Education, Naruto University of Education, Japan.

Lake Suigetsu, Honshu Island, central Japan (35°35’N, 135°53’E) provides an ideal sedimentary
sequence from which to derive a wholly terrestrial radiocarbon calibration curve back to the limits
of radiocarbon detection (ca. 60 cal kyr BP). The presence of well-defined, annually-deposited lam-
inae (varves) throughout the entirety of this period provides an independent, high-resolution chro-
nometer from which radiocarbon measurements of plant macrofossils from the sediment column can
be directly related. The importance of this site for radiocarbon calibration purposes was brought to
the attention of the community through the works of Kitagawa and van der Plicht (1998a, 1998b,
2000). However, data from this initial Lake Suigetsu project were found to diverge significantly
from alternative, marine-based calibration data sets released soon thereafter (e.g. Beck et al. 2001;
Hughen et al. 2004). The reasons for such divergence lie in the absolute age chronology of the initial
Suigetsu project – the result of both missing sections of the retrieved sedimentary column, as well
as varve counting uncertainties.
Lake Suigetsu was re-cored in summer 2006, with material obtained from four separate bore-holes,
producing a composite core of 73.19 m length, lacking any of the age gaps of the former study. The
purpose of the present paper is therefore to refresh the community’s knowledge of the Lake Suigetsu
site; to discuss the multi-disciplinary techniques being utilized to derive both chronological (Schlo-
laut et al., this conference) and paleoenvironmental data from throughout the sedimentary profile;
and to demonstrate the AMS radiocarbon results thus far obtained from the Suigetsu 2006 project.
Beck JW, Richards DA, Edwards RL, Silverman BW, Smart PL, Donahue DJ, Hererra-Osterheld S, Burr GS, Calsoyas L, Jull
AJT, Biddulph D. 2001. Extremely large variations of atmospheric 14C concentration during the Last Glacial period. Sci-
ence 292:2453–2458.
Hughen K, Lehman S, Southon J, Overpeck J, Marchal O, Herring C, Turnbull J. 2004. 14C activity and global carbon cycle
changes over the past 50,000 years. Science 303:202–207.
Kitagawa H, van der Plicht J. 1998a. Atmospheric radiocarbon calibration to 45,000 yr B.P.: Late Glacial fluctuations and
cosmogenic isotope production. Science 279:1187–1190.
Kitagawa H, van der Plicht J. 1998b. A 40,000-year varve chronology from Lake Suigetsu, Japan: extension of the 14C cali-
bration curve. Radiocarbon 40(1):505–515.
Kitagawa H, van der Plicht J. 2000. Atmospheric radiocarbon calibration beyond 11,900 cal BP from Lake Suigetsu lami-
nated sediments. Radiocarbon 42(3):369–380.

#9

ASSESSMENT OF THE INTEGRITY OF THE SOUTHERN HEMISPHERE 14C CALIBRATION

CURVE AND ITS EXTENSION FROM AD 785 TO 195 BC, WITH PARTICULAR EMPHASIS ON
THE INTERHEMISPHERIC OFFSET
Alan Hogg1, Jonathan Palmer2, Gretel Boswijk3, Christopher Bronk Ramsey4, Chris Turney5,
Gerry McCormac6, David Brown6
1. Radiocarbon Laboratory, University of Waikato, PB 3105, Hamilton 3240, New Zealand.
2. Gondwana Tree-Ring Laboratory, P.O. Box 14, Little River, Canterbury 7546, New Zealand.
3. School of Geography & Environmental Science, University of Auckland, PB 92019, Auckland, New Zealand.
 Abstracts 7

4. Oxford Radiocarbon Accelerator Unit, Research Lab for Archaeology, Dyson Perrins Building, South Parks Road, Oxford
OX1 3QY, UK.
5. School of Geography, Archaeology and Earth Resources, University of Exeter, Devon EX4 4RJ, UK.
6. Centre for Climate, the Environment & Chronology (14CHRONO), School of Geography, Archaeology and Palaeoecol-
ogy, Queen’s University Belfast, Northern Ireland BT7 1NN, UK.

Numerous studies show a measurable difference in Δ14C in wood from trees growing in the same
year in opposite hemispheres. Although this interhemispheric offset was assumed to be constant in
time, duplicate high-precision Δ14C measurements on contemporaneous Southern Hemisphere (SH)
and Northern Hemisphere (NH) sample pairs from AD 955–1845 demonstrate that it is variable,
with SH wood 8–80 14C years older during this interval. The SH atmospheric 14C calibration curve,
SHCal04, consists of a measured phase (AD 955–1955) and a modeled phase (the remainder of the
Holocene), with the latter based upon the NH curve, IntCal04, with an applied variable offset (based
upon that observed from AD 955–1845) generated by a random effects model.
We present >100 high-precision 14C measurements on decadal samples of dendrochronologically-
dated New Zealand kauri to extend the measured phase of SHCal04 and to test the integrity of the
modeled phase between AD 785–195 BC. The interhemispheric offset is investigated in this interval
with 15 high-precision 14C measurements on selected paired decadal samples of dendrochronologi-
cally-dated Irish oak (Quercus petraea). Five decadal samples (AD 955–995) overlap with the mea-
sured phase of SHCal04 for QA purposes. We also introduce a new method for assessing published
SH Holocene data sets to determine their usefulness in estimating the offset for these earlier time
periods.
Our kauri measurements are in very close agreement where they overlap with the measured phase
of SHCal04 and with the modeled phase of SHCal04 from AD 785–195 BC and confirm the integ-
rity of SHCal04 for this period. The kauri-oak decadal sample pairs indicate a continuance of the
interhemispheric offset, which averages 27–58 years for the measured intervals. We show that many
of the published SH Holocene data sets are too imprecise to detect an offset of this magnitude.

#10

DEVELOPMENTS IN THE CALIBRATION AND MODELING OF RADIOCARBON DATES

C Bronk Ramsey1, M Dee1, S Lee1, T Nakagawa2, R A Staff1, Suigetsu 2006 Project Members
1. Research Laboratory for Archaeology & the History of Art, University of Oxford, UK.
2. Department of Geography, University of Newcastle upon Tyne, Newcastle upon Tyne, UK.

Calibration is a core element of radiocarbon dating and is undergoing rapid development on a num-
ber of different fronts. This is most obvious in the area of radiocarbon archives suitable for calibra-
tion purposes, which are now demonstrating significantly greater coherence over the earlier age
range of the technique. Of particular significance to this end are the development of purely terres-
trial archives such as those from the Lake Suigetsu sedimentary profile (Staff et al., this conference)
and kauri tree rings from New Zealand, in addition to the groundwater records from speleothems.
Equally important, however, is the development of statistical tools that can be used with, and
develop, such calibration data. In the context of sedimentary deposition, age-depth modeling pro-
vides a very useful way to analyze series of measurements from cores, with or without varve infor-
mation. New methods are under development, making use of model averaging, that generate more
8 Abstracts

robust age models. In addition, all calibration requires a coherent approach to outliers, for both sin-
gle samples and where entire data sets might be offset relative to the calibration curve. This paper
will look at current developments in these areas.

#11

PROGRESS IN THE EXTENSION AND UPDATE OF THE INTCAL04, MARINE04, AND SHCAL04
RADIOCARBON CALIBRATION CURVES
P J Reimer1, M G L Baillie1, E Bard2, A Bayliss3, J W Beck4, M Blaauw1, P G Blackwell5,
C Bronk Ramsey6, C E Buck5, G S Burr4, R L Edwards7, M Friedrich8,9, P M Grootes10,
T P Guilderson11,12, I Hajdas13, T J Heaton5, A G Hogg14, K A Hughen15, K F Kaiser16,17,
B Kromer9, F G McCormac1, S W Manning18, R W Reimer1, D A Richards19, J R Southon20,
S Talamo21, F W Taylor22, C S M Turney23, J van der Plicht24,25, C E Weyhenmeyer26
1. 14CHRONO Centre for Climate, the Environment and Chronology, School of Geography, Archaeology and Palaeoecology,
Queen’s University Belfast, Belfast BT7 1NN, UK.
2. CEREGE, UMR-6635, Europole de l’Arbois BP80, 13545 Aix-en-Provence cdx 4, France.
3. English Heritage, 23 Savile Row, London W1S 2ET, UK.
4. Department of Physics, University of Arizona, Tucson, AZ 85721, USA.
5. Department of Probability and Statistics, University of Sheffield, Sheffield S3 7RH, UK.
6. Research Laboratory for Archaeology and the History of Art, Dyson Perrins Building, South Parks Road, Oxford OX1
3QY, UK.
7. Department of Geology and Geophysics, University of Minnesota, MN 55455-0219, USA.
8. Institut für Botanik-210, D-70593 Stuttgart, Germany.
9. Heidelberger Akademie der Wissenschaften, Im Neuenheimer Feld 229, D-69120 Heidelberg, Germany.
10. Leibniz Laboratory, Christian-Albrechts-Universität zu Kiel 24098, Germany.
11. Center for Accelerator Mass Spectrometry L-397, Lawrence Livermore National Laboratory, Livermore, CA 94550,
USA.
12. Ocean Sciences Department, University of California-Santa Cruz, Santa Cruz, CA 92697, USA.
13. Labor für Ionenstrahlphysik, ETH, 8092 Zurich, Switzerland.
14. Radiocarbon Dating Laboratory, University of Waikato, Private Bag 3105, Hamilton, New Zealand.
15. Woods Hole Oceanographic Institution, Department of Marine Chemistry & Geochemistry, Woods Hole, MA 02543
USA.
16. Swiss Federal Institute for Forest, Snow and Landscape Research WSL, Zurcherst. 111, 8903 Birmensdorf, Switzerland.
17. Department of Geography, University of Zurich-Irchel, 8057 Zurich, Switzerland.
18. Malcolm and Carolyn Wiener Laboratory for Aegean and Near Eastern Dendrochronology, Cornell Tree Ring Labora
tory, Cornell University, Ithaca, NY 14853, USA.
19. School of Geographical Sciences, University of Bristol, University Road, Bristol BS8 1SS, UK.
20. Department of Earth System Science, University of California-Irvine, Irvine, CA 92697, USA.
21. Max Planck Institute for Evolutionary Anthropology, Department of Human Evolution, Deutscher Platz 6, D-04103
Leipzig, Germany.
22. Institute for Geophysics, University of Texas, Austin, TX 78758-4445, USA.
23. School of Geography, Archaeology and Earth Resources, University of Exeter, Exeter EX4 4QJ, UK.
24. Centrum voor Isotopen Onderzoek, Rijksuniversiteit Groningen, Nijenborgh 4, 9747 AG Groningen, Netherlands.
25. Faculty of Archaeology, Leiden University, P.O.Box 9515, 2300 RA Leiden, Netherlands.
26. Department of Earth Sciences, Syracuse University, Syracuse, NY 13244-1070 USA.

There are currently few purely atmospheric 14C calibration archives with absolutely known calendar
ages beyond the end of the European tree-ring chronologies; however, earth scientists and archaeol-
ogists need to estimate calendar age ranges throughout the radiocarbon dating range. The IntCal
Working Group plans to provide a fully updated curve taking into consideration new Holocene tree-
ring data, a floating European tree-ring record anchored with quantifiable uncertainty, U/Th-dated
 Abstracts 9

corals, speleothems, varved sediments, and other marine and terrestrial data sets with climatic cor-
relation to independently dated timescales. To utilize many of the archives requires assumptions
about the size and time-dependency of reservoir offsets. Considerable progress has been made in our
understanding of these systems which will aid in this process.
Progress has also been made modeling the sources of uncertainty and error dependence within the
data. The random walk model (rwm) used to construct IntCal04, Marine04, and SHCal04 has been
re-implemented using a Markov Chain Monte Carlo algorithm (Blackwell and Buck 2008) to allow
greater flexibility. This no longer requires us to treat the value of the calibration curve at adjacent
points as independent. Furthermore we are able to incorporate ordering constraints and adaptively
estimate the rwm’s variability. Future work will include investigating reservoir offset dependency
and permitting more inhomogeneous changes in the calibration curve’s value through time.
In the meantime, an interim curve will be provided which includes the absolute tree-ring extension
to ca. 12550 cal BP, Cariaco Basin and Iberian margin data sets on the Hulu Cave U/Th timescale,
and additional published coral data. Atlantic marine data at the beginning of the Younger Dryas will
be omitted for ca. 400 years based on evidence of a decline in the surface reservoir age. The tree-ring
based calibration curve to ca. 12,000 cal BP will remain unchanged in the interim curve.

Plenary Session 3: First Detection of Radiocarbon Using AMS: A Retrospective

#12
FIRST AMS MEASUREMENTS USING A CYCLOTRON - A REVIEW

Richard Muller
Lawrence Berkeley Laboratory, University of California, Berkeley CA 94720, USA.

My journey to AMS began in 1974, when Will Happer suggested the possibility of using lasers to
detect minute radioactive trails from a nuclear submarine. Luis Alvarez was familiar with the unsuc-
cessful work by Michael Anbar using a mass spectrometer to attempt direction detection of 14C, and
he suggested to me the laser method might work for radiocarbon. I tried hard to make it succeed, and
failed, but that’s how I became aware of the potential of direct detection.
Later, I became involved in a quark search with Alvarez using a cyclotron. We found no quarks, but
I recognized that the high-energy accelerator allowed enormous background suppression and could
be used for 14C. The first experimental successes we had with this new method (soon to be called
AMS) was with tritium dating; later we were the first to succeed with 10Be. Shortly after our Berke-
ley team announced the idea and the tritium success, and several other groups immediately recog-
nized the potential advantage of using a tandem van de Graaff, since negative ions suppressed the
nitrogen backgrounds.
The tandem quickly dominated the field. Eventually, we built a table-top negative ion cyclotron at
Berkeley, and managed to make AMS for 14C work on this small device. But the field exploded, and
left me way behind, watching in amazement as AMS developed into a standard tool for not only for
radioisotope dating, but for applications in medicine and geophysics.
10 Abstracts

#13

ANCIENT AMS: DEVELOPMENT OF AMS AT MCMASTER UNIVERSITY

Erle Nelson
Simon Fraser University, Burnaby, BC, Canada.

As requested, this talk will be a personal and subjective recounting of the circumstances and the
details of the experiment described in Nelson, Korteling, and Stott (1977), Science 198, p 507.

Plenary Session 4: Intercomparisons

#14

THE RADIOCARBON INTERCOMPARISON (VIRI): (PART 1) THE FINAL STAGE

P Naysmith1, E M Scott2, G T Cook1
1. SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, Scotland, UK.
2. Statistics Department, University of Glasgow, Glasgow G12 8QQ, Scotland, UK.

The Fifth International Radiocarbon Intercomparison (VIRI) continued the tradition of TIRI (third)
and FIRI (fourth) providing an independent check on laboratory procedures. VIRI was designed to
have three stages, spread over several years, involving two sets of specific sample types (grain and
bone), and then a final stage involving a wide variety of common sample materials.
In stage 3, 7 samples were provided to all laboratories, comprising 3 wood samples, as well as cel-
lulose, shell, barley mash and humic acid samples (samples K, L, M, O, R, S, U). Radiometric lab-
oratories received a further charcoal sample (sample P). while a further 4 samples were provided for
AMS laboratories, comprising one further wood sample, a charcoal and two humic acid samples
(sample N, Q, T, J). More than 50 laboratories, of which 19 were radiometric) participated. In addi-
tion, many AMS facilities reported replicate results.
In the preliminary analysis, initial consensus values were calculated using the median, which also
led to the identification of a number of outlying values. Then, the consensus or assigned values were
calculated using a weighted average.
For the analysis, for the first time, we have reported z scores, calculated as
Z = (XM – XA) / σp

where XM, is the reported result, XA is the assigned or true value for the material, and σp is the target
value for the standard deviation for values of X. The value for σp is determined by fitness for pur-
pose and represents the amount of uncertainty in the results that is tolerable in relation to the purpose
of the analysis, although more commonly the laboratory quoted error is used. XA may be known or
assessed as the consensus value. Interpretation of the z score reflects the accuracy achieved and pro-
vides a means of making a judgement concerning fitness for purpose.
Results from the third stage of VIRI are reported and assigned values given for the 12 samples used.
 Abstracts 11

#15

THE RADIOCARBON INTERCOMPARISON (VIRI): (PART 2) AN OVERALL ASSESSMENT OF

LABORATORY PERFORMANCE
E M Scott1, G T Cook2, P Naysmith2
1. Statistics Department, University of Glasgow, Glasgow G12 8QQ, Scotland, UK.
2. SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, Scotland, UK.

Proficiency testing is a widely used, international procedure common within the analytical chemis-
try community. A proficiency trial (which VIRI is) often follows a standard protocol, including the
analysis which is typically based on z scores, with one key quantity σp.
From a laboratory intercomparison (sometimes called a proficiency trial), we hope to gain an assess-
ment of accuracy (in this case from the dendrodated samples), laboratory precision (from the dupli-
cate samples), and generally measurement variability and hence realistic estimates of uncertainty. In
addition, given our stated aim of creating an archive of reference materials, we also gain a determi-
nation of consensus values for new reference materials.
What do laboratories gain from taking part? Firstly, participation helps ensure the results from a lab-
oratory are meaningful; it contributes to and enhances the laboratory’s reputation. What do partici-
pating labs want? They want to have relevant test material (samples), confidence in the homogeneity
of test material and confidence in the assigned value.
VIRI samples have been chosen to deliver these objectives and the sample ages included in the dif-
ferent stages, by design, spanned modern to background. With regard to pretreatment, some samples
required intensive pretreatment (e.g. bone), while others required none (e.g. cellulose and humic
acid). Sample size was not optimized, and indeed some samples were provided solely for AMS mea-
surement. In this sense, VIRI presented a more challenging exercise than previous intercompari-
sons, since by its design in stages, one can explore improvements (or deteriorations) over time in
laboratory performance. The analysis of the results makes use of the idea of z scores, which are
widely used in the general area of proficiency testing and delivers a set of assigned ages for each
sample (and their uncertainties). At each stage, more than 50 laboratories have participated, with an
increasing demographic shift towards more AMS and fewer radiometric laboratories. In this paper,
an overview of the complete exercise will be presented.

#16

AMS SAMPLE COMPARISON CLOSE TO THE LIMIT OF THE METHOD: A LIMIT OR CHAL-
LENGE?
Sahra Talamo and Mike Richards
Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, Deutscher Platz 6, D-04103 Leipzig,
Germany.

Bone is a commonly used material in radiocarbon dating yet, at ages close to the limit of the method,
it presents a substantial challenge to remove contamination and produce accurate ages. We report
here preliminary results of a dating study of two bones older than 30,000 years each treated with a
suite of pretreatment procedures, especially ultrafiltration (Brown et al. 1988). Substantial differ-
12 Abstracts

ences in the age determination were observed, which is most likely linked to crucial steps in the
removal of contamination both in the bone and in the laboratory. In short, using a comprehensive
sequence of pretreatment procedures, including ultrafiltration, we obtain generally older and more
consistent ages.
Brown TA, Nelson DE, Vogel JS, Southon JR. 1988. Improved collagen extraction by modified Longin method. Radiocarbon
30(2):171–7.

#17

ARE UNIVERSAL COMPACT AMS FACILITIES A COMPETITIVE ALTERNATIVE TO LARGER

TANDEM ACCELERATORS?
M Suter, A M Müller, V Alfimov, C Christl, T Schulze-König, H-A Synal, L Wacker
Laboratory of Ion Beam Physics, Schafmattstrasse 20, CH-8093 Zürich, Switzerland.

In the last decade, small and compact AMS systems became available which are operated at termi-
nal voltages of 1 MV and below. This new category of instruments has become competitive for
radiocarbon detection to larger tandem accelerators and meanwhile many of these instruments are
successfully used for radiocarbon dating or biomedical applications. The AMS group in Zurich has
also demonstrated that small universal instruments can be built, which allow measurements for
many other radioisotopes such as 10Be, 26Al, 129I, and actinides. In addition, sufficient sensitivity is
obtained to perform 41Ca measurements for biomedical applications. A summary of the recent
developments made at the 600kV Pelletron in Zurich is given and the performance is compared with
that of a commercial compact instrument from HVEE operating at 1 MV at CNA in Seville, Spain,
as well as with that of larger AMS facilities.
It turns out that the ion optics and the detection system are very essential for the performance. The
most challenging problem is the measurement of 10Be because of the high intensity of 10B, which
has to be suppressed by 8–10 orders of magnitude. A foil degrader method combined with a high-
resolution gas ionization detector provides the suppression power needed. With a second magnet on
the high-energy side, isotopic background (9Be) can be eliminated. An overall transmission of more
than 10% for 10Be can be expected. At the same time, this second magnet improves also the mass
resolution for heavy isotopes.
 Abstracts 13

PARALLEL ORAL SESSIONS

Parallel Session 2-1: Bone Dating and Chemistry
#18

THE CARBON ORIGIN OF STRUCTURAL CARBONATE IN BONE APATITE OF CREMATED

BONES
Mark Van Strydonck1, Mathieu Boudin1, Guy De Mulder2
1. Royal Institute for Cultural Heritage, Jubelpark 1, 1000 Brussels, Belgium.
2. Department of Archaeology, Ghent University, Blandijnberg 2, 9000 Ghent, Belgium

The validity of 14C dating of cremated bones has been proven in laboratory intercomparison tests as
well as in the dating of duplicate samples of charcoal/cremated bone. Post-depositional contamina-
tion of well-cremated bones seems to be limited to the deposition of secondary carbonate and no CO3
radicals are trapped by substitution. Nevertheless, the fact that charred bones showed aberrant results
even after proper pretreatment means that the possibility of CO3 substitution must be reconsidered.
The most observable chemical changes in the bone, however, do not occur during the post-deposi-
tional phase, but during the cremation phases: dehydration, burning of the collagen, changes in crys-
tallinity, and loss of structural carbonate. So if any exchange between the environment and the bone
carbonate is happening, it is most probably taking place during the cremation. Because the radiocar-
bon age of the fuel in a pyre (relatively young trees) is comparable with the age of the cremated
body, a possible exchange is not recognizable in the date. In order to circumvent this problem, fresh
animal bones were cremated under laboratory conditions designed to emulate as much as possible a
normal cremation, but with the use of a fossil fuel. In the first experiment, bone was cremated in an
electric furnace filled with coal (anthracite) and in the second experiment bones were cremated in
the flame of a bunsen burner using natural gas. The bunsen and the bone were placed in a brick hous-
ing to create a protected atmosphere. In both experiments, the CO2 concentration in the ovens was
higher than in the free atmosphere and the CO2 was depleted in 14C. Also, the 14C content of the cre-
mated bones was lower (lower pMC) than the 14C content of a bone from the same animal cremated
in the electric furnace under natural conditions. Duplicate measurements gave similar results, but
important differences in 14C content of the cremated bones in the electric oven with the coal and the
cremation in the flame of the bunsen were observed.

#19

DEVELOPMENT OF A NEW ANALYTICAL METHOD FOR RADIOCARBON DATING BONES

USING THE AMINO ACID HYDROXYPROLINE
Anat Marom, James S McCullagh, Robert E M Hedges
Research Laboratory for Archaeology and the History of Art, University of Oxford, UK.

Bones are widely found in archaeological sites, and have been radiocarbon dated successfully for
the past several decades. Isolation and analysis of bone-specific molecules continues to be of inter-
est, however, as a way of eliminating contamination from soil and museum treatment. 4-hydrox-
yproline (Hyp) serves as an ideal candidate, consisting of about 10% of bone collagen but not found
in significant amounts elsewhere in nature.
14 Abstracts

A method for the separation and isolation of amino acids from bone collagen has been developed in
the Oxford lab, using a mixed-mode (i.e. ion-exchange combined with hydrophobic mechanisms),
semipreparative HPLC. The amino acids do not require derivatization, and no organic solvents are
used, thereby avoiding addition of carbon. The Hyp fraction can therefore be isolated from hydro-
lyzed bone reasonably easily, in principle providing reliable dates.
We describe the developments necessary to minimize laboratory-derived contamination in sample
processing and to optimize the transfer from HPLC output to the graphitization process. We will
present data demonstrating the isolation of Hyp does not add significant background carbon to the
radiocarbon measurement, together with preliminary dates. In order to verify the accuracy and pre-
cision of the technique, we intend to obtain radiocarbon dates on archaeological bone samples which
are very old (>45 kyr), highly contaminated, or have too low collagen for normal chemical methods.

#20

RADIOCARBON AGE OF BONE FRACTIONS FROM UNDERWATER-COLLECTED GRAY

WHALE BONES, JY REEF, OFFSHORE GEORGIA, USA
Alexander Cherkinsky1, Scott E Noakes1, Ervan G Garrison1, Greg B McFall2
1. The University of Georgia, Athens, GA 30602, USA.
2. The Grays Reef National Marine Sanctuary, NOAA, Savannah, GA 31411, USA.

The gray whale exists now only in the North Pacific Ocean and is completely extinct from the North
Atlantic Ocean since the end of the 18th century. To date, there is very a little information about
radiocarbon dating of underwater bone samples in the literature. During the last 3 years, scientific
divers from UGA and NOAA have been conducting research at JY reef, approximately 20 nautical
miles offshore Georgia, USA. JY Reef is an ancient shell bed reef that is slowly eroding due to nat-
ural ocean currents and storm events. The shell bed reef is primarily composed of large cold-water
scallops and has been dated in the range of 33 to 40 kyr. As a result of the erosion, research divers
have recovered four mineralized bones of gray whales in about 17 m of seawater.
The samples have been dated on the organic fraction and mineral fraction of bioapatite. The bone
shape was very well preserved; however, the collagen fraction was almost completely destroyed and
the concentration of organic carbon was below 0.3%. In all four cases, the bioapatite fractions have
an older radiocarbon age in the range of 33–37.5 kyr, whereas the organic, collagen-like fractions
were significantly younger with radiocarbon ages from 8.3 to 23 kyr. The oldest organic fraction age
is in the most dense rib bone, which is also youngest on the bioapatite fraction. The dense bone is
less exposed to attacks by microorganisms and apparently still preserved some original collagen,
which certainly has been contaminated by products of decomposition. Thus, the Pleistocene bones
buried underwater condition could not be dated by the organic fraction due to high microbiological
activity, decomposition of original collagen, and replacement of it with foreign organic matter,
which is significantly younger than the bone itself.
 Abstracts 15

#21

DIETARY RECONSTRUCTION AND RESERVOIR CORRECTION OF 14C DATES ON BONES

FROM PAGAN AND EARLY CHRISTIAN GRAVES FROM ICELAND
Árný E Sveinbjörnsdóttir1, Jan Heinemeier2, Jette Arneborg3, Niels Lynnerup4,
Gudmundur Ólafsson5, Gudný Zoëga6
1. Institute of Earth Sciences, University of Iceland, Sturlugata 7, IS-101 Reykjavík, Iceland.
2. AMS 14C Dating Centre, Department of Physics and Astronomy, Aarhus University, DK-8000 Aarhus C, Denmark.
3. The National Museum Of Denmark, Frederiksholms Kanal 12, DK-1220 Copenhagen K, Denmark.
4. Laboratory of Biological Anthropology, Panum Institute, Blegdamsvej 3 DK-2200, Copenhagen N, Denmark.
5. National Museum of Iceland, Sudurgata 41, IS-101 Reykjavík, Iceland.
6. The Skagafjordur Heritage Museum, 560 Varmahlíd, Iceland.

In this study, δ13C and δ15N of bone samples from 60 skeletons (57 humans, 2 horses, and a dog)
excavated from pagan and early Christian graves from 20 localities in Iceland are used to reconstruct
diet of the first settlers in Iceland and possible differences in diet depending on distance between
excavation site and the seashore. We have 14C dated 19 of these skeletons and used the carbon iso-
topic composition (δ13C) to correct for the marine reservoir effect (the 14C difference between ter-
restrial and mixed marine organisms). The reservoir-corrected ages lie in the range of 783 ± 36 to
1159 ± 35 BP, or from AD 887 to 1258. The reservoir age corrections were checked by comparing
14C dates of a horse (terrestrial), a dog (highly marine), and a human (mixed diet) from the same

burial. The range in measured marine protein percentage in individual diet is from about 10% up to
60%, mostly depending on the geographical position (distance from the sea) of the excavation site.
We had access to the skeleton (AAR-5908) of the Skálholt bishop Páll Jónsson, whose remains are
enshrined at the Episcopal residence in Skálholt, southern Iceland. According to written sources, the
bishop died in AD 1212. According to our dietary reconstruction, his bones were about 17% marine,
which is within the range of human skeletons from the same geographical location. The reservoir
corrected radiocarbon age of the skeleton agrees with the historical date. We also report measure-
ments of 24 skeletons from the Christian/conversion period cemetery site of Keldudalur, in Ska-
gafjördur, northern Iceland, and some pagan graves close by with bones from humans and a dog,
excavated in 2002–3. AMS reservoir-corrected 14C datings of the 24 skeletons confirm that the cem-
etery dates from just after AD 1000 up to the early 12th century, suggesting that it was in use only
for 100–150 years. About 20% of individual diet was from marine resources.

#22

EXPERIMENTAL STUDY ON THE ORIGIN OF CREMATED BONE APATITE CARBON

Matthias Huels, Pieter M Grootes, Marie-Josée Nadeau, Helmut Erlenkeuser, Nils Andersen
Leibniz Laboratory for Radiometric Dating and Isotope Research, Christian Albrecht University, Kiel, Germany.

Bones which underwent burning at high temperatures, i.e. cremation, do not contain organic carbon
anymore. Lanting et al. (2001) proposed that some of the original structural carbonate, formed dur-
ing bioapatite formation, survived. Parallel radiocarbon dating of cremated bone apatite and con-
temporary charcoal presented in their study seems to confirm this assumption. However, stable car-
bon isotope composition of carbonate in cremated bones is consistently lighter than in uncremated
material and is closer to the δ13C values seen in C3 plant material. This raises the question of the ori-
gin of carbonate in cremated bone apatite, i.e. is the isotope signal caused by isotopic fractionation
16 Abstracts

during cremation or by an exchange of carbon with the local cremation atmosphere and thus carbon
from the burning fuel.
To study the changes in carbon isotopes (14C, 13C) of bone apatite during burning up to 800 °C a
modern bovine bone (14Ccol: 106.2 pMC, δ13Ccol: –13.4‰, δ13Cap: –5.0‰) was exposed to a contin-
uous flow of an artificial atmosphere (basically a high-purity O2/N2 gas mix) under defined condi-
tions (temperature, gas composition). To simulate the influence of the fuel carbon available under
real cremation conditions, fossil CO2 (14C: 0.18 ± 0.02 pMC, δ13C: –32.64 ± 0.01‰) was added at
different concentration. Infrared vibrational spectra revealed details on the crystal configuration of
the burned material. To get cremated bone apatite material similar to archaeological cremated bone
apatite, according to crystallographical criteria, it was necessary to add water vapor (ultra pure) to
the atmosphere within the oven. The isotope results indicate an effective exchange of carbon
between bone apatite and atmosphere depending on temperature and CO2 concentration.

#23

DIETARY HABITS AND FRESHWATER RESERVOIR EFFECTS IN BONES FROM A NEOLITHIC

NORTHERN GERMAN CEMETERY
Jesper Olsen1, Jan Heinemeier2, Harald Lübke3, Friedlich Lüth4, Thomas Terberger5
1. Department of Earth Sciences, Aarhus University, Denmark.
2. Department of Physics and Astronomy, Aarhus University, Denmark.
3. Archaeological State Museum, ZBSA, Germany.
4. Römisch-Germanische Kommission Frankfurt, Germany.
5. Chair of Prehistory. University of Greifswald, Germany.

Within a project on Stone Age sites of northeastern Germany, 26 burials from the Ostorf cemetery
and some further Neolithic sites have been analyzed by more than 40 AMS dates. Eighteen individ-
uals have also been studied on stable isotopes and will be discussed in this presentation.
We here present the results of stable isotope and radiocarbon measurements together with control
radiocarbon dates on grave goods from terrestrial animals such as tooth pendants found in 11 of the
graves. Age differences between human individuals and their associated pendants are used to calcu-
late 14C reservoir effects. The resulting substantial reservoir effects have lead to misleadingly high
radiocarbon ages of their remains, which originally indicated a surprisingly early occurrence of
graves and long-term use of this Neolithic burial site. We demonstrate that in order to radiocarbon
date the human bones from Ostorf cemetery, it is of utmost importance to distinguish between a ter-
restrial- versus freshwater-influenced diet. The latter may result in significant reservoir ages with
apparent radiocarbon ages up to ca. 800 yrs too old. The carbon and nitrogen isotopic composition
may provide a basis for or an indicator of necessary corrections of dates on humans where no datable
grave goods of terrestrial origin such as tooth pendants or tusks are available. Based on the associ-
ated control animals, there is no evidence that the dated earliest burials occurred any earlier than
3300 BC, in contrast to the original first impression of the grave site.
 Abstracts 17

Parallel Session 3-1: Radiocarbon & AMS - Small and Large I

#24

MICADAS: ROUTINE AND HIGH-PRECISION RADIOCARBON DATING

L Wacker1, G Bonani1, I Hajdas1, B Kromer2, M NÏmec1,3, M Ruff1,3, H-A Synal1

1. Ion Beam Physics, Paul Scherrer Institute and ETH Zurich, 8093 Zurich, Switzerland.
2. Heidelberg Academy of Sciences, 69120 Heidelberg, Germany.
3. Department of Chemistry and Biochemistry, University of Bern, 3012 Bern, Switzerland.

The MIni CArbon DAting System (MICADAS) is in routine operation at ETH Zurich for more than
1½ years. The instrumental limits and their impact on real measurements were extensively explored
during this time and will be presented. A series of measured reference materials and samples with a
known age in the range from modern back to 45 kyr BP will demonstrate the reliability and stability.
The MICADAS performs extremely well and goes beyond what is generally accepted as high-pre-
cision radiocarbon measurements (<2‰). The main reason for this originates in the compact design,
which shows more similarities to a stable isotope mass spectrometer than to an accelerator-based
AMS system and thus runs very stable over time. We will support our findings on several examples
of high-precision measurements and discuss the required effort in detail.

#25

TWO YEARS OF SSAMS AT ANU

L K Fifield1, S J Fallon2, K James2, R Norman2
1. Department of Nuclear Physics, Research School of Physics and Engineering, The Australian National University, ACT
0200, Australia.
2. Research School of Earth Sciences, The Australian National University, ACT 0200, Australia.

We now have two years of operational experience with the NEC single-stage accelerator mass spec-
trometer (SSAMS) at the ANU. In this time, ~1400 unknowns have been measured. The principal
issue that we have confronted is variability in the δ13C measured by the machine, as a consequence
of variable 12C transmission with intensity. In the search for a solution, we replaced the 150 MΩ
resistors on the acceleration tube with 30 MΩ resistors in order to increase the stabilizing current
from 40 to 200 μA. This did not, however, result in any noticeable improvement. A possibly related
observation is that, at intensities above ~40 μA, the 12C– beam appears not to be centered in the off-
axis low-energy cup when 13C– is injected into the acceleration tube. We are still searching for an
explanation for this phenomenon. At present, 12C– beam currents are being limited to ~30 μA in
order to achieve consistent operation, even though the ion source will deliver 100 μA comfortably.
Machine background and sample preparation blanks have also been studied. The failure of a turbo-
pump resulted in a small but measurable increase in the SSAMS machine background, emphasizing
the importance of high vacuum in these instruments. We have also investigated the blanks from dif-
ferent irons, and see significant differences. Because splits of the same graphite blanks can be mea-
sured on both the 14UD accelerator and the SSAMS, it is possible to disentangle the contributions
from sample preparation and from instrumental background. Results for ultrafiltered collagen from
ancient bison bones will be presented and discussed.
18 Abstracts

#26

A NEW 1 MV AMS FACILITY IN KIGAM (KOREA)

Wan Hong, Jung Hun Park, Hyung Joo Woo, Jun Kon Kim, Han Woo Choi, Gi Dong Kim
Korea Institute of Geoscience and Mineral Resources (KIGAM), 92 Gwahakno, Yuseong-gu, Daejeon 305-350, Korea.

A 1MV AMS system was installed at Korea Institute of Geoscience and Mineral Resources
(KIGAM) at the end of 2007. It occupies only a 7-m-long by 6-m-wide room. The installation of the
AMS took 4 months. It is expected that the running cost is much lower than the large AMS
machines. In addition, one of main differences of this system from 500 kV AMS is the capability to
measure not only 14C but also 10Be and 26Al, even though the background level of Be and Al is
slightly higher than larger AMS system. Our AMS system is still showing reasonable and acceptable
results of Be measurement for most application fields, except for a few cases needing a very low
background level.
This system has a 50-multitarget ion source and a 100-Hz bouncer system to measure several iso-
topes sequentially such as 12C, 13C, and 14C. The carbon beam of several tens of μA can be easily
obtained by the ion source. And beam transport yield of the accelerator tube was around 40% for 12C
isotope with the stripper gas condition of 2.1 × 10–2 torr. The counting system was composed of dE-
E detector using a gas ionization chamber with a 75-nm-thick Si3N4 window and a two-dimensional
data acquisition system.
Since terminal voltage is small, C2+ ions are measured for age dating. In this case, the lithium molec-
ular beam, Li22+, interferes the measurement the same as if using C4+ ions. The optimum combina-
tion of analyzing magnet and ESA set values is needed to obtain good sensitivity for carbon iso-
topes. Typical 14C/12C ratios of a background sample and NIST oxalic II are 2.5 × 10–15 and 1.30 ×
10–12, respectively. For the 10Be measurement, a Si3N4 film with the thickness of 150 nm is
employed to separate beryllium isotopes from boron isotopes. A typical 10Be/9Be ratio of a blank
was measured to be 1.63 × 10–14, and the linearity of activity and measured values relation obtained
using ICN standard samples was good.
This AMS system has been applied to geological environmental research since January 2008 after
test runs. As of 2008, around 1000 carbon samples have been measured with stable background and
oxalic acid levels. The performance of this system will be given in the conference.

#27

DEVELOPING ULTRA SMALL-SCALE RADIOCARBON SAMPLE MEASUREMENT AT MALT

AMS, TOKYO UNIVERSITY
Yusuke Yokoyama1,2,4, Mamito Koizumi1,2, Hiroyuki Matsuzaki3, Yosuke Miyairi3,
Naohiko Ohkouchi4
1. Ocean Research Institute, University of Tokyo.
2. Department of Earth and Planetary Science, University of Tokyo.
3. Department of Nuclear Engineering and Management, University of Tokyo.
4. Institute for Research on Earth Evolution, Japan Agency for Marine-Earth Sciences and Technology.

We have developed AMS measurement techniques for ultra small-size samples ranging 0.01 to 0.10
mg C with new type MC-SNICS(TM) ion source system. We could generate higher ion beam cur-
rent intensity for ultra-small samples by an optimization of graphite position in the target holder
 Abstracts 19

with the new ionizer geometry. CO2 gas from larger combusted standards was split to ultra-small
samples in the updated graphitization vacuum line. After conversion of CO2 gas to graphite, graph-
ite is pressed into a depth of 1.5 mm from the front of the target holder. This is much deeper than the
usual 0.35-mm-depth position. We measured 12C4+ beam currents generated by small standards
using a AMS system (5UD PelletronTM tandem van de Graaf, NEC) at the University of Tokyo. Ion
beam currents (15–30 μA) from the targets in optimized position lasted 20 min for 0.01 mg C and
65 min for 0.10 mg C. We observed that the measured 14C/12C ratios are unaffected by the difference
of ion beam currents ranging from 5 to 30 μA. This enables measurement of ultra small-size samples
with high precision. Examination of the background samples revealed 1.1 μg of modern and 1 μg of
dead carbon contaminations in the process of target graphite preparation. We make corrections for
both background components; however, reduction of the contamination is necessary for more accu-
rate measurement.

#28

STATUS REPORT FOR THE AMS FACILITY IN TRONDHEIM

Steinar Gulliksen and Einar Værnes
National Laboratory for 14C dating, Norwegian University of Science and Technology, NO-7491 Trondheim, Norway.

The Norwegian University of Science and Technology ordered in August 2005 a 1MV system from
HVE. Due to extremely unfortunate circumstances, the acquisition of a suitable site for the instru-
ment suffered severe and multiple delays. In September 2006, the components were delivered and
stored under stable conditions with regard to temperature and humidity in a submarine bunker built
during WWII. Installation was started at the end of January 2009.
As a matter of fact, it can be argued that maturing processes during storage have improved the
instrument; some important modifications based on operational experiences from other facilities
have been implemented by HVE.
The first results will be presented.

#29

THE ERLANGEN AMS FACILITY: STATUS, APPLICATIONS, AND THE PROSPECT OF 14C

MEASUREMENTS OF MICROGRAM SAMPLES

Wolfgang Kretschmer, Andreas Scharf, Stefanie Gierl, Katrin Leichmann, Alexander Stuhl,
Andreas Wiedenhofer
Physics Institute, University of Erlangen-Nuremberg, Germany.

The Erlangen AMS facility consists of a negative ion source for solid and gas samples, an injection
spectrometer with 90° electrostatic deflector and 90° magnet with fast isotope switching, an EN tan-
dem accelerator, an analyzing spectrometer with 15° electrostatic deflector, 55° magnet, and a heavy
ion detector. The combustion of chemically pretreated samples to CO2 is performed in an elemental
20 Abstracts

analyzer (EA), followed by a stable mass spectrometer for precision measurement of δ13C from 1%
of the CO2. The CO2 outlet of the EA is connected to a tenfold reduction unit for the production of
carbon sputter targets. The use of an EA has the big advantage that both combustion and reduction
are connected in a closed vacuum line with minimum contamination from outside, and that both pro-
cesses can be performed in a fully automated and time-saving way. For a typical 1-mg sample, the
background from procession and machine is about 0.08 pMC, corresponding to an apparent age of
59.000 years. To extend the range of applications beyond our standard samples (charcoal, soil, pol-
len, bones, mussels, etc.), attempts have been made for the radiocarbon dating of iron artifacts and
lime mortar, which involves different pretreatment techniques.
For applications concerning the biogenic or anthropogenic origin of environmental compounds the
amount of carbon for radiocarbon measurements can be extremely low which makes an extension of
the carbon mass range down to microgram highly desirable. To minimize any carbon contamination,
we have developed a closed system from gas chromatograph, used for fraction separation, via EA
and a fully automatic gas handling system to the sputter ion source. With this new method, modern
and old calibration samples of monochloroacetic acid (MCA) with carbon masses down to 10 μg
have been measured with reasonable results.

Parallel Session 2-2: Pacific Studies

#30

CALIBRATED RADIOCARBON AGES OF JOMON ISEKI, NE JAPAN, AND THEIR SIGNIFI-

CANCE
Kunio Omoto1, Kenji Takeishi1, Shigeru Nishida2, Junichi Fukui2
1. College of Humanities and Sciences, Nihon University. Setagaya, Tokyo 156-8550, Japan.
2. Inquiry Section, Hokkaido Archaeological Operations Center, Ebetsu, Hokkaido, 069-0832, Japan.

Archaeological chronology in the Japanese Islands during the Jomon period is essentially based on
the relative age given to straw-rope patterns marked on earthenware. Moreover, many Japanese
archaeologists still believe their own relative chronology instead of using calibrated radiocarbon
ages. However, some archaeologists recently have begun to use AMS data.
We aimed to correlate calibrated radiocarbon ages with classical archaeological chronology and to
reconstruct new chronology of the Jomon period. We collected shell samples from nine shell
mounds and one human ruins in Hokkaido Island, as well as from four shell mounds in the Tohoku
district. The archaeological sites investigated were Tokoroasahi, Tenneru, Kitakogane, Ponma,
Takasago, and Hokuto Iseki on Hokkaido Island and Sakiyama, Satohama, Nishihata, and Daigika-
koi Iseki on NE Honshu Island, respectively.
A total of 39 sample materials were radiocarbon dated and their isotope fractionations were mea-
sured. The figures of the local reservoir collection (ΔR) on Hokkaido Island reached 300~500 yr for
marine samples, indicating a strong influence of the Oyashio or Liman currents and an apparently
larger local reservoir collection than that of SW Japan (Nakamura et al. 2007). The calibrated radio-
carbon ages ranged between ca. 6000 and 3000 yr, correlative with the Early Jomon and the Last
Jomon stages as indicated by classical earthenware chronology of Honshu Island.
 Abstracts 21

#31

THE IMPACT OF OCEAN BOUNDARIES AND ISLAND CHAINS ON SHELL 14C DETERMINA-
TIONS: KNOWN-AGE AND ARCHAEOLOGICAL ΔR EXAMPLES FROM THE BISMARCK SEA
AND HAWAIIAN ISLANDS
Fiona Petchey
Waikato Radiocarbon Dating Laboratory, University of Waikato, Private Bag 3240, Hamilton, New Zealand.

The use of the modeled marine reservoir curve with associated regional offsets (known as ΔR) has
been widely accepted as the best method for calibrating marine shell 14C determinations (Stuiver et
al. 1998). The application of these ΔR values to marine shell calibrations, however, requires an
understanding of the habitat and dietary preferences of the shellfish concerned, as well as detailed
knowledge of regional ocean circulation patterns. In particular, recent research has indicated that
considerable ΔR variation is possible where ocean currents come into contact with continental land-
masses, opposing currents, or island chains (Petchey et al. 2008). Two regions that have been iden-
tified as problematic in this respect are the Hawaiian Islands and the Bismarck Sea.
In Hawaiian waters, interaction between the island chain, the North Equatorial Current and the
northeasterly tradewinds cause large-scale eddies and upwelling in the lee of the islands. Petchey
(2009) has speculated that these conditions may, in part, be responsible for the range in ΔR values
(from –29 ± 4 to 280 ± 80 14C yrs) observed by Dye (1994) in shells collected pre-AD 1950. ΔR val-
ues ranging from 365 ± 50 to 43 ± 68 14C yrs have also been recorded around New Ireland and the
Bismarck Sea. Petchey et al. (2004) suggested that this spread was caused by seasonal reversals in
the South Equatorial Current and North Equatorial Counter Current system in combination with
weak equatorial upwelling (see also McGregor et al. 2008). Radiocarbon data from archaeological
sites and marine shells collected in the early 20th century have been obtained to further investigate
these hypotheses. Preliminary results of this research are presented here.

REFERENCES
Dye T. 1994. Apparent ages of marine shells: implications for archaeological dating in Hawaii. Radiocarbon 36(1):51–57.
McGregor HV, Gagan MK, McCulloch MT, Hodge E, Mortimer G. 2008. Mid-Holocene variability in the marine 14C reservoir
age for northern coastal Papua New Guinea. Quaternary Geochronology 3:213–225.
Petchey F, Phelan M, White P. 2004. New ÄR values for the southwest Pacific Ocean. Radiocarbon 46(2):1005–1014.
Petchey F, Anderson A, Zondervan A, Ulm S, Hogg A. 2008. New marine ÄR values for the South Pacific Subtropical Gyre
region. Radiocarbon 50(3):373–397.
Petchey F. 2009. Dating marine shell in Oceania: issues and prospects. In: Fairbairn A, O’Connor S and Marwick B (eds),
New Directions in Archaeological Science. Terra Australis, 28: Canberra: Pandanus books, Research School of Pacific and
Asian Studies, ANU.
Stuiver M, Reimer PL, Braziunas TF. 1998. High-precision radiocarbon age calibration for terrestrial and marine samples.
Radiocarbon 40(3):1127–1154.

#32

LATE HOLOCENE ΔR VALUES FOR HAWAII DERIVED FROM U-TH DATED ARCHAEOLOGI-
CAL CORALS
Quan Hua1, Marshall Weisler2, Jian-xin Zhao3
1. Australian Nuclear Science and Technology Organisation, Menai, NSW 2234, Australia.
22 Abstracts

2. School of Social Science, University of Queensland, Brisbane, QLD 4072, Australia.

3. Radiogenic Isotope Laboratory, University of Queensland, Brisbane, QLD 4072, Australia.

We present the first application of U-series dating and AMS radiocarbon assay of Polynesian
archaeological Pocillopora spp. branch corals for the investigation of possible temporal and spatial
variations in local marine reservoir corrections (ΔR). Our study area is Moloka‘i Island (Hawaiian
Archipelago) where archaeological surveys and excavations have been conducted over the 20,000-
hectare region during the past 20 years. Twelve samples of TIMS U-Th dated corals spanning the
entire cultural-historic sequence for the Hawaiian Islands (about AD 700–1800) were used for AMS
radiocarbon analysis. Six corals, dated to within ~100 years by U-Th, were sampled from 3 coast-
lines of west Moloka‘i with different offshore conditions to examine spatial variability in ΔR. The
remaining 6 samples were collected from the south coast, near the west end of Moloka‘i.
High-precision TIMS U-Th dating was performed at the Radiogenic Isotope Laboratory, University
of Queensland, and AMS 14C measurement was carried out using the STAR facility at ANSTO. The
14C results indicate that there is no significant temporal variability in ΔR between about AD 700 to

1800 for Moloka‘i Island and the weighted average ΔR value for that period is 52 ± 25 yrs. In com-
bination with 4 selected previously published ΔR values based on pre-bomb known-age marine
shells, a revised ΔR of 66 ± 54 yrs is established for the Hawaiian Islands. Our results also show no
spatial variability in ΔR during the 17th century for Moloka’i Island. However, future research
should examine the archipelago-wide spatial variability in ΔR with the analysis of additional dated
archaeological coral samples.
This work was supported by Ainse grants 06/190 & 08/068.

#33

SPORORMIELLA AND THE NEW ZEALAND AVIFAUNAL EXTINCTIONS

Jamie R Wood and Janet M Wilmshurst
Landcare Research, PO Box 40, Lincoln 7640, New Zealand.

Spores of the coprophilus fungi genus Sporormiella are a palynological proxy for vertebrate herbi-
vore biomass, and have provided evidence for the timing of megafaunal extinctions on several con-
tinents and islands across the globe. Until now, Sporormiella have not been used in New Zealand
paleoecological studies, yet they offer the possibility of answering several interesting questions
relating to the extinctions that followed human settlement. For example, Sporormiella spores may
allow accurate testing of the temporal linkage between herbivore extinctions and anthropogenic
deforestation, providing new insights into the order and process of human impacts on the fauna and
flora following initial settlement in the late 13th century. Previously, large age ranges produced by
calibrations of radiocarbon dates in this period have masked the relative timing and order of extinc-
tion and deforestation. We present preliminary results of Sporormiella counts from New Zealand
peat cores, and discuss the implications of the spore abundance curves for understanding the timing
of, and roles of, hunting and habitat loss in New Zealand’s avifaunal extinctions.
 Abstracts 23

#34

BOMB RADIOCARBON DATING OF WHITE ABALONE (HALIOTIS SORENSENI): INVESTIGA-

TIONS OF AGE, GROWTH, AND LIFESPAN
Allen H Andrews1, Gregor M Cailliet1, Robert T Leaf2, Thomas Brown3, Laura Rogers-Bennett4,
Kenneth H Coale1
1. Moss Landing Marine Laboratories - Age and Longevity Research, California State University, 8272 Moss Landing Road,
Moss Landing, CA 95039, USA.
2. Department of Fisheries and Wildlife Sciences, Virginia Tech, 101 Cheatham Hall, Blacksburg, VA 24061, USA.
3. Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, CA
94551, USA.
4. Bodega Marine Laboratory, California Dept. Fish and Game, 2099 Westside Road, Bodega Bay, CA 94923, USA.

Bomb radiocarbon dating was used to answer unresolved questions about age, growth, and lifespan
of white abalone (Haliotis sorenseni). Abalone species in central and southern California have suf-
fered population declines resulting in the closure of both commercial and recreational fisheries. Of
the seven Haliotis species in California, white abalone is perhaps the most threatened and was the
first marine invertebrate listed as an endangered species under the federal Endangered Species Act.
An understanding of life history information is critical to restoring the species from its extirpated
state and will be necessary, if the population can recover, for harvest management. A significant
issue in this regard is to determine accurate age and growth parameters that describe ontogenetic
growth and to estimate lifespan. It is possible that remnant white abalone populations may be com-
posed of senescent individuals that are incapable of contributing to population growth, and that nat-
ural mortality may take the remaining population due to recruitment failure. Age, growth rates, and
lifespan were estimated in previous studies from limited observations of growth in the field and cap-
tive environments. Growth was described as initially rapid, slower with increasing size, and highly
variable. From those observations, age near maximum size (~210 mm MSL) was extrapolated to
~35 years; however, these estimates of age, growth, and lifespan remain unconfirmed. Bomb radio-
carbon dating has provided well-constrained age determinations from a series of white abalone
shells with known collection dates. In this work, a series of shells with a range of lengths (148–200
mm MSL) and collection years (1952–1996) were used in concert with a reference bomb radiocar-
bon data series to examine growth parameters for this species. The lifespan of a large shell (193 mm
MSL) was confirmed at 24 ± 2 years, and a larger shell (200 mm MSL) may have lived 29 ± 2 years.
These independent estimates of age provide support for early age and growth estimates, and greater
confidence in the development of growth parameters necessary for population recovery and respon-
sible management of white abalone.

Parallel Session 3-2: Radiocarbon & AMS - Small and Large II

#35

DATING MICADAS AT THE NEW AMS LABORATORY AT CEZA, MANNHEIM, GERMANY

B Kromer1, H-A Synal, L Wacker, I Levin, E Pernicka
1. Heidelberg Academy of Sciences, INF 229, Heidelberg, Germany, D-69120, Germany.

We present a status report on the newly founded AMS laboratory (Klaus Tschira Laboratory, KTL)
at the Curt-Engelhorn Centre for Archaeometry, CEZA, Mannheim, Germany. The AMS machine is
24 Abstracts

based on the MICADAS concept (200kV terminal voltage). It will be the first one of this design
operated as dating facility outside the ETH Zurich. For more than a year, we investigated procedures
to lead to a high performance, mainly with respect to dating precision and long-term stability. We
compared modern 14CO2 samples, measured by gas counting and AMS. Our results indicate that the
Dating MICADAS system is capable of routine analyses at the 2–3‰ level in unattended operation.
Background levels (in the 48–50 kyr 14C age range) are considered completely adequate for the
range of our projects.

#36

RADIOCARBON AND RADIONUCLIDE STUDIES USING TWO DIFFERENT AMS MACHINES:

ADVANTAGES AND COMPLICATIONS
A J T Jull, J W Beck, D L Biddulph, G S Burr, R Cruz, N A Lifton, J Quade
NSF Arizona AMS Laboratory, Departments of Geosciences and Physics, University of Arizona, Tucson, AZ 85721, USA.

At the Arizona laboratory, we have 2 operational accelerator mass spectrometers (AMS). A large
project was a thorough overhaul of our “original” machine, a General Ionex Tandetron, including
installation of new HVEE accelerator tubes. Our second machine, a 3MV NEC Pelletron AMS, has
had a high level of operational time over the last 2 years. We have also instituted programs in 129I
and 26Al measurements, in addition to 14C and 10Be. We introduce some improvements in AMS sys-
tem and chemical processing features as well as some interesting studies in 14C dating of corals, spe-
leothems, and bones; 129I studies; new records of 10Be from marine sediments and extraterrestrial
materials; and 26Al measurements on extraterrestrial materials.
In collaboration with Quade et al., we have developed new low-level 14C extraction and graphitiza-
tion systems at the University of Arizona. We have also identified sources of variable radiocarbon
blanks from both chemical and instrumental sources. The recent successful construction of these
systems has reduced laboratory-induced contamination during the combustion and graphitization
phases of sample preparation and has demonstrated that it is possible to extend the 14C dating hori-
zon from about 48,000 to about 60,000 years before present. This advance can only be exploited if
it is coupled with improvements in the chemical decontamination of samples prior to combustion.
In addition, these studies have identified some previously unexpected sources of background in AMS
systems. We are able to identify these effects by using the same target material run under different
conditions on different machines. We believe these sources of background, apparently from molec-
ular interferences, have not been well-quantified in the past and have been the source of problems in
some other AMS systems. We discuss our approach to resolve these issues at the Arizona laboratory.
 Abstracts 25

#37

THE LOW-ENERGY ISOBAR SEPARATOR FOR ANIONS – PROGRESS REPORT

W E Kieser1, J Eliades1, A E Litherland1, X-L Zhao1, L Cousins2, S J Ye2
1. IsoTrace Laboratory, University of Toronto, Ontario, Canada.
2. Ionics Mass Spectrometry Group, Bolton, Ontario, Canada.

The suppression of interferences from atomic isobars is a key requirement for the extension of accel-
erator mass spectrometry (AMS) to the analysis of new cosmogenic isotopes and for increasing the
range of applications for small AMS systems. In earlier work, we have shown that unwanted isobars
can be eliminated by anion-gas reactions [1]. Recently, we described a prototype system, the Isobar
Separator for Anions (ISA), which decelerates the beam of rare anions from an AMS ion source to
eV energies, guides them through a radiofrequency quadrupole (RFQ) gas cell, and re-accelerates
them for further analysis in a 2.5MV AMS system. Demonstrations of this system for the analysis
of 36Cl showed a suppression of 36S by over six orders of magnitude, with the transmission of ~30%
of the 36Cl beam [2]. Since that time, tests with a range of standard reference materials have shown
the linearity of the system for measuring the 36Cl/35Cl ratio over a span of two orders of magnitude.
Further tests, using the AMS system as a diagnostic tool, have provided clues to the nature of the
remaining low level of 36S transmission. Tests for the suppression of atomic isobars for other species
of interest in AMS have been carried out, for example, the suppression of 41K in the measurement
of 41Ca. A general description of the ISA system and the results of these recent tests will be pre-
sented.
[1] A. E. Litherland et al. 2007. Nuclear Instruments & Methods B 259:230.
[2] J. Eliades et al. 2008. 11th Int. Conf. on Accelerator Mass Spectrometry, Rome, Sept 2008, submitted to Nuclear Instru-
ments & Methods B.

#38

A BEAM PROFILE MONITOR FOR RARE ISOTOPES IN AMS MEASUREMENTS

F Taccetti, M E Fedi, P Mariani, L Carraresi, P A Mandò
INFN Sezione di Firenze e Dipartimento di Fisica dell’Università di Firenze, via Sansone 1, 50019 Sesto Fiorentino (Fi),
Italy.

In accelerator mass spectrometry (AMS), it is usually difficult to monitor the beam shape of the rare
isotopes due to their small current. A beam profile monitor (BPM) based on a multiwire propor-
tional chamber (MWPC) has been developed at the LABEC laboratory in Florence. The BPM con-
sists of a central cathode between two grids of anode wires, oriented perpendicular to each other.
The active volume of the BPM is filled with heptane, continuously flowing. Tests were performed
using the pulsed beam IBA facility present at LABEC. Carbon beams (12C ions), 10 MeV energy,
were produced and used for tests. The ions rate was chosen to range between 1 and 200 Hz to repro-
duce the same condition of AMS measurements. Tests show a good spatial resolution (0.5 mm
FWHM) and excellent long-term stability. A short description of the BPM operating principles and
of the front-end electronics will be also given.
26 Abstracts

#39

ΔE-Q3D ISOBAR IDENTIFICATION METHOD IN AMS MEASUREMENT

Ming He, Wei Wang, Chaoli Li, Jie Gong, Shan Jiang, Hongtao Shen, Xianggao Wang,
Kejun Dong, Shaoyong Wu
China Institute of Atomic Energy, P.O. Box 275(50), Beijing 102413, China

In AMS measurement, isobar identification is a key point. Many techniques were developed for iso-
bar identification, such as the ionization chamber, gas-filled magnet, projectile X-ray, and so on.
Recently, the ΔE-TOF method was suggested for isobar identification. Now, we want to develop a
new isobar identification method based on Q3D magnet spectrometry. In our lab, Q3D magnet spec-
trometry has very high momentum resolution power (0.01%). So, after isobars passi through a suit-
able thickness foil, located at the entrance of the Q3D magnet, they will lose energy (ΔE) in the foil.
The energy loss (ΔE) is different for the isobars; thus, Q3D magnet spectrometry can separate the
isobars according to their different energies. This is the idea of ΔE-Q3D. In order to further identify
isobars, a multi-anode ionization chamber will be set at the focal plane of the Q3D magnet spec-
trometer. In our lab, there is a big tandem accelerator, which can be set for a terminal voltage of
12MV. According to our calculations, this ΔE-Q3D method can separate isobars at atomic numbers
of about 45. Thus, this is effective method for the measurement of 63Ni, 79Se, 94Nb, etc. This method
also can be used to measure 53Mn and 32Si.
This work is supported by the National Natural Science Foundation of China (No.10875176).

Parallel Session 2-3: Archaeological Applications

#40

THE ARTEMIDORUS PAPYRUS: SOLVING AN ANCIENT PUZZLE WITH RADIOCARBON AND

ION BEAM ANALYSIS MEASUREMENTS
M E Fedi1, L Carraresi1, N Grassi1, A Migliori1, F Taccetti1, F Terrasi2, P A Mandò1
1. INFN Sezione di Firenze e Dipartimento di Fisica dell’Università, via Bruno Rossi 1, 50019 Sesto Fiorentino (Fi), Italy.
2. Dipartimento di Scienze Ambientali, Seconda Università di Napoli, via Vivaldi 43, 81100 Caserta, Italy.

Ancient papyrus manuscripts are one of the most fascinating sources for reconstructing not only
ancient life habits but also past literature. In recent years, an amazing document has come to the fore
due to the heated debates it raised: the so-called Artemidorus papyrus. It is a very long roll (about
2.5 meters), composed of several fragments of different dimensions, with inscriptions and drawings
on both sides. On the recto of the document, a text about geography and some drawings of heads,
feet, and hands are present, while on the verso there are many sketches of animals, both real and fan-
tastic. Its importance in classical studies comes from the fact that some scholars claim that it is the
first known transcription of a relatively large fragment of the Greek geographer Artemidorus. On the
other hand, other scholars think that the papyrus is a fake, drawn in 19th century AD by a well-
known forger. In order to overcome all possible ambiguities, the papyrus has been studied not only
on the basis of historical and paleographic criteria but also by scientific techniques. We have con-
tributed to deepening the knowledge about the papyrus by radiocarbon dating the support and by
analyzing the composition of the ink using ion beam analysis (IBA). Results are compatible with the
 Abstracts 27

originality of the roll as an ancient manuscript: AMS radiocarbon measurements have dated the
papyrus to a period between 1st century BC and 1st century AD; IBA measurements have pointed
out the use of an organic (carbon-based) ink, which was typical of ancient Roman and Greek times.
Details of the measurements are presented to emphasize the importance of combining AMS and
IBA results.

#41

MARCO POLO AND THE DISCOVERY OF THE AMERICAS: RADIOCARBON DATING THE
ROSSI MAP-WITH-SHIP
Greg Hodgins1 and Eric Hansen2
1. NSF-Arizona AMS Laboratory, Department of Anthropology, University of Arizona, Tucson, Arizona, USA.
2. Department of Preservation Research and Testing, Library of Congress, Washington DC, USA.

Library of Congress has within its collection a parchment document known as the Rossi Map-with-
Ship. The document is associated with the early Renaissance explorer Marco Polo, who traveled the
Far East between AD 1275 and 1295. The document is significant because it contains a map of the
Pacific Rim, including the coast of China and eastern Siberia, the Bering Strait, the Aleutian Islands,
Alaska, and the West Coast of North America all in distorted but recognizable form. Adjacent the
map are drawings of a Venetian sailing vessel, and the Polo family coat-of-arms, notes written in a
Venetian dialect of Italian, and toponyms in Arabic and Chinese scripts. Parchment sampled from
the Map-with-Ship was carbon dated to the periods AD 1463 to 1526 and AD 1556 to 1633 at the
95% confidence interval. Consequently, the association with Marco Polo cannot be direct. However,
the document remains intriguing for a variety of reasons. For example, if the parchment date reflects
the age of the map, it predates European exploration of the Bering Strait and the coast of Alaska by
nearly a century. Moreover, several lines of evidence suggest it might be based upon Chinese rather
than European knowledge of geography. The map was originally part of a collection of 14 related
documents, 11 of which survive in private hands. The interrelationship of these documents and sto-
ries they tell beg further inquiry.

#42

EARLY BRONZE AGE CHRONOLOGY DEDUCED FROM RADIOCARBON AGES OF REMAINS

FROM TELL GHANEM AL-ALI SITE, RAQQA, SYRIA
Toshio Nakamura1, Mitsuo Hoshino2, Tsuyoshi Tanaka2, Hidekazu Yoshida3, Takeshi Saito4,
Kazuhiro Tsukada3, Yusuke Katsurada5, Yoshiyuki Aoki2, Tomoko Ohta1, Atsunori Hasegawa6,
Katsuhiko Ohnuma7, Anas Al Khabour8, Michel Al Maqdissi8
1. Center for Chronological Research, Nagoya University, Japan.
2. Graduate School of Environmental Research, Nagoya University, Japan.
3. University Museum, Nagoya University, Japan.
4. Faculty of Science and Technology, Meijo University, Japan.
5. Research Fellow, Nagoya University, Japan.
6. Graduate School of Humanities and Social Sciences, University of Tsukuba, Japan.
28 Abstracts

7. The Institute for Cultural Studies of Ancient Iraq, Kokugakuin University, Japan.
8. Syrian Directorate General of Antiquities and Musems, Syria.

We have conducted a field survey on archaeological sites along the middle Euphrates, as well as
geological environment and archaeological remains in the Bishri region, southeast of Raqqa, Syria,
in the Near and Middle East. The main aim of this project is to conduct chronological analyses of the
archaeological sites there and establish the formation period of the terraces on which the sites are
located, by 14C dating with AMS at Nagoya University.
During the survey, we have visited the Tell Ghanem al-Ali site that is located on the lowest terrace
of the Euphrates River, where extensive archaeological excavations have been performed at two
locations in the northeast side of the tell, Square-1 (9 × 9 m2) and Square-2 (3 × 15 m2). We have
detected several fragments of pottery as well as black layers composed of charred soil and charcoal
fragments. The layers were quite clear and we can believe that these layers have resulted from human
activities at the site. We have collected in total 19 samples such as charcoal and wood fragments from
a Square-1 trench and adjacent outcrop to the site. We also collected some charcoal and organic soil
samples to investigate the formation age of the lowest terrace on which the site is situated.
14C ages for 18 samples related with human habitation at the site are concentrated in a range from
4001 ± 30 to 4114 ± 35 BP, calibrated dates from 2850 to 2500 cal BC, 1σ error. These results are
almost consistent with an archaeological view on chronology of Early Bronze Age period estimated
based on typological analysis. 14C ages for samples collected from the sediments that formed the
lowest terrace ranged from 927 ± 25 to 4494 ± 37 BP. These ages are informational to estimate the
formation period of the lowest terrace.

#43

A NEW 14C CHRONOLOGY FOR NEW KINGDOM EGYPT

M W Dee1, A J Shortland, T F G Higham, J M Rowland, F Brock, S Harris, C Bronk Ramsey
1. University of Oxford, RLAHA, Oxford, Oxfordshire, OX1 3QY, United Kingdom.

The effectiveness of 14C dating for Egyptian chronology has come under significant criticism in
recent years. Indeed, the agreement of scientific dates with the established historical record has at
times been unsatisfactory. Further, the broadness of calibrated 14C dates has led some Egyptologists
to doubt the utility of the method for disentangling the intricate chronological issues of the Dynastic
era. On the basis of historical analysis, one of the most securely dated periods of the entire chronol-
ogy is the New Kingdom (1550–1069 BC), considered to be accurate to within 10–20 years. Hence,
it represents the definitive test for the performance of 14C dating in Egypt. The findings presented
here comprise more than 100 new dates made on short-lived plant samples that were able to be his-
torically allocated to individual reigns of the New Kingdom. The dates were modeled using the very
latest Bayesian techniques, and the outputs subsequently compared with the traditional Egyptian
chronology.
 Abstracts 29

#44

DATING THE COLLAPSE OF THE FUNERARY TEMPLE OF AMENHOTEP III AT LUXOR WEST
BANK, EGYPT
Mohammed Mahran1, Arkadi Karakhanyan2, Michel Wuttmann3, Hourig Sourouzian4
1. Institut français d’archéologie orientale (IFAO), Cairo, Egypt.
2. The Colossi of Memnon and Amenhotep III Temple Conservation Project.
3. Institut français d’archéologie orientale (IFAO), Paris, France.
4. Director of The Colossi of Memnon and Amenhotep III Temple Conservation Project.

The funerary temple of Amenhotep III at Luxor—built under the reign of Amenhotep III during the
18th dynasty (first half of the 14th century BC), the highest, most prosperous, and peaceful period
of the ancient Egyptian civilizatio—was the largest temple of all the area. Having collapsed during
an earthquake, this once prestigious temple complex had served as quarry for surrounding temples
and other constructions until modern times. Stones and bricks had been quarried away. Later, more
rubble and Nile alluvia have covered the original temple ground level up to 3 m. To confirm the dat-
ing of the strong earthquake that caused the collapse of the temple, 27 radiocarbon measurement
were made on soil samples taken from several of the rubble and destruction layers. This paper pre-
sents the interpretation of the results combined with the geomorphological studies and proposes a
date for the earthquake which destroyed the temple.

Parallel Session 3-3: Bayesian Analysis and Databases

#45

ROBUST BAYESIAN ANALYSIS, AN ATTEMPT TO IMPROVE BAYESIAN SEQUENCING

Franz Weninger, Peter Steier, Walter Kutschera, Eva Maria Wild
Vienna Environmental Research Accelerator (VERA), Faculty of Physics, Isotope Research, University of Vienna,
Währinger Straße 17, A-1090 Vienna, Austria.

Bayesian sequencing has become a generally accepted and very successful tool to reduce the uncer-
tainty of radiocarbon measurements, when additional information (the so-called “prior informa-
tion”) about the temporal relationship within a series of radiocarbon dates from an archaeological
context exist. However, there is a well-known fundamental problem in applying Bayesian statistics:
The archaeological evidence often does not provide enough information to unambiguously define a
prior function. Nevertheless, in the standard approach, a specific prior function is selected in a
canonical way. The gaps in the prior knowledge are filled with assumed information, which is
intended not to influence the final result. This may, however, not necessarily be the case. A possible
way to overcome this uncertainty is the use of several different prior functions that are in agreement
with the archaeological information, and then superimpose the results; this is called “robust Baye-
sian analysis.”
We will demonstrate the difference between the usual Bayesian sequencing and the robust analysis
with a few illustrative examples (e.g. the Iceman and his axe). Our results indicate that it is possible
to establish robust analysis for practical applications. As a general result, robust Bayesian analysis
30 Abstracts

leads to a higher accuracy, however paid for with a reduced precision. Even though some method-
ological shortcomings of the robust analysis are not completely solved, we will present a compari-
son with the usual method in a more elaborate case, where a long sequence of radiocarbon dates in
a secure archaeological context is available.

#46

USING CAR4AMS, THE BAYESIAN AMS DATA-ANALYSIS CODE

V Palonen, P Tikkanen, J Keinonen
Department of Physics, P.O.B. 43, FIN-00014 University of Helsinki, Finland.

The Bayesian CAR model for AMS data analysis delivers uncertainties with less scatter and bias
[1]. Better detection and estimation of the instrumental error of the AMS machine are also achieved.
Presently, the main disadvantage is the several-hour duration of the analysis. CAR4AMS, a MCMC
program for CAR-model analysis, has been made freely available under the GPL license [2].
Included in the package is an R program, which analyzes the CAR4AMS output and transfers the
information to graphical and spreadsheet formats. We describe the main properties of the CAR anal-
ysis and the use of the two parts of the program package for radiocarbon AMS data analysis.
[1] V. Palonen. 2008. Accelerator mass spectrometry and bayesian data analysis [PhD thesis]. University of Helsinki Report
Series in Physics D, HU-P-D 148. Available at http://urn.fi/URN:ISBN:978-952-10-3262-2.
[2] V. Palonen, P. Tikkanen, J. Keinonen. CAR4AMS: A program for Bayesian analysis of accelerator mass spectrometry data.
Computer Physics Communications, submitted. Available at http://beam.acclab.helsinki.fi/~vpalonen/car4ams.htm.

#47

A NEW ATTEMPT TO ESTABLISH THE INTERNATIONAL RADIOCARBON SOILS DATABASE

(IRSDB)
Peter Becker-Heidmann
Institut für Bodenkunde, Allende-Platz 2, Hamburg 21465, Germany.

Twenty years after the first International Radiocarbon Database Workshop, and 13 years after the
setup of a preliminary structure for a worldwide database on radiocarbon dates of soils, sound rea-
sons and excuses for not establishing a real and globally accessible database have diminished. Cli-
mate change itself is widely accepted as reality now, and the strong demand of the modeling com-
munity for reliable data of the carbon pool—especially in soils—has been maintained. With the
steadily increasing capacity of radiocarbon dating facilities, these data can be and are produced.
Nevertheless, they still cannot be accessed easily and equally enough. Now, decreased costs of hard-
ware and the recent developments of the internet enable the IRSDB to be implemented, in a joint
effort. As a seed, a test server has been set up, with open-source software, housing the database in
alpha-stage, a web interface and a community portal. Thus, the development of the design as well as
the data input of the database is done in close collaboration of the users of the database, the labora-
tories, soil scientists, archaeologists, modelers, other scientists, and interested lay persons. In order
to guarantee the long-time independence of the availability and usability of the database from ven-
 Abstracts 31

dors or changing standards, only widely used open-source software and open standards are used.
Therefore, the development of plugins for data input from laboratory databases or output to different
required formats as well as interfaces to GIS and other software is possible. A version control sys-
tem takes care of the integrity of the data. The actual stage of implementation of the IRSDB is pre-
sented and demonstrated, and the prospects of the development are discussed.

#48

ZAGRADA – THE NEW ZAGREB RADIOCARBON DATABASE

A Portner, B ObeliÊ, I Krajcar BroniÊ
Rudjer BoškoviÊ Institute, BijeniËka 54, HR-10000 Zagreb, Croatia
In the Radiocarbon and Tritium Laboratory at the Rudjer BoškoviÊ Institute in Zagreb, Croatia,
three different techniques for 14C dating have been used: gas proportional counting (GPC); liquid
scintillation counting (LSC), both CO2 absorption and benzene synthesis method; and preparation
of milligram-sized samples for accelerator mass spectrometry (AMS) dating. The use of several
measuring techniques has initiated a need for development of a new relational database
ZAGRADA (Zagreb Radiocarbon Database) since the existing software package CARBO,
designed in 1988 for the GPC technique only, could not satisfy the requirements for parallel pro-
cessing of several techniques.
Using the SQL procedures, and constraints defined by primary and foreign keys, ZAGRADA
enforces high data integrity and provides better performances in data filtering and sorting. The struc-
tural scheme of this database conceptually comprises 5 basic modules with data about samples,
chemical pretreatment and preparation, measurements, results and data on standards. Additionally,
the new database for 14C samples is a multi-user oriented application that can be accessed from
remote computers in the workgroup, thus providing better efficiency of laboratory activities. In
order to facilitate data handling and processing in ZAGRADA, the graphical user interface is
designed to be user-friendly and to perform various actions on data like input, corrections, search-
ing, sorting and output to printer. A new software package also provides the capability to export data
regarding certain samples to an .xls file, which thereafter can be used as a custom-made report doc-
ument in further correspondence with customers. All invalid actions performed in the user interface
are registered with a short textual description of an error occurred and appearing on screen in mes-
sage boxes. Unauthorized access is also prevented by login control and each application window has
implemented support to track last changes made by the user. The implementation of a new database
for 14C samples has significant contribution to scientific research performed in the Radiocarbon and
Tritium Laboratory and to quality assurance and quality control, and will provide better and easier
communication with customers.
32 Abstracts

Parallel Session 2-4: Marine Studies

#49

AN IMPROVED RADIOCARBON CHRONOLOGY FOR MD-982167, EASTERN INDIAN OCEAN:

SYNCHRONIZING PALEOMAGNETIC INTENSITY AND 10Be ACROSS THE LASCHAMP EVENT
Vladimir Levchenko1, David Fink1, Bradley Opdyke2, Charles Mifsud1, Alan Williams1,
Chris Klootwijk2
1. Australian Nuclear Science and Technology Organisation [ANSTO], PMB 1, Menai NSW 2234, Australia.
2. Australian National University, Research School of Earth Sciences, ACT 0200, Australia.

The production rate of cosmogenic isotopes 10Be and 14C is influenced by variations in the primary
cosmic-ray flux and by changes of the Earth’s magnetic field, which has experienced a number of
significant perturbations during the late Quaternary (e.g. Laschamp excursion, ca. 41 kyr BP).
Understanding these changes and synchronizing variations in paleomagnetic intensity derived from
various marine sediment cores can often help constrain the quality of 14C dating, particularly for
times beyond IntCal04 between 30 and 50 kyr ago, leading to improved overall core chronologies
and determination of ocean sedimentation rates. An often used proxy for these paleomagnetic
changes is the concentration of atmospheric fallout of 10Be in marine sediments. The ocean core
MD-982167 recovered from the Scott Plateau in the eastern Indian Ocean (13°09′S, 121°35′E;
depth of 1981 m) has an established δ18O stable isotope stratigraphy and a high sedimentation rate
of 10 to 20 cm/kyr based on a preliminary chronology tied to ice-core records. This in turn makes
MD-982167 amenable to high-resolution 10Be assay across the Laschamp period. We have com-
pleted measurements of 10Be and 14C (in both foraminifera and fine sediments) and paleomagnetic
intensity in core MD-982167. We are able to correlate variations in 10Be with paleomagnetic inten-
sity recorded in these ocean sediments based on our detailed radiocarbon age-depth profile and con-
firm a strong presence of the Laschamp excursion at 39.9 ± 1.3 kyr BP. Comparing the variations of
atmospheric-derived 10Be in MD-982167 tied to our improved radiocarbon chronology with 10Be
established from the ice cores and will allow much better correlation between marine sedimentary
and ice-core records across recent stadial and interstadial climatic events.

#50

SPATIAL VARIATION IN THE MRE THROUGHOUT THE SCOTTISH POST-ROMAN TO LATE

MEDIEVAL PERIOD: NORTH SEA VALUES (500–1350 BP)
N Russell1, G T Cook1, P L Ascough1, A J Dugmore2
1. SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, Scotland, UK.
2. Institute of Geography, School of Geosciences, University of Edinburgh, Drummond St., Edinburgh EH8 9XP, Scotland,
UK.

The marine 14C reservoir effect (MRE) occurs as a spatially and temporally dependent variable
owing to localized changes in oceanic water composition. Previous studies have endeavored to
quantify the MRE for a variety of locations and time periods. By determining the MRE for a specific
time and place, the deviation (known as ΔR) from the global average MRE of zero can be calculated,
and therefore a more accurate calibration of ages based on marine-derived carbon can be obtained.
Previous research in this laboratory investigated the MRE for the Atlantic seaboard from the west
 Abstracts 33

coast of Ireland, through to the Faroe Isles. This facilitated a comprehensive investigation of tempo-
ral 14C variations for this region during the Holocene and produced revised ΔR values for the west
coast of Scotland, northern Iceland, and the Faroe Islands. Significant re-evaluation and re-interpre-
tation of Scottish archaeological sites followed, where conclusions had been drawn that were based
on dates from marine-derived carbon. This current study extends the investigation to the east coast
of Scotland, where a more complicated topographical estuarine system exists, draining into the
semi-enclosed North Sea basin.
Owing to the availability of suitable archaeological samples, the data set has a distinct Medieval
focus that spans an extended geographical area from Aberdeen in the north, to East Lothian in the
south. Many of the ΔR values are not significantly different from zero (the global average), but there
are occasional excursions to negative values indicating the presence of younger water. These values
are in stark contrast to the average value determined for the Middle Iron Age on the west coast of
Scotland (–79 ± 17) (Ascough et al. 2004), suggesting that considerable care must be taken when
dating marine-derived samples from Scottish archaeological sites.

REFERENCE
Ascough P. et al. 2004. Holocene variations in the Scottish marine radiocarbon reservoir effect. Radiocarbon 46(2):611–20.

#51

MARINE CHRONOLOGY BASED ON 14C DATING ON DIATOMS PROTEINS

Christine Hatté1, Gregory Hodgins2, A J Timothy Jull2,3, Bongi Bishop2, Benoit Tesson4
1. Laboratoire des Sciences du Climat et de l’Environnement, UMR CEA-CNRS-UVSQ 1532, Domaine du CNRS, F-91198
Gif-sur-Yvette, France.
2. NSF Arizona AMS Laboratory, Physics Building, University of Arizona, Tucson, AZ 85721, USA.
3. Department of Geosciences, University of Arizona, Tucson, AZ 85721, USA.
4- ISOMer, Laboratoire de Biologie Marine, UPRES EA 2663 “Ecophysiologie Marine Intégrée”, Université de Nantes, 2
rue de la Houssinière, BP 92208, F-44322 Nantes Cedex 03, France.

We present a new method of obtaining radiocarbon dates for proteins intrinsic to diatom frustules
(sillafin). By asserting age models for sediment core samples that lack calcium carbonate, this
method will improve interpretations of diatom-based paleoproxies for either marine or lacustrine. In
preparation for radiocarbon dating by accelerator mass spectrometry, diatoms were concentrated
from core sediment samples. The diatoms frustules were freed of any surface-bound organic matter,
through chemical and physical treatments that will be discussed later in this paper. Compounds
intrinsic to diatoms frustules were released from their opal matrix by HF dissolution. This method
differs from approaches based on specific compound extractions from complex organic mixture by
preparative chromatography because we eliminated most of the contaminated organic matter. The
advantage of our method is that it does not require heavy cost investment and it can be added to a
preparative chromatography protocol to ensure dating reliability.
This method was applied to samples from a marine core collected in the Southern Ocean that spans
the last climatic cycle. We report on the radiocarbon dating results obtained on organic matter at
each step of the chemical treatment, from bulk to sillafin and their interpretation.
34 Abstracts

#52

RADIOCARBON AND THE OCEAN SCIENCES – CONTRIBUTIONS FROM A DEDICATED 14C

AMS FACILITY
Karl F von Reden, representing current and former staff colleagues at NOSAMS
National Ocean Sciences AMS Facility Geology and Geophysics Department Woods Hole Oceanographic Institution Woods
Hole, MA 02543, USA.

In the late 1980s, a Request for Proposals was issued by the National Science Foundation to estab-
lish a dedicated facility to analyze samples of marine science origin for radiocarbon, using acceler-
ator mass spectrometry. The main driver for this project was the World Ocean Circulation Experi-
ment (WOCE) that was expected to generate in excess of 13,000 seawater samples to be analyzed
for dissolved inorganic carbon at such a laboratory. The Woods Hole Oceanographic Institution was
awarded the project and the National Ocean Sciences Accelerator Mass Spectrometry Facility
(NOSAMS) was established in 1989. During the two decades, over 1100 scientific papers have been
published that cite or rely on radiocarbon results produced at this facility. In some instances, other
AMS laboratories were also involved in the measurements. This review will attempt to show how
progress in the marine sciences was affected by the data that were obtained at NOSAMS and distrib-
uted to numerous researchers all over the world. Only publications that explicitly or implicitly
acknowledge AMS measurements that were performed at NOSAMS are included in this review.
Clearly, there will not be enough space or time to mention all of our “clients’” research in this review
and it should be emphasized that this does not constitute judgment over the importance of any spe-
cific topic. There will be coverage of four main areas: tracer mapping programs, paleo research, bio-
geochemical applications, and future trends.

#53

RELATION BETWEEN SEA SURFACE VARIABILITY OF NORTHEAST ATLANTIC Δ14C AND

THE NORTH ATLANTIC OSCILLATION
N Tisnérat-Laborde1, M Paterne1, B Métivier2, M Arnold3, P Yiou1, G Reverdin4
1. Laboratoire des Sciences du Climat et de l’Environnement, IPSL, CEA-CNRS-UVSQ, Gif-sur-Yvette, France.
2. Muséum National d’Histoire Naturelle, Département Milieux et Peuplements Aquatiques, UMS 0401, UMR-CNRS
BOME, Paris, France.
3. CEREGE, Europôle Méditerranéen de l’Arbois, Aix en Provence, France.
4. Laboratoire d’Océanographie et de climat, Expérimentation et approches numériques, IPSL, UPMC, Paris, France.

Radiocarbon in the surface ocean depends on ocean-atmosphere exchange of 14CO2 and ocean
dynamics and is thus linked to climate forcing. We investigate variations of surface water radiocar-
bon related to climate change. Here we present the first quasi-annual pre-bomb radiocarbon time
series of the North East Atlantic (NEA) sea surface covering 1823 to 1952. It is obtained from a
compilation of new radiocarbon measurements on mollusk shells from the collection of the Muséum
National d’Histoire Naturelle de Paris and of previously published data. The NEA Δ14C record
shows two major features: 1) a decrease of 12‰ due to the huge injection of fossil fuel CO2 into the
atmosphere, known as the Suess effect; and 2) 14C fluctuations of some 10‰ up to 15‰ between AD
1885 and 1950. We find that these variations cannot be attributed to the mollusk species, depth hab-
itats, diet and latitudinal distribution. Spectral analyses of the NEA Δ14C exhibit a quasi-periodic
 Abstracts 35

cycle around ~7.4 years. This can be associated with the ~6.5-year period of winter North Atlantic
Oscillation (NAO). Moreover, a direct comparison shows that Δ14C lows correspond to more intense
phases of winter NAO, with a phase lag. Our study suggests that these Δ14C variations are not
related to local changes but reflect more likely advective events.

#54

DECADAL CHANGES OF RADIOCARBON IN THE SURFACE BAY OF BENGAL: THREE

DECADES AFTER GEOSECS AND ONE DECADE AFTER WOCE
Koushik Dutta1, G V Ravi Prasad1,2, Dinesh K Ray1, Sanjeev Raghav3
1. Institute of Physics, Sachivalaya Marg, Bhubaneswar 751 005, India.
2. Center for Applied Isotope Studies, University of Georgia, Athens, GA, USA.
3. Marine Wing, Geological Survey of India, Kolkata 700 064, India.

Temporal changes of radiocarbon in the oceanic dissolved inorganic carbon (DIC) provide valuable
clues for air-sea CO2 exchange and circulation. Detailed measurements of radiocarbon in the north-
ern Indian Ocean were carried out during the GEOSECS expeditions of 1977–78 (Stuiver and
Ostlund 1983), and two decades later during the WOCE expedition of 1994–95. DIC 14C measure-
ments were also reported by Bhushan et al. (1997) and Dutta (2001) for several other important loca-
tions in the northern Indian Ocean, which were not covered during the GEOSECS or WOCE expe-
ditions. Monitoring decadal changes of 14C in the oceanic DIC offers a way for validating ocean
circulation models, which predict the spatial and temporal changes in the distribution of radiotracers.
Surface seawater samples were collected onboard ORV Samudra Manthan during November 2006,
approximately following a meridional transect along 90°E in the central Bay of Bengal from 21°N
to about 5°N. Samples were collected with a PVC Niskin sampler and stored refrigerated in airtight
bottles after poisoning them with HgCl2. CO2 was extracted from these samples by acidifying with
H3PO4 under vacuum. The liberated CO2 was cryogenically purified and converted to graphite for
AMS 14C analysis. An aliquot of CO2 was used for δ13C analysis. 14C measurements were carried
out at the AMS facility of the Center for Applied Isotope Studies, University of Georgia. The Δ14C
of these surface seawater samples ranged from 38‰ to 55‰ (mean 50.5 ± 1.5‰, n = 13). Compar-
ison of the 14C results in these samples with that of 1977–78 GEOSECS, 1994–96 WOCE, and other
results of 1993–99 will give us information on temporal and spatial changes of surface ocean 14C,
which can address circulation changes in the northern Bay of Bengal three decades since GEOSECS
and one decade after WOCE and PRL surveys. This data would be useful for testing numerical
ocean circulation models that simulates oceanic tracer distribution (Guilderson et al. 2000).

REFERENCES
Bhushan R, Somayajulu BLK, Chakraborty S, Krishnaswami S. 2000. Radiocarbon in the Arabian Sea water column: CO2
exchange rates and temporal variations in bomb 14C inventory since GEOSECS. Journal of Geophysical Research 105:
14,273–82.
Dutta K. 2001. Study of marine processes in the northern Indian Ocean using radiocarbon [PhD thesis]. M.S. University of
Baroda, Vadodara, India. 142 p.
Guilderson TP, Caldeira K, Duffy PB. 2000. Radiocarbon as diagnostic tracer in ocean and carbon cycle modeling. Global
Biogeochemical Cycles 14:887–902.
Stuiver M, Ostlund HG. 1983. GEOSECS Indian Ocean and Mediterranean radiocarbon. Radiocarbon 25(1):1–29.
36 Abstracts

Parallel Session 3-4: Calibration and Calculation

#55

EXTENSION OF THE SOUTHERN HEMISPHERE RADIOCARBON CURVE IN TASMANIAN

PINE, 955–2200 YR BP
Susan Zimmerman1, Thomas Guilderson1,2, Brendan Buckley3, Edward Cook3
1. Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA.
2. Department of Ocean Sciences and Institute of Marine Sciences, University of California, Santa Cruz, CA 94064, USA.
3. Tree Ring Laboratory of the Lamont Doherty Earth Observatory, Palisades, NY 10964, USA.

Published comparisons of radiocarbon-calendar calibrations for the Northern and Southern hemi-
spheres have shown disagreement in whether a hemispheric offset exists, and if so, how large and
variable the offset is. A recent cross-laboratory, interhemispheric comparison showed with very
high precision that the atmosphere in the Southern Hemisphere was on average 41 (±14) years older
than in the Northern Hemisphere over nearly 1000 years. Here we present radiocarbon dating results
of decadal samples of dendrochronologically-dated pine (Lagorostrobos franklinii) from the Stanley
River basin, Tasmania, between 955–2200 yr BP. A δ13C of –25‰ is assumed while awaiting sam-
ple-specific results, which will be available in time for the conference.
This data set overlaps and extends the current Southern Hemisphere record, which covers the period
110–995 yr BP. There is good agreement between the two records between 995–955 yr BP, and
between sample replicates and with consensus values for several standards. However, unlike the
published data set, we find little evidence of offset between the Southern And Northern hemispheres
for the time interval studied. One possible explanation for both our results and other published stud-
ies that find no offset may be that laboratory differences create or obscure offsets when Northern and
Southern hemisphere samples are measured in different labs. Alternatively, there may be real tem-
poral variability that results in periods of zero offset as well as periods of distinct difference between
hemispheres.

#56

COMPARISON OF 14C AGES BETWEEN LSC AND AMS MEASUREMENTS FOR CHOUKAI JIN-
DAI CEDAR TREE RINGS AT THE RAPID CHANGE IN 2600 CAL BP
Yui Takahashi1, Hirohisa Sakurai1, Kayo Suzuki1, Taiichi Sato1, Shuichi Gunji1,
Fuyuki Tokanai1, Hiroyuki Matsuzaki2, Yoko Sunohara2
1. Department of Physics, Yamagata University, 1-4-12 Kojirakawa, Yamagata 990-8560, Japan.
2. Research Center for Nuclear Science and Technology, The University of Tokyo, Tokyo 113-0032, Japan.
14C concentrations of single-year tree rings indicate solar modulations of cosmic rays such as the 11-

yr solar cycle, the Gleissberg cycle with a period of 80 years, etc. Hence, old tree rings are a power-
ful tool for detecting past variations in cosmic rays. From 14C dating, calendar ages of Choukai Jin-
dai cedars in Japan (39°N) ranged from 2757 to 2437 cal BP with 320 tree rings. According to the
IntCal04 calibration curve, this wood sample is located in a very interesting era because the 14C con-
centrations form a peak as well as Maunder minimum at which the sunspot number was absent.
Moreover, IntCal04 shows a characteristic peak between 2650 and 2600 cal BP. However, the mea-
sured data set of this peak area is poor and it is for a 10-yr span. Since the IntCal04 calibration curve
in the episode was compiled using the data sets from Irish oak, German oak, etc., the data set of
Choukai Jindai cedar allows for comparison of the regional effects between Japan and Europe, and
 Abstracts 37

also allows investigating the solar activity at about 2600 cal BP. We have measured the 14C ages of
100 single-year tree rings of Choukai Jindai cedar using two methods for the same tree rings, the
highly accurate liquid scintillation counting system (LSC) by Quantulus and accelerator mass spec-
trometry (AMS) by 5MV tandem accelerator. The statistical accuracies were typically 16 and 30 14C
yr for the LSC and the AMS, respectively. The 14C ages were cross-checked during the characteris-
tic peak area and the following comparably flat area. The difference of the 14C ages between the
LSC and the AMS methods was 3.7 ± 5.2 14C yr on average. We will describe the variation in both
measurements in detail.

#57

TREE-RING DATING AND AMS WIGGLE-MATCHING OF WOODEN STATUES IN KOREA

Won-Kyu Park1, Yojung Kim1, Ah-Reum Jeong1, Sang-Kyu Kim2, Kwang-Hee Lee2,
Byung-Hwa Son2, Jung-Ae Oh2, Suh-Young Park3, Sunil Choi4
1. Department of Wood and Paper Science, Chungbuk National University, Cheongju 361-763, Korea.
2. Tree-Ring Material Bank, BLD #504-216, Chungbuk National University, Cheongju 361-763, Korea.
3. Department of Art History, Hongik University, Seoul 121-791, Korea.
4. Office of Cultural Properties Appraisal Incheon International Airport, Incheon 400-715, Korea.

Studying Buddhist statues could be the one of the best ways to understand Buddhist art. Species
analyses and dating is helpful to find the preferred woods at that time and to support relic value. We
are conducting a 3-year project on dating Korean Buddhist statues. This paper reports initial studies
on tree-ring dating and AMS wiggle-matching for several wooden Buddhist statues at various tem-
ples in Korea.
The first statue is a wooden Buddhist statue from Jogae Temple, Seoul. The date of this statue is con-
troversial: some art historians estimate its age as less than 150 years, but the temple claims a much
older age, about 600 year based on historical records.
When we observed the bottom of the statue, the only part available for sampling, 2 back and 2 bot-
tom boards were taken for sampling. The back boards are Zelkova serrata (Thunb.) Makino and
Pinus koraiensis S. et Z., and the bottom boards Pinus koraiensis S. et Z. and Tilia species. The gilt
layer of the statue body is considerably removed. It was possible to observe tree rings on a back
board (Z. serrata). We measured 170 ring widths, made a curve, and compared it with master curves.
Our curve did not agree with the master curves, so we tried to date the statue using wiggle-match
radiocarbon dating.
We took 5 ring samples at every 10th ring for AMS radiocarbon dating. Wiggle-matching of AMS
results produced a date of AD 1405 ± 10 for the outermost ring. If we assume that about 50 rings had
been eliminated during processing, the estimated cutting date is AD 1455 ± 10. The AMS dates sup-
port the temple’s claim. We will discuss several other cases of wooden statues (17th–18th century
AD) in the southern part of Korea.
38 Abstracts

Parallel Session 2-5: Radionuclide Tracers

#58

MEASUREMENTS OF ACTINIDES AT CIRCE IN CASERTA (ITALY)

M De Cesare1,3, N De Cesare2,3, A D’Onofrio1,3, L Gialanella3, A Petraglia1, Y Guan4,
F Quinto1,5, V Roca3,6, D Rogalla1,3, C Sabbarese1, F Terrasi1,3
1. CIRCE, INNOVA, and Dipartimento di Scienze Ambientali, Seconda Università di Napoli, Caserta, Italy.
2. CIRCE, INNOVA, and Dipartimento di Scienze della Vita, Seconda Università di Napoli, Caserta, Italy.
3. INFN Sezione di Napoli, Naples, Italy.
4. College of Physical Science and Engineering Technology, Guangxi University, Nanning 530004, China; Also: ICTP, Tri-
este, Italy.
5. Faculty of Physics, University of Vienna, Austria.
6. Dipartimento di Scienze Fisiche, Università Federico II, Naples, Italy.

Nuclear power plants (NPPs) convert nuclear energy into electric power by thermal reaction, while
nuclear weapons (NW) use the destructive effects of supercritical chain reactions. Different sources
of contamination affect the environment on large (atmospheric NW test fallout and Chernobyl-like
accidents) and local (NPPs) scales. Both of these applications of nuclear energy, together with all the
activities correlated with production, transport, and reprocessing of nuclear fuel, lead to the release
of a wide range of radioactive elements in the environment, such as uranium, plutonium, fission and
activation products. Since the relative concentrations of plutonium and uranium isotopes depend on
the nature of the source material and on its subsequent irradiation history, all these sources do not
contribute equally to the contamination in a given site.
These nuclides are present in environmental samples at ultra-trace levels and their detection requires
the resolution of MS techniques, but only AMS has the sensitivity required. In order to perform the
isotopic ratio measurements of actinides, e.g. 236U5+/238U, an upgrade of the CIRCE (Center for Iso-
topic Research on Cultural and Environmental Heritage, Caserta, Italy) accelerator, originally
equipped for radiocarbon AMS, has been planned. The main difficulty of AMS measurement of
236U and of Pu isotopes is the intense neighboring beam of 238U. Although most of the 238U ions are

suppressed by magnetic and electrostatic elements, a small fraction of this intense beam can inter-
fere with measuring the rare isotopes.
We report on the results of simulations and test measurements (obtained with a 16-strip large silicon
detector) aimed to define the best ion optics and to understand the origin of background ions. A high
resolution time of flight-energy (TOF-E) detector system will described and used, to identify the
rare isotopes from the unsuppressed interfering ions.

#59

QUANTIFICATION OF FOSSIL FUEL CO 2 EMISSIONS FROM EAST ASIA USING ATMO-

SPHERIC OBSERVATIONS OF Δ14CO2
Jocelyn Turnbull1, Pieter Tans1, John Miller1,2, Scott Lehman2, Gabrielle Petron1,2, Jay Gregg3
1. NOAA Earth System Research Laboratory, 325 Broadway, Boulder, CO 80305, USA.
2. University of Colorado, Boulder 80309, USA.
3. Dept. of Geography, University of Maryland, USA.

Fossil fuel CO2 emissions are the largest net source of CO2 to the atmosphere, and accurate quanti-
fication of these emissions is essential to furthering our understanding of the global carbon cycle.
 Abstracts 39

Currently, fossil fuel CO2 (CO2ff) emissions are estimated from economic inventories of fossil fuel
use reported by governments and industry. Measurements of the radioactive isotope 14C in CO2 pro-
vide an independent, objective method of constraining CO2ff emissions, since fossil fuel-derived
CO2, unlike other CO2 sources, is devoid of 14C. These measurements can be used to validate eco-
nomic inventories, and have the potential to improve on them by: quantifying uncertainty estimates;
providing emissions estimates at smaller scales (in both space and time); and may be available to
researchers more quickly than inventory data.
Emissions from East Asia are of particular interest, since they contribute ~25% of global total CO2ff
emissions, and are growing rapidly by 8–10% per year, but are quite uncertain compared to other
regions. We quantitatively determine recently added CO2ff in samples from the NOAA/ESRL coop-
erative sampling network site at Tae-Ahn Peninsula, South Korea, which usually sees air that has
recently passed over northern China and Korea. Samples typically contain a few ppm of CO2ff from
China, and Korea contributes additional CO2ff in some samples; most measured values are much
larger than the detection limits of the method. The observed CO2 also shows large variability due to
biospheric CO2 exchange, even in winter, so that CO2 measurements alone cannot accurately esti-
mate CO2ff. We compare the observational results with estimates of CO2ff from a Lagrangian parti-
cle dispersion model and a prior estimate of CO2ff emissions, examining our ability to validate the
reported emissions and determine quantitative uncertainties to these emissions.

#60

ATMOSPHERIC FOSSIL FUEL CO2 RECORD IN A CENTRAL EUROPEAN CITY DURING THE
WINTER OF 2008/09

M Molnár1, L Haszpra2, É Svingor1, M Veres3, I Major4, I Svetlík5

1. Hertelendi Ede Laboratory of Environmental Studies, MTA ATOMKI, Debrecen, Hungary.
2. Hungarian Meteorological Service, Budapest, Hungary.
3. Isotoptech Zrt, Debrecen, Hungary.
4. University of Debrecen, Hungary.
5. Nuclear Physics Institute AS CR, Prague, Czech Republic.

Fossil fuel CO2 content in the air of a major Hungarian city (Debrecen) was determined using mea-
surements of CO2 mixing ratio and 14C content of air. For this project, we developed a high-preci-
sion atmospheric CO2 monitoring station in Debrecen. An integrating sampling system (developed
by ATOMKI) was applied for 14C measurements. One sampler was installed in Debrecen station and
two independent 14CO2 sampling lines were installed ~300 km from Debrecen at Hegyhátsál station
as independent background references, where high-precision atmospheric CO2 mixing ratio mea-
surements have been conducted since 1997.
During the winter of 2008/09, we measured the mixing ratio and 14C content of atmospheric CO2 at
Debrecen and the reference station simultaneously. It was concluded that trends in CO2 mixing ratio
variations in time are very similar at the three different sampling points (2 m above ground in Debre-
cen, 10 m and 115 m above ground in Hegyhátsál). Air quality in Debrecen during September 2008
seemed to be relatively clear, at least regarding its CO2 content. When winter approached in Octo-
ber, with lower outside temperatures and less sunshine hours, the CO2 content of air increased in
general at all three sampling points, but this effect was more intensive closer to ground level.
40 Abstracts

Our 14C measurements clearly indicated that there was not a significant amount of fossil fuel CO2 in
the air of Debrecen in September 2008. But during the winter of 2008/09, the 14C value of atmo-
spheric CO2 of Debrecen decreased by more than 40‰ relative to the September results. According
to our calculations, this was caused by ~20 ppm fossil fuel CO2, which appeared as a surplus amount
in the air above the September level.

#61

C/N RATIO AND RADIOCARBON AS TRACERS FOR STUDY OF AN OIL-SPILLED COASTAL

AREA, TAEAN IN KOREA

K J Kim, S W Park, H Song

Korea Institute of Geoscience and Mineral Resources, 92 Gwahang-no, Yuseong-gu, Daejeon 305-350, Korea.

On 7 December 2007, a major oil spill in Taean, Korea, happened when a Samsung crane barge
towed by a tug boat collided with the anchored Hong Kong registered crude oil carrier Hebei Spirit.
This collision resulted in a leakage of some 10,900 tons of oil. The spill occurred near Mallipo
Beach in Taean County, one of the most beautiful and popular beaches in South Korea. The oil spill
has impacted this region of large coastal areas where a national maritime park and 445 sea farms
were located. We used both radiocarbon and the stable isotope ratio of carbon to nitrogen (C/N) to
monitor the affected areas, especially for biological samples from April to October 2008. Because
crude oil carbon contains much less 14C than that of modern biological samples at present day, we
expected that 14C can be an useful tracer to monitor the recovery process of oil-contaminated bio-
logical environments. We have monitored biological samples such as clam, shell, seaweed, crab,
oyster, fish, etc., to measure its C/N ratio and 14C amount for the six-month period. Our results indi-
cated that C/N ratios of biological samples tend to decrease as time passes, while the 14C amounts
increase. Our study confirmed that the degree of 14C absorption in biological samples varies based
on species. This could be due to the difference in trophic level of carbon in each species. The 14C
amount as pMC (percent Modern Carbon) among the above biological samples for the 6 months
from April to October 2008, ranged from 103.61 to 108.56 pMC. The C/N ratio ranged from 1.97 to
15.44 among all species. Our presentation will demonstrate the detailed monitored values of C/N
ratio and 14C amount for each species as well as those for crude oil, sand, and mud samples of a few
Taean costal regions with respect to both C/N ratios and 14C as environmental tracers.

#62

RADIOCARBON AND OTHER GLOBAL FALLOUT RADIONUCLIDES IN INDIAN OCEAN

WATERS
P P Povinec1, M Aoyama2, R Breier1, J Gastaud3, M Groening4, K Hirose2, A J T Jull5, S-H Lee6,
I Levy-Palomo3, U Morgenstern7, P Roos8, J A Sanchez-Cabeza3, I Sýkora1, Z Top9
1. Comenius University, Faculty of Mathematics, Physics and Informatics, 84248 Bratislava, Slovakia.
2. Meteorological Research Institute, Department of Geochemistry, Tsukuba, Japan.
3. International Atomic Energy Agency, Marine Environment Laboratories, Monaco.
 Abstracts 41

4. International Atomic Energy Agency, Isotope Hydrology Laboratory, Vienna, Austria.

5. University of Arizona, Departments of Physics and Geosciences, Tucson, USA.
6. Korea Research Institute of Standards and Science, Daejeon, Republic of Korea.
7. Institute of Geological and Nuclear Sciences, Lower Hutt, New Zealand.
8. Risø National Laboratory, Roskilde, Denmark.
9. University of Miami, Rosentiel School of Marine and Atmospheric Sciences, Miami, USA.

The distribution of global fallout 14C, 3H, 137Cs, and 239,240Pu in seawater of the Indian Ocean about
four decades after their main injection from atmospheric nuclear weapons tests on the ocean surface
is presented and discussed. We evaluate results obtained in the GEOSECS, WOMARS, and WOCE
projects, and compare them with more recent data (2003–2004) obtained in the framework of the
SHOTS (Southern Hemisphere Ocean Tracer Studies) project. The observed results indicate trans-
port of radionuclides from the Northwest Pacific via the Indonesian Seas to the Indian Ocean, as
well as their accumulation in the South Indian Ocean Subtropical gyre. The 137Cs concentrations
observed in surface waters southwest of Australia decreased by about a factor of two when com-
pared with 1976–1978 data, in agreement with the estimated effective half-life of 137Cs in the Indian
Ocean. A different situation is in the case of transuranics, where the observed 239,240Pu concentra-
tions are by about a factor of two lower than previously reported. In the case of 241Am, they
decreased only by ~30%, in spite of relative shorter effective half-life of 241Am. A slower decrease
in transuranics levels may be due to their additional transport from Marshall Islands testing grounds,
and in the case of 241Am also due to its in-growth from the decay of 241Pu. On the basis of radionu-
clide signatures, different water masses have been identified in the southern Indian Ocean. While the
Subtropical gyre has been maintaining high radionuclide levels in surface and subsurface waters of
the basin, the deep waters (between 1000–2500 m) have been influenced by North Atlantic Deep
Water (NADW), and the bottom waters by Antarctic Bottom Water (AABW).

Parallel Session 3-5: Eurasian Archaeology I

#63

WHAT HAS CHANGED BECAUSE OF THE YAYOI DATING CONTROVERSY?

Shinya Shoda
University of Tokyo, Japan.

Since the sensational 2003 announcement by the National Museum of Japanese History that AMS
14C dating of charred pottery remains pushed the start of the Yayoi period back by 500 years, archae-

ologists working on 1st millennium BCE material have had to switch from the older “short chronol-
ogy” to a new “long chronology.” This announcement helped Japanese archaeologists notice the
failings of traditional period datings and re-examine them. Despite a small number of staunch
defenders of traditional dating, most Japanese archaeologists have accepted this “shift” to varying
degrees. However, this change need not apply to the entire northeast Asian region as China’s chro-
nology for the 1st millennium is tied to written records. Considering the social and material links
across northeast Asia, the chronology of the Korean peninsula, intermediate between Japan with its
new long chronology and China with its text-based one, needs to be re-examined. These changes
also mean that the traditional “inclined chronology,” in which archaeological artifacts from the cen-
42 Abstracts

ter are considered to be older while those from peripheral areas would be younger, must be re-exam-
ined. The chronology of the 1st millennium BCE in the Korean peninsula is still in dispute, in part
because of the flood of 14C data lacking clear archaeological contexts. The author summarizes these
problems, demonstrates a reliable material relationship existed at this time linking northeast Asia
from China to Japan, and documents its absolute dating.

#64

THE ORIGIN OF POTTERY IN EAST ASIA AND ITS RELATIONSHIP TO ENVIRONMENTAL

CHANGES IN THE LATE GLACIAL: WHO CAME FIRST, THE CHICKEN OR EGG?
Yaroslav V Kuzmin
Institute of Geology & Mineralogy SB RAS, Novosibirsk, Russia.

Chronometry of the pottery origin in East Asia can provide clues to the eternal question in Old
World archaeology: did environmental changes trigger and/or accelerate the emergence of cultural
innovations such as pottery-making, fishing, maritime adaptation, and agriculture? Recent results
show that pottery appeared almost simultaneously in three regions of East Asia: South China, Japa-
nese Islands, and the Russian Far East. The time period of ca. 13,700–13,300 BP (in south China
could be even as old as ca. 15,000 BP) witnessed the beginning of ceramics production in East Asia.
This was in general the Late Glacial progressive warming after the Last Glacial Maximum, with
several warmer and cooler oscillations. It should be noted that the first pottery-containing sites pre-
ceded the Bølling-Allerød warm event (ca. 13,000–11,500 BP), as noted by Taniguchi (2002). Thus,
the relationship between the origin of pottery and climatic and vegetation changes was not “linear,”
and the question of cause and consequence cannot be solved easily to the present day. It seems that
the combination of environmental changes (appearance of deciduous species pushing out conifers in
south China and Japan and the slow migration of broadleaved species to the Russian Far East) and
the need to process freshwater food resources (fish, including anadromous species; and mollusks)
and nuts resulted in the introduction of pottery-making. An important feature is the quite non-uni-
form nature of the Neolithization process (i.e. the beginning of pottery manufacture) in East and
Southeast Asia, where often in two neighboring regions (such as south China and Southeast Asia)
pottery appeared at very different times (at ca. 15,000-–4,000 BP vs. ca. 4000 BP, respectively).
Thus, the question “Who came first: the chicken or egg?” still stands in terms of what was the driv-
ing force for the emergence of pottery in East Asia and worldwide.
This research is supported by FBBR (09-06-00126).
 Abstracts 43

#65

THE AGE OF EARLY HOMO SAPIENS IN CHINA

Susan G Keates
United Kingdom. Email: archres2008@yahoo.com.

Pleistocene fossils of early modern humans are known from the northern and southern parts of the
People’s Republic of China. Their chronological status is based almost exclusively on indirect dat-
ing of the stratigraphic context and faunal and radiocarbon dating of associated or supposedly asso-
ciated archaeological specimens. The human fossils comprise isolated teeth, crania as well as more
complete skeletal remains, such as Liujiang and Shandingdong (Upper Cave). This talk will exam-
ine the radiocarbon chronology of early modern humans to ascertain their actual age.

#66

THE EFFECT OF ACCURACY AND PRECISION ON ATTEMPTS TO DATE THE FIRST PEO-
PLING OF SAHUL RADIOMETRICALLY
R Esmée Webb
School of Natural Sciences, Edith Cowan University, Joondalup, WA 6027, Australia.

Precisely when Sahul was first colonized continues to be contentious partly because people appear
to have reached the landmass before they managed to invade Europe; yet getting to Sahul required
over-water dispersal, whereas Europe was accessible on foot. European archaeologists, in particular,
feel that the difficulties inherent to reaching Sahul militate against its early settlement. Conse-
quently, they seek to discredit any dates from archaeological sites that are older than those obtained
from the same sites by radiocarbon. The latter indicate an arrival time of about 40 kyr; whereas
luminescence and electron spin resonance suggest an arrival time of about 60–50 kyr.
Here, I argue that those who espouse the “short” chronology appear to have overlooked the temporal
limitations of radiocarbon dating, the fact that different techniques measure “time” in different
ways, yielding “years” of different length, and that all radiometric dates are probability statements.
I show that the archaeological evidence cannot be sequenced sufficiently precisely to address such
questions as how (mode) and how quickly (tempo) people colonized Sahul. When all the radiomet-
ric dates from the oldest sites are considered as age ranges, sites in the northern part of Australia can-
not definitively be shown to have been occupied before sites in southern Tasmania, suggesting that
the whole landmass was occupied within the limits of accuracy of the dating techniques employed,
that is, “instantaneously.” That “radiometrically instantaneous” period lasted for many millennia,
however. It was sufficiently long that the colonizing population would have been able to adapt to liv-
ing even in the central deserts quite slowly. Hence, until as many as possible of the oldest Sahulian
sites are redated using longer-lived isotopes than radiocarbon, we will not know when people first
reached Sahul. Moreover, we need to rethink our models of how people colonize and adapt to pris-
tine landmasses.
44 Abstracts

#67

RADIOCARBON DATES DOCUMENTING THE NEOLITHIC-BRONZE AGE TRANSITION IN

KOREA
Jongchan Kim1 and Christopher J Norton2
1. School of Physics, Seoul National University, Korea.
2. Department of Anthropology, University of Hawai’i at Manoa, USA.

One of the most debated topics in Korean archaeology is the nature of the Neolithic-Bronze Age
transition, primarily from a hunting-gathering-fishing lifestyle to reliance on full-scale plant and
animal domesticates. (In Korea, the Neolithic is sometimes referred to as the Chulmun period, while
the Bronze Age is often referred to as the Mumun period—similar to the Japanese Jomon and
Yayoi.) In particular, there are two primary arguments currently being debated: 1) the transition is
due primarily to an influx of farmers migrating from the north; and 2) the transition is due to increas-
ing indigenous population growth. However, these interpretations have often relied on a limited
number of excavations and a paucity of 14C dates from the time period in question (ca. 2800–4900
BP). The intent of our paper is to present an updated compilation of 14C dates from sites in Korea
that date to the Late Neolithic and Early Bronze Age. These data are primarily derived from the
SNU-AMS laboratory and culled from published literature when applicable. Thus, we present a
comprehensive range of 14C data currently available in Korea for this time period. Based on the dis-
tribution of 14C dates, we see that from ca. 3900 BP the number of Neolithic sites slowly decreases
while the number of Bronze Age sites slowly increases. Neolithic sites disappear from the archaeo-
logical record by 3400 BP. Although the full-blown Bronze Age is observed by ca. 3200 BP, we are
able to document evidence that it may have started as early as ca. 3800~3900 BP. After 3000 BP,
only Bronze Age sites are present in Korea. It is interesting to note that there are two major increases
in the number of sites during the Neolithic-Bronze Age transition. The first one occurs at ca. 4500
BP, doubling at ca. 4200 BP and remaining stable until ca. 3700 BP. The second major increase
begins at ca. 3600 BP and more than triples by 3000 BP. Thus, the 14C data support the claims for a
significant increase in population during the Neolithic-Bronze Age transition in Korea. We explore
possible causal factors for this increase in population in Korea in our paper.
 Abstracts 45

#68

THE LATE PALEOLITHIC-NEOLITHIC TRANSITION IN KOREA: CURRENT ARCHAEOLOGICAL

AND RADIOCARBON PERSPECTIVES
Christopher J Norton1 and Jongchan Kim2
1. Department of Anthropology, University of Hawai’i at Manoa, USA.
2. School of Physics, Seoul National University, 151-742 Seoul, Korea.

The Late Paleolithic-Neolithica transition in Korea include a number of behavioral changes in

human foraging groups that parallel or differ from other regions of the world. The Korean Late Pale-
olithic is represented by mobile foraging groups, and stone toolkits that include blades and micro-
blade technologies, while the Neolithic is represented by the appearance of sedentary villages, stone
toolkits that include chipped stone axes, and pottery for storage. Broad spectrum foraging is thought
to be representative of both cultural periods, though research in Korea concerning Late Paleolithic
and Neolithic diet is still in its infancy. The primary question being asked is what are the causal fac-
tors that led to these behavioral changes? Nevertheless, previous studies that have discussed this
behavioral transition in Korea have relied on a relatively sparse data set, including 14C data. Utiliz-
ing current 14C data analyzed primarily by the SNU-AMS laboratory we contribute the most up-to-
date data for dating archaeological sites in Korea that derive from this behavioral transitional period.
We evaluate how these 14C data influence our view of particular sites and behavioral complexes,
particularly when placed in a broader perspective that includes variation in the paleoenvironment,
paleobathymetry, mobility and foraging strategies.
a. As suggested by Gao and Norton (2002, Antiquity), a distinct “Middle” Paleolithic is absent from much of East Asia. Thus,
we use the two-stage model (Early and Late Paleolithic) here. The Neolithic is also sometimes referred to as the Chulmun
period (similar to the Japanese Jomon).

Parallel Session 2-6: Soils and Sediments

#69

OVERVIEW OF RAMPED PYROLYSIS RADIOCARBON DATING AND APPLICATION TO CAR-

BON CYCLING AND CHRONOLOGY
Brad E Rosenheim1, Eugene Domack2, Ann McNichol3, Valier Galy3, John M Hayes3,
Kimberly Roe1, Jessica Adamic1, Mead A Allison4
1. Tulane University, New Orleans, LA, USA.
2. Hamilton College, Clinton, NY, USA.
3. WHOI, Woods Hole, MA, USA.
4. University of Texas, Austin, TX, USA.

We present an overview of a ramped pyrolysis technique aimed at analyzing acid insoluble sedimen-
tary carbon by exploiting differences in thermochemical stability of carbon-bearing compounds.
The method is based on the idea that the younger organic materials will be more likely to yield vol-
atile, carbon-containing products at lower temperatures than more mature organic materials. The
abundance of 14C in those products will yield a more accurate estimate of minimum age of the sed-
iment as well as the proportions of carbon from reservoirs of different ages. This hypothesis is sup-
ported in all depositional environments thus far analyzed; yet some environments suggest a decou-
46 Abstracts

pling of chemical decomposition from radiocarbon age within discernible carbon-bearing

components. In all cases examined thus far, analysis of radiocarbon in pyrolysis products has pro-
vided important, new information about the diversity of organic material present in the sediment.
We pose several scientific and technical questions that are being investigated using ramped pyroly-
sis radiocarbon dating.

#70

EXCHANGE PROCESSES OF CARBON ISOTOPES 14C AND 13C IN THE KARST ENVIRON-
MENT
Nada HorvatinËiÊ, Bogomil ObeliË, Ines Krajcar BroniË, Jadranka BarešiË
Rudjer BoškoviÊ Institute, P.O.Box 180, 10002 Zagreb, Croatia.

Natural cosmic-ray produced radiocarbon 14C is a part of the CO2 cycle and through the atmosphere-
biosphere-hydrosphere exchange process has been introduced to the environment. Anthropogenic
disturbance of the natural 14C activity caused by nuclear tests, nuclear power plants and fossil fuel
combustion has been recorded in the atmospheric CO2. In the karst environment, which includes
water, carbonate sediments (speleothem, tufa, lake sediment), soil, and vegetation (terrestrial and
aquatic plants), the exchange processes within the CO2—dissolved inorganic carbon (DIC)—
CaCO3 system play an important role. This system is very sensitive to any change in the atmosphere,
and response to the contamination is very fast. δ13C values of each material give additional informa-
tion of carbon origin and geochemical processes in the system.
Few examples of isotopic studies in karst environment from several sites in Croatia and Slovenia
will be presented: 1) the exchange processes of 14C and 13C between atmospheric CO2, DIC in water
and recent carbonate precipitated in surface water in form of tufa and lake sediment in the Plitvice
Lakes and Korana River; 2) 14C activity in recent tufa in Krka and Zrmanja rivers in Croatia, and
Krka River in Slovenia; 3) 14C activity in recent speleothems from caves of the Dinaric Karst in
Croatia and the Postojna Cave in Slovenia; 4) distribution of 14C activity in surface soil in Plitvice
Lakes and Postojna area; 5) 14C activity of atmospheric CO2 inside and outside the Postojna Cave,
6) 14C activity in recent plants (terrestrial and aquatic) and shells. Environmental samples from dif-
ferent sites were collected in the period 1974–2008. Distribution of 14C in the karst environment will
be compared with 14C activity of atmospheric CO2 that has been continuously monitored in monthly
samples in Zagreb area. The response of different environmental materials to the bomb 14C activity
will be discussed.

#71

DATING HUMAN OCCUPATION ON DIATOM-PHYTOLITH-RICH SEDIMENT: STUDY CASE

MUSTANG SPRING AND LUBBOCK LAKE, TEXAS, USA
Christine Hatté1, Gregory Hodgins2, Vance T Holliday3, A J Timothy Jull2
1. Laboratoire des Sciences du Climat et de l’Environnement, UMR CEA-CNRS-UVSQ 1532, Domaine du CNRS, F-91198
Gif-sur-Yvette, France.
 Abstracts 47

2. NSF Arizona AMS Laboratory, Physics Building, University of Arizona, 1118 E Fourth St., Tucson, AZ 85721, USA.
3. Departments of Anthropology and Geosciences, University of Arizona, Tucson, AZ 85721, USA.

The Great Plains of North America has a rich archaeological record that spans the period from the
Late Glacial to the Historic, a period of time that also witnessed significant changes in climate and
ecology. Chronometric dating of archaeological sites in many areas of the Great Plains, however, is
often problematic, largely because charcoal and wood—the preferred material for radiocarbon dat-
ing—are scarce in this grassland environment with few trees. Two reference archaeological sites are
here studied: Mustang Spring and Lubbock Lake, Texas, USA. We carry out a geochronological
approach based on a cross study of carbon-derived data: combustion yield, δ13C, 14C ages difference
between high temperature and low temperature released carbon and 14C age itself. These analyses
within the geological context allow us to draw a succession model of dry and wet episodes and to
associate it to a chronological framework. In this way, we can assert for Mustang Spring site, that
several human occupations existed from ca. 11 kyr BP to ca. 8.7 kyr BP along the 110-cm-long
series with an interruption of ~150 years associated to palustrine environment between the Plain-
view and Firstview occupation.

#72

RADIOCARBON-ESTIMATED RATES OF NORTHERN PEATLAND EXPANSION AND THEIR

SIGNIFICANCE TO THE LATE-HOLOCENE RISE OF ATMOSPHERIC METHANE
Atte Korhola1, Meri Ruppel1, Heikki Seppä2, Minna Väliranta1, Tarmo Virtanen1,
Jan Weckström1
1. Environmental Change Research Unit (ECRU), Department of Biological and Environmental Sciences, University of Hel-
sinki, P.O. Box 65 (Viikinkaari 1), 00014 Helsinki, Finland.
2. Department of Geology, University of Helsinki, P.O. Box 64 (Gustaf Hällströminkatu 2a), 00014 Helsinki, Finland.

Methane in the Earth’s atmosphere is an important greenhouse gas, so far accounting for about 20%
of the global warming caused by human activity. According to ice cores, atmospheric concentrations
of methane (ACM) varied significantly during the Holocene, the causes of which are not completely
understood. In particular, the reasons for the increased ACM during the late Holocene (from ~5 kyr
onwards) have been debated widely, including an anthropogenic explanation. Initially, this increase
was associated with increased emissions from northern wetlands, but estimated peat initiation rates
seem not to support the conclusion. Based on a new data set of 784 basal peat radiocarbon dates that
accounts for more properly the horizontal growth dynamics of northern peatlands (by containing
only sites with multiple basal dates per site), we show here that the most extensive lateral expansion
of high-latitude peatlands occurred only after 5 kyr, parallel with the rise of CH4 in the ice cores.
This explosive increase in the extent of peatlands took place at a time when the mires were still
largely moist minerotrophic fens that emit high amounts of CH4. Hence, northern peatlands cannot
be neglected when seeking cause(s) for the late Holocene rise in CH4. A similar event in future could
worsen climate change by causing a rapid shift in atmospheric greenhouse gas concentrations.
48 Abstracts

#73

SEDIMENT MOBILIZATION IN THE BAY OF BENGAL DURING LGM TO HOLOCENE

TRANSITION
R Bhushan1, A J T Jull2, G S Burr2
1. Physical Research Laboratory, Ahmedabad 380009, India.
2. NSF-AMS Facility, University of Arizona, Tucson, AZ, USA.

One of the major components of the tropical climate system is the Indian summer monsoon, which
results in a differential land-sea sensible heating, producing seasonal reversals in wind direction and
intense rainfall. The Indian summer monsoon represents an important factor driving weathering and
erosion of the Himalayan and Burmese mountain ranges. On geological time scales, changes in the
strength of the summer and winter monsoon rainfall are likely to affect the chemical and physical
weathering intensity of the Ganga–Brahmaputra and the Irrawaddy catchment and their drainage
areas. Sediments of the Bay of Bengal and the Andaman Sea thus provide a record of the variability
of the intensity of erosion of the Himalayan and Burmese mountain ranges that in turn relate to pale-
oclimatic and paleoenvironmental variations that have affected the dynamics of the monsoon. Two
cores were radiocarbon dated using accelerated mass spectrometry (AMS) and were analyzed to
reconstruct the Late Quaternary climatic history spanning the last 50 kyr using geochemical and iso-
topic parameters to ascertain the spatial and temporal changes in sediment provenance.
In the central Bay of Bengal, a discernible increase in the sedimentation rate around 40 kyr is
observed. A prominent increase is also observed in the mass accumulation rate after 20 kyr (post-
LGM), i.e. between 18 kyr and 14 kyr, implying sediment mobilization during the transitional cli-
matic condition following the LGM. In the southern Bay of Bengal, a monotonous sedimentation
rate is seen except between 17 kyr and 15 kyr, when a marginal increase is observed. Increased sed-
iment flux in the Bay of Bengal occurred during the transitional climatic condition between the
LGM and Holocene. This is attributed to the combination of low sea level, exposed coastal zone,
and the strengthening of southwest summer monsoon. A general decrease in the ratios of detrital
proxies since the LGM to the present is due to increased productivity as a result of gradual strength-
ening of the southwest summer monsoon. It has been observed that during periods of weakened
northeast winter monsoon, contribution is dominant from the Irrawaddy River, whereas during
enhanced southwest summer monsoon periods, the Himalayan source became dominant. Results
from the two cores would be discussed.

#74

RADIOCARBON DATING OF SMALL TERRESTRIAL GASTROPODS IN NORTH AMERICA

Jeffrey S Pigati1, Jason A Rech2, Jeffrey C Nekola3
1. US Geological Survey, Denver Federal Center, Box 25046, MS-980, Denver, CO 80225, USA.
2. Department of Geology, Miami University, Oxford, OH 45056, USA.
3. Department of Biology, University of New Mexico, Albuquerque, NM 87131, USA

Fossil shells of small terrestrial gastropods are commonly preserved in wetland, alluvial, loess, and
glacial deposits, as well as in sediments at many archaeological sites. These shells are composed
largely of aragonite (CaCO3) and could be used for radiocarbon dating, but they must meet two cri-
 Abstracts 49

teria before their 14C ages can be considered to be reliable: (1) when gastropods are alive, the 14C
activity of their shells must be in equilibrium with the 14C activity of the atmosphere, and (2) after
burial, their shells must behave as closed systems with respect to carbon. We conducted a thorough
examination of the 14C content of the most common small gastropods (<10 mm in maximum dimen-
sion) in North America. Here we report the results of nearly 250 14C analyses of modern shell mate-
rial (3754 individual shells) from 46 species of gastropods that were collected live from habitats on
carbonate terrain; thus, the data represent “worst-case scenarios.” In all, ~78% of the shells did not
contain any dead carbon from rocks even though it was readily available at all sites, which included
shells from 20 different species. Approximately 18% of the gastropod shells contained less than 5%
dead carbon and the remaining few aliquots contained 5–15% dead carbon. Fossil shells of small
terrestrial gastropods recovered from well-dated, late-Pleistocene sediments in the American Mid-
west yielded 14C ages that were as precise and accurate as 14C ages from well-preserved plant mac-
rofossils, which suggests that the shells behave as closed systems with respect to carbon over geo-
logic timescales in humid environments. Taken together, our results indicate that fossil shells of
some small terrestrial gastropods, including at least five common genera, Catinella, Columella, Dis-
cus, Gastrocopta, and Succinea, should yield reliable 14C ages for the last 25,000 years, regardless
of the geologic substrate, environmental conditions, climate, or ecological habitat.

Parallel Session 3-6: Eurasian Archaeology II

#75

RADIOCARBON CHRONOLOGY OF THE EARLIEST POTTERY AND ENVIRONMENTAL

CHANGES DURING THE LATE GLACIAL IN THE EASTERN JAPAN
Yuichiro Kudo
National Museum of Japanese History, 117 Jonai-cho, Sakura City, Chiba, Japan.

The author examined the temporal correspondence between the geochronological settings and
archaeological chronology during the late glacial using calibrated radiocarbon dates of archaeolog-
ical sites in order to reconsider the human-environment interactions in eastern Honshu Island, Japan.
The archaeological chronology during this period has been roughly divided into five phases: 1)
microblade industry, 2) biface industry and plain pottery group, 3) linear-relief (Ryuki-sen-mon)
pottery group, 4) nail-impressed (Tsume-gata-mon) and cords-marked (Ta-jo-mon) pottery group,
and 5) cord-wrapped-stick-pattern (Yoriito-mon) pottery group. Based on calibrated 14C dates, the
beginning pottery use seems to be placed around 17–15 kyr cal BP, preceding the abrupt warming
and vegetation changes of the late glacial. It is generally explained that the beginning of pottery use
enabled the use of many foods found in cool-temperate deciduous broadleaved forest such as nuts,
acorns, and so on. Contrary to such a common perception, the emergence and spread of pottery and
the expansion of cool-temperate deciduous broadleaved forest shows no obvious causal connection
in the Japanese islands. The succeeding linear-relief pottery group seems to coincide with the
warmer period (ca. 15–13 kyr cal BP). The number of sites with linear-relief pottery increased much
more than the plain pottery phase. The nail-impressed and cords-marked pottery group seems to cor-
respond to the Younger Dryas cooling period.
50 Abstracts

Comparison between relative archaeological frameworks and geochronological ones is rather com-
prehensive. Better dating of the archaeological sequence is needed to obtain a more reliable picture
of the detailed correspondences between geochronological changes and cultural changes.

#76

RADIOCARBON DATING OF THE GUCHENGZHAI SITE IN CENTRAL CHINA

Xiaohong Wu2, Kexin Liu1, Quanfa Cai3, Xingfang Ding1, Dongpo Fu1, Yan Pan2, Zhiyu Guo1
1. State Key Laboratory of Nuclear Physics and Technology & School of Physics, Peking University, Beijing, 100871, China.
2. School of Archaeology and Museology, Peking University, Beijing, 100871, China.
3. Henan Institute of Cultural Relics and Archaeology, Zhengzhou, Henan, 450000, China.

Guchengzhai means the village around an ancient city wall. The Guchengzhai site is located on a
highland on the east side of the Qin River, 35 km from Xinmi City, Henan Province, China. Large
houses and the verandas around the houses were discovered at the site. Large city walls and gates to
the north and south were found outside of the structure’s area. The walls and gates were arranged in
a rectangular shape with west and east sides longer than north and south sides. There are small sites
around this site, located in a large area about 0.3 km2. The depth of the cultural deposits at this site
is about 1 meter. Pottery, glazed pottery, stone tools, jade wares, oracle bones, and other animal
bones were excavated. The oracle bones, a kind of special jade article (“Cong”), and sacrificial
places show the complexity of the society at that time. The archaeological cultures of the site are
mainly of the late Longshan period according to the types of the excavated pottery. Radiocarbon dat-
ing of this site gave an age duration of about 2200–2000 BC. It shows that the site belongs to the
transition period from Longshan culture to Erlitou culture. Because this region is located in the cen-
tral area of influence of the Xia Dynasty, research of this site is important to understand the estab-
lishment of the Xia Dynasty and the development of Chinese civilization.

#77

PRECISE CHRONOLOGY OF PREHISTORIC JAPAN BY COMBINATION OF RADIOCARBON

DATING AND POTTERY TYPOLOGY
Minoru Sakamoto1,2, Shin’ichiro Fujio1,2, Ken’ichi Kobayashi3, Hiromasa Ozaki1,4,
Mineo Imamura1,2
1. National Museum of Japanese History, The National Institutes for Humanities, Japan.
2. Department of Japanese History, The Graduate University for Advanced Studies, Japan.
3. Faculty of Letters, Chuo University, Japan.
4. AMS Dating Facility, Paleo Labo Co. Ltd., Japan.

AMS radiocarbon dating of the prehistoric Japanese Archipelago has been intensively carried out in
the course of a research project on the chronology of Jomon and Yayoi periods. We focused on
charred material on potsherds as well as charcoal and carbonized seeds. Such material is considered
as food residue or soot of firewood and should represent the age of pottery usage. Generally, the
results are harmonious with the order of pottery sequence established by archaeological inspection.
 Abstracts 51

Ancient earthenware in Japan is typologically classified and its classification has been frequently
used as the means of chronology in prehistoric Japan. Each typological classification is well exam-
ined in terms of chronological and regional relationship. Although its absolute age remains uncer-
tain, the resolution of relative age is sometimes even higher than that of calibrated age, since the
radiocarbon calibration curve is statistically smoothed and there are some wiggles as well. More-
over, possible local offsets of the curve cannot be ignored. In this study, we discuss the precise chro-
nology of the Jomon, Yayoi, and Kofun periods via a combination of 14C dates and pottery typology.

#78

THE ORIGIN OF MILLET CULTIVATION IN JAPAN: COMPARISON WITH IRRIGATED RICE

CULTIVATION
Y Miyata1,4, M Sumida1, Z Zhao2, H Matsuzaki3, T Nishimoto1
1. National Museum of Japanese History, National Institutes for Humanities, Japan.
2. Institute of Archaeology, Chinese Academy of Social Sciences, China.
3. Department of Nuclear Engineering and Management, School of Engineering, The University of Tokyo, Japan.
4. Present address: Center for Chronological Research, Nagoya University, Japan.

Recent archaeobotanical achievements in Japan suggest that millet (Setaria italica, Echinochloa uti-
lis, and Panicum miliaceum) cultivation preceded the adoption of irrigated rice cultivation, which
occurred in the Yayoi period (10th century BC to 3rd century AD) from the Korean Peninsula. We
present the results on radiocarbon dating of archaeological millets in light of the origin of millet cul-
tivation in Japan. This include the world’s oldest domesticated broomcorn millet (Panicum milia-
ceum) excavated from the Xinglonggou site (China), “Jomon barnyard grass” (Echinochloa crus-
galli) from the Tominosawa 2 site (Aomori, Japan), the oldest remains of common millet found in
western Japan from the Ryugasaki A site (Shiga, Japan), and millet samples from storage pits in the
Yayoi period, as well as charred materials on potsherds from the Jomon to Yayoi periods. Particu-
larly noted are many barnyard grass grains excavated from dwelling pit No. 321 of the Tominosawa
2 site (Middle Jomon, 2900–2600 cal BC). Although it is not well known whether these grains are
wild or domesticated, this plant seems to have used frequently in this stage. Except for the “Jomon
barnyard grass” described above, there is no reliable evidence so far for millet utilization in the
Jomon period. Therefore, we also review the dates of charred remains on potsherds, which are
thought to be C4 plants based on stable isotopic analysis. The dates of millets are compared with the
date of the adoption of irrigated rice cultivation in order to discuss the routes and times of their dif-
fusion into the Japanese Archipelago.

#79

AMS 14C DATING OF PLANT REMAINS FROM ARCHAEOLOGICAL SITES IN THE WESTERN
LIAOHE RIVER BASIN, NORTHEASTERN CHINA
Y Y Li and L P Zhou
Laboratory for Earth Surface Processes, Department of Geography, Peking University, Beijing 100871, China.

A variety of bio-remains from archaeological sites have been used for dating. These materials
include macrofossil charcoal, microfossil charcoal, pollen, phytoliths, and animal bone may have
52 Abstracts

different origins, and hence represent different events of the past. Few studies have been undertaken
to evaluate the dating results based on different materials in terms of the accuracy required in the
archaeological context. Here, we present preliminary results of an ongoing systematic comparison
of AMS 14C dating results of different materials from a wide range of archaeological sites in western
Liaohe River Basin, NE China. This area is known to have witnessed one of the earliest civilizations
in China. Numerous archaeological sites have been found and dated by 14C. Our dating results on
plant remains from 18 archaeological sites show that the dates of macrofossil charcoal overlap well
with culture stages; macrofossil charcoal from different archaeological sites with the same culture
character gave the same ages even these macrofossil charcoals were identified to different species
under SEM. The advantage of the detailed study of macrofossil charcoal structures is that accurate
information about the environment may be obtained. This information is useful not only for pale-
oenvironmental reconstruction but also for evaluation of age consistency among archaeological
sites. The interpretation of the ages based on phytolith, microfossil charcoal and pollen may be more
complicated due to the potential variability in their origins. This will be illustrated by some exam-
ples of the dating studies at the archaeological sites and some Quaternary sedimentary profiles.

#80

RADIOCARBON DATING OF CHARCOAL AND BONE COLLAGEN ASSOCIATED WITH EARLY

POTTERY AT YUCHANYAN CAVE, HUNAN PROVINCE, CHINA
E Boaretto1,2, X Wu3, J Yuan4, O Bar-Yosef5, V Chu2, Y Pan3, K Liu6, D Cohen7, T Jiao8, S Li3,
H Gu4, P Goldberg9, S Weiner10
1. Department of Land of Israel Studies, Bar Ilan University, 52900 Ramat Gan, Israel.
2. Radiocarbon Dating and Cosmogenic Isotopes Laboratory, Kimmel Center of Archaeological Science, Weizmann Institute
of Science, 76100 Rehovot, Israel.
3. School of Archaeology and Museology, Peking University, Beijing 100871, China.
4. Hunan Institute of Archaeology, China.
5. Anthropology Department, Harvard University, Cambridge, MA 02138, USA.
6. School of Physics, Peking University, Beijing 100871, China.
7. International Center for East Asian Archaeology and Cultural History, Boston University, Boston, MA 02215, USA
8. Department of Anthropology, Bishop Museum, Honolulu, Hawaii, USA.
9. Department of Archaeology, Boston University, Boston, MA 02215, USA.
10. Department of Structural Biology and the Kimmel Center for Archaeological Science, Weizmann Institute of Science,
76100 Rehovot, Israel.

Yuchanyan Cave in Daoxian County, Hunan Province (China), yielded fragmentary remains of two
or more ceramic vessels, in addition to large amounts of ash, a rich animal bone assemblage, cobble
and flake artifacts, bone tools, and shell tools. The artifacts indicate that the cave was a Late Paleo-
lithic foragers’ camp. Here we report on the radiocarbon ages of the sediments based on analyses of
charcoal and bone collagen. The best-preserved charcoal and bone samples were identified by pre-
screening in the field and laboratory. The dates range from around 13,800 to 21,000 cal BP. We show
that the age of the ancient pottery ranges between 15,430 and 18,300 cal BP. Charcoal and bone col-
lagen samples located above and below one of the fragments all produced dates of ~18,000 cal BP.
These ceramic potsherds therefore provide some of the earliest evidence for pottery making in China.
 Abstracts 53

Parallel Session 2-7: Freshwater and Groundwater

#81

EFFECT OF STORAGE ON THE RADIOCARBON AND STABLE CARBON AND NITROGEN ISO-
TOPIC SIGNATURES IN RIVERINE DOM
P Gulliver1, S Waldron, C L Bryant, E M Scott
1. NERC Radiocarbon Facility (environment), SETP, Rankine Avenue, East Kilbride, North Lanarkshire, G75 0QF, UK.

Chemical preservatives (e.g. HgCl2) are frequently added to freshwater samples to prevent biologi-
cal activity, with the aim of preserving the isotopic signature of dissolved organic matter (DOM)
over time. However, alternative preservation methods are needed due to regulations restricting the
use of preservatives with potentially adverse environmental and health impacts. This study was
designed to investigate whether a chemical preservative is needed to maintain the radiocarbon and
stable carbon and nitrogen signatures of freshwater DOM from a river system draining a peaty
catchment.
Fifty liters of first-order stream water was collected into one plastic carbuoy, within 24 hrs. One-liter
aliquots from the 50-L sample were transferred to acid-washed plastic bottles. Five aliquots were
analyzed straight away to determine the baseline values for 14C (%mc), 13C (‰VPDB), 15N (‰), %C
(mg L–1), and %N (mg L–1). Of the remaining subsamples, 20 were frozen and a further 20 refriger-
ated at <40 °C. After 7, 30, 90, and 180 days, 5 frozen and 5 refrigerated aliquots were analyzed in
the same manner as the baseline aliquots. Analysis of the results shows that length of storage has an
effect on some of the parameters measured and that the effect may depend on storage method.

#82

FRESHWATER RESERVOIR EFFECTS IN AN ICELANDIC ECOSYSTEM

P L Ascough1, G T Cook1, M J Church2, E Dunbar1, A Einarsson3, T H McGovern4,
A J Dugmore5, S Perdikaris4
1. SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, Scotland, UK.
2. Department of Archaeology, University of Durham, South Road, Durham, DH1 3LE, England, UK.
3. Mývatn Research Station, Skutustir, Iceland.
4. Hunter Bioarchaeology Laboratory, Hunter College CUNY, New York, NY 10021, USA.
5. Institute of Geography, School of GeoSciences, University of Edinburgh, Drummond Street, Edinburgh EH9 8XP, Scot-
land, UK.

Lake Mývatn is an interior highland lake in northern Iceland, forming a unique ecosystem of inter-
national scientific importance that is surrounded by a landscape rich in archaeological and paleoen-
vironmental sites relating to the Norse settlement (1,2). A significant 14C age offset exists between
contemporaneous human and terrestrial herbivore bones at some Norse (ca. AD 870–1000) sites in
the wider region of Mývatnssveit. This may partly reflect the marine 14C reservoir effect (MRE), due
to consumption of marine resources by the settlers, and previously a ΔR of between +111 ± 14 and
+132 ± 21 14C years was established for the period AD 890–1400 (3). However, the size of the age
offset and the fact that the stable isotope (δ13C and δ15N) ratios of the human bone do not reflect a
dominantly marine diet suggests the existence of a freshwater 14C reservoir effect (FRE). Previously
reported AMS measurements indicated an FRE of ~1500–1900 14C years in Mývatn Lake carbon
using archaeological samples (3). A new study has used stable isotope and 14C measurements to
54 Abstracts

quantify the Mývatn FRE for both the Norse and modern periods. This work has identified a tempo-
rally variable FRE greatly in excess of previous assessments. New, paired samples of Norse cattle
and pig bones demonstrate that at least some pig diets incorporated freshwater resources, resulting
in an age offset of up to 400 14C yrs, and Norse period freshwater fish indicate a FRE of >4000 14C
yrs. Modern samples of benthic detritus, aquatic plants, zooplankton, invertebrates, and freshwater
fish indicate an FRE in excess of 5000 14C yrs in some species. Likely mechanisms for this large
FRE are discussed, along with implications for both chronological reconstruction and integrated
investigation of stable and radiogenic isotopes.
1. Einarsson et al. 2004. Aquatic Ecology 38:317–348.
2. McGovern et al. 2007. American Anthropologist 108:27–51.
3. Ascough et al. 2007. Radiocarbon 49(2):947–961.

#83

HIGH CONTRIBUTION OF RECALCITRANT ORGANIC MATTER TO DISSOLVED ORGANIC

CARBON (DOC) IN JAPANESE OLIGOTROPHIC LAKE REVEALED BY RADIOCARBON MEA-
SUREMENTS
Fumiko Watanabe Nara1,4, Akio Imai1, Masao Uchida1, Kazuo Matsushige1, Kazuhiro Komatsu1,
Nobuyuki Kawasaki1, Yasuyuki Shibata1, Kunihiko Amano2, Hajime Mikami3, Ryuji Hanaishi3
1. National Institute for Environmental Studies, 16-2 Onogawa, Tsukuba, Ibaraki 305-8506, Japan.
2. Public Works Research Institute, 1-6 Minamihara, Tsukuba City, Ibaraki 305-8516, Japan.
3. Aomori Prefectural Institute of Public Health and Environment, Aomori, Aomori 030-8566, Japan.
4. Present address: Graduate School of Science, Tohoku University 980-0845, Japan.

Carbon isotopes (14C and 13C) of dissolved organic carbon (DOC) in a Japanese oligotrophic lake
(Lake Towada) were measured to study the origin and cycling of dissolved organic matter (DOM)
in Lake Towada. Lake water samples were collected at three depths (0 m, 30 m, and 80 m, or 85 m)
in 4 months (April, June, August, and October) in 2006. The measurements of 14C for DOC were
performed by an accelerator mass spectrometer (AMS) at the National Institute for Environmental
Studies (NIES-TERRA) in Japan. Δ14C and δ13C values of DOC in Lake Towada showed light car-
bon isotopic values ranging from –750 to –514‰ and –29.0 to –27.8‰, respectively. These values
are similar to those of humic substances reported. The very low carbon isotopic values of DOC in
Lake Towada suggest a very small contribution of DOC derived from fresh phytoplankton to the
lake DOC. Furthermore, it is implied that most of the phytoplankton-derived DOC is rapidly decom-
posed in the water column of Lake Towada, and the DOC in Lake Towada is mainly composed of
recalcitrant organic matter such as humic substances. There is an extremely high linear relationship
between the Δ14C and δ13C of DOC in Lake Towada when all data points are plotted (r2 = 0.818,
p<0.01), suggesting that the DOC in Lake Towada have two specific sources between heavy and
light carbon isotopes. Although the freshly produced DOC of phytoplankton origin can be decom-
posed easily, the variation in the autochthonous DOC should influence the carbon isotopic values of
DOC in Lake Towada.
 Abstracts 55

#84

VARIATIONS IN Δ14C OF POC IN WATERS FROM THE UPPER TO LOWER ISHIKARI RIVER
Seiya Nagao1, Masato Mikami2, Takayuki Tanaka3
1. Low Level Radioactivity Laboratory, KU-INET, Kanazawa University, Japan.
2. Graduate School of Environmental Science, Hokkaido University, Japan.
3. Aomori Research and Development Center, Japan Atomic Energy Agency, Japan.

The dynamics of bulk particulate organic carbon (POC) discharged from rivers are important to
understand the role of rivers in global biogeochemical cycles. We have been investigated dynamics
of riverine POC at the Ishikari River system, which is 268 km long from the source to mouth with a
drainage basin of 14,330 km2, in northern Japan. The Ishikari River has two peaks of water dis-
charge at spring snowmelt and autumn rain events. To better understand the origin, forms, and fate
of POC, it is important to study the characteristics of riverine POC because of its heterogeneous
mixtures of organic matter.
This study reports the variations in Δ14C values of POC at three observation sites from the Upper to
Lower Ishikari River during spring to autumn in 2006. Riverine suspended particles were collected
with a single-bowl continuous-flow centrifuge. The organic 14C measurements were performed by
accelerator mass spectrometry at the Aomori Research Center of Japan Atomic Energy Agency.
Variations in Δ14C values of riverine POC are different from each station. In the Lower Ishikari, the
Δ14C values were –260‰ for the spring snowmelt season, and –148‰ to –122‰ for the normal flow
condition. On the other hand, POC of the Middle site shows opposite variation trend. The POC had
Δ14C of –47‰ and –148‰ to –122‰ at the spring snowmelt and low water discharge, respectively.
In the upper Ishikari, the Δ14C values were +19‰ for the spring snowmelt season, but –148 and
+208‰ for the normal flow condition. The variation range of Δ14C value is in the order of the Middle
≤ Lower < Upper stations during spring to autumn. These results suggest that the sources and supply
processes of POC from watershed are different from the Upper, Middle, and Lower Ishikari River.

#85

RADIOCARBON RESERVOIR CORRECTIONS FOR FRESHWATER FISH IN BRITAIN AND

IRELAND
Evelyn M Keaveney, Paula J Reimer, Gerry McCormac
Centre for Climate, the Environment & Chronology (14CHRONO), School of Geography, Archaeology and Palaeoecology,
Queen’s University Belfast, Northern Ireland, UK.

The importance of fish in past diets leads to problems when attempts are made to ascertain the age
of the consumers of an aquatic diet. Radiocarbon reservoir effects lead to age offsets when the bone
of these consumers is radiocarbon dated. The aim of this project is to investigate whether a method-
ology can be developed to allow freshwater reservoir effects to be predicted from measurements on
modern fish bone and water samples, especially in the absence of well-preserved archaeological fish
bone. Modern and archaeological samples were radiocarbon dated using accelerator mass spectrom-
etry (AMS). Results showed reservoir age offsets ranging from 1 to 1638 14C yrs, which were related
to lake alkalinity in Irish sites, although variability was high, potentially from ecological differences.
56 Abstracts

Parallel Session 3-7: 14C Methods Including Archaeological Dating

#86

SUCCESSFUL AMS 14C DATING OF NON-HYDRAULIC LIME MORTARS FROM THE MEDIEVAL
CHURCHES OF THE ÅLAND ISLANDS
Jan Heinemeier1, Åsa Ringbom2, Alf Lindroos3, Árny E Sveinbjörnsdóttir4
1. AMS 14C Dating Centre, Aarhus University, DK-8000 Aarhus C, Denmark.
2. Art History, Åbo Akademi University, Finland.
3. Geology and Mineralogy, Åbo Akademi University, Finland.
4. Institute of Earth Sciences, University of Iceland, IS-101 Reykjavík, Iceland.

Fifteen years of research on AMS 14C dating of non-hydraulic mortar has now led to the establish-
ment of a chronology for the medieval stone churches of the Åland Islands (Finland), where no con-
temporary written records could shed light on the first building phases. In contrast to other material
for dating, well-preserved mortar is abundantly available from every building stage.
We have gathered experience from AMS dating of some 130 Åland mortar samples. Two thirds of
these have age control from dendrochronology or from 14C analysis of wooden fragments in direct
contact with the mortar. Of the samples with age control, 96% of the results agree with the known
age.
We have been able to define criteria of reliability to interpret the results when mortar dating is the
only possibility to constrain the buildings in time. Originally, the samples were analyzed in two suc-
cessive CO2 fractions evolved from acid dissolution, but later in five CO2 fractions, to form an age
profile reflecting the entire dissolution process. Highest reliability, or Criterion I, was defined when
the two first CO2 fractions yielded the same result, thus indicating the absence of both readily soluble
re-crystallizations and contaminating effects from slowly dissolving unburned limestone. Alterna-
tively, in a multi-fraction profile a horizontal plateau reflects a contamination-free phase in the dis-
solution process. Criterion II, almost equally convincing, occurs when the first CO2 fractions in three
or more samples from the same building unit yield the same results. Criterion II reveals the presence
of contamination of unburned limestone as varying results for the second fractions, but also that the
acid hydrolysis has been successful in eliminating the effects in the first fractions. With these crite-
ria, 86% of all samples reached conclusive results, and we have thus so far been able to establish the
chronology of 13 out the 14 Åland churches and chapels, while one remains to be analyzed.

#87

LIME PLASTER, MORTAR, AND WOOD ASH: MATERIALS CHARACTERIZATION AND POTEN-
TIAL FOR RADIOCARBON DATING
L Regev1, S Weiner2, E Boaretto3
1. Radiocarbon and Cosmogenic Isotopes Laboratory and Department of Structural Biology, Weizmann Institute of Science,
Rehovot, Israel.
2. Kimmel Center of Archaeological Science and Department of Structural Biology, Weizmann Institute of Science, Rehovot,
Israel.
3. Department of Land of Israel Studies and Archaeology, Bar Ilan University, Israel; and Radiocarbon and Cosmogenic Iso-
topes Laboratory, Weizmann Institute of Science, Rehovot, Israel.

Most of the radiocarbon dating in archaeology is based on charred materials and bone collagen. In
many cases the relation between the sample and the archaeological context is not very secure. Lime
 Abstracts 57

plaster, mortar and wood ash are common in many archaeological sites, and are often present in
solid archaeological contexts. Even though these materials all contain radiocarbon from the atmo-
sphere, they are difficult to date due to their complex nature and tendency to undergo diagenetic
change that can alter their 14C contents. Here we propose new approaches for dating well preserved
plaster, mortar and wood ash that are based on the integration of a variety of structural parameters
that characterize the preservation and purity of the archaeological materials. The techniques used are
mainly FTIR, SEM-EDS, and radiocarbon dating. Although challenging technical problems remain,
the results obtained clarify better the conditions under which the original radiocarbon signal is pre-
served, and highlight just how complex the diagenetic processes are that these materials undergo.

#88

HIGH-PRECISION RADIOCARBON DATING OF THE PRIMARY CONSTRUCTION PHASE OF

OAKBANK CRANNOG, LOCH TAY, PERTHSHIRE
G T Cook1, T N Dixon2,3, N Russell1, P Naysmith1, S Xu1, B Andrian3
1. SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, Scotland, UK.
2. Archaeology, School of History, Classics and Archaeology, University of Edinburgh, Infirmary Street, Edinburgh EH1
1LT, Scotland, UK.
3. Scottish Trust for Underwater Archaeology, Scottish Crannog Centre, Kenmore, Perthshire, PH15 2HY, Scotland, UK.

At the last conference, we presented 14C ages for 13 of the 18 crannogs situated in Loch Tay (Dixon
et al. 2007). Oakbank Crannog, which is the most extensively studied, used to be considered by
itself, but it became clear from the study that all 13 sites originated in the latter half of the 1st mil-
lennium BC or the 1st or 2nd centuries AD. Within that group it is possible to see divisions, with
nine sites ranging within the period 820–350 BC, two within the period 400–50 BC, and two from
170 BC–AD 180.
Unfortunately, since many of the crannogs were built in the Early Iron Age, calibration of the 14C
ages produces very broad calendar age ranges due to the well-documented plateau in the calibration
curve. However, the large oak timbers that were used in the primary construction phase of some of
the crannogs potentially provide a means of improving the precision of the dating through sub-
dividing them into decadal or subdecadal increments, dating them to high precision and wiggle-
matching the resulting data to the master 14C calibration curve. We obtained one oak timber from
Oakbank comprising 70 rings (sample OB06 WMS 1) including sapwood that was complete to the
bark edge. This was sectioned into 5-year increments and dated to a precision of between 1 and 2‰.
We present these results and those of a standard, the latter being used to justify this precision. The
wiggle-match predicts that the last ring dated was formed around 500 BC (maximum range of 520
to 465 BC) and should be taken as indicative of the time of construction of Oakbank Crannog. This
is a considerable improvement on the estimates based on single radiocarbon ages made on oak sam-
ples, which typically encompassed the period from around 800 to 400 BC.

REFERENCE
Dixon TN, Cook GT, Andrian B, Garety LS, Russell N, Menard T. 2007. Radiocarbon dating of the crannogs of Loch Tay,
Perthshire. Radiocarbon 49(2):673–684.
58 Abstracts

#89

IMPROVED AMS RADIOCARBON DATING OF SHELL CARBONATES USING HIGH-PRECISION

XRD AND DENSITY SEPARATION
Katerina Douka1, Robert E M Hedges1, Thomas F G Higham2
1. Research Laboratory for Archaeology and the History of Art, University of Oxford, Dyson Perrins Building, South Parks
Road, Oxford, OX1 3QY, UK.
2. Oxford Radiocarbon Accelerator Unit, Research Laboratory for Archaeology and the History of Art, University of Oxford,
Dyson Perrins Building, South Parks Road, Oxford, OX1 3QY, UK.

One critical variable in the successful application of radiocarbon dating, especially to very old mate-
rial, is the effective removal of carbonaceous contaminants. In the case of marine shell carbonates,
contamination appears usually in the form of secondary calcite, the stable polymorph of CaCO3 and
byproduct of the post-mortem recrystallization of the autochthonous CaCO3 (either in the form of
calcite or aragonite). Depending on the nature of the depositional environment, the contaminant may
be contemporary in its age, or being younger or older. The limited ability of current pretreatment
protocols to remove the secondary mineralogical phases prior to dating carbonates has been the
main reason shell and coral 14C results are sometimes difficult to validate in terms of their reliability.
We have developed a new pretreatment protocol designed to achieve greater reliability and accuracy
in the dating of this material. The method involves an improved detection and quantification of sec-
ondary calcite in aragonite (at a precision of ~0.1% and ~0.6%, respectively) using XRD and SEM
methods. This is followed by a novel density fractionation step of the two polymorphs using non-
toxic heavy liquids, where required. This enables the clear separation of calcite and aragonite, with
only the latter kept for dating.
We have applied the new protocol (screening and separation) using standard and archaeological
examples, and the initial results suggest that it is successful and reproducible. We will describe the
method and preliminary results.

#90

A THERMAL AND ACID TREATMENT FOR CARBON EXTRACTION FROM CAST IRON AND ITS
APPLICATION TO AMS DATING OF CAST IRON OBJECTS FROM ANCIENT KOREA
Jang-Sik Park1 and G S Burr2
1. Department of Metallurgical Engineering, Hongik University, Chochiwon Choongnam, 339-701, Korea.
2. NSF-Arizona AMS Facility, University of Arizona, Department of Physics, Tucson, AZ 85721, USA.

A method of thermal and acid treatments was developed at the Archaeometallurgy Laboratory of
Hongik University in Korea to extract carbon from cast iron, and carbon samples thus prepared from
cast iron artifacts of ancient Korea were dated at the University of Arizona’s AMS Facility. The ther-
mal treatments consist of the heating of a specimen at around 1000 °C in a controlled environment
with reduced oxygen potential, followed by rapid cooling to room temperature. The heating causes
the cementite phase in white cast iron to be graphitized and the quenching suppresses pearlite for-
mation. The specimen then consists of flakes of graphite embedded in the matrix of martensite. The
matrix is dissolved in a solution of nitric acid plus hydrochloric acid, leaving graphite powder,
which is collected and cleansed to be made into graphite targets for AMS. This method was applied
 Abstracts 59

to some artifacts from the Korean Three Kingdoms period (ca. AD 300–668) and their AMS results
were compared with the chronology estimated by independent means.

Parallel Session 2-8: Paleoclimate Applications

#91

IS EARLY HOLOCENE OPTIMUM TIME-TRANSGRESSIVE IN THE MONSOON CHINA?:

EVIDENCE FROM HIGH-RESOLUTION PEAT RECORD
Weijian Zhou1,2,3, Peng Cheng1,2, Yizhi Zhu1, Xuefeng Lu1,2,3, George Burr4, A J Timothy Jull4,
J Warren Beck4
1. State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences,
Xi’an 710075, China.
2. Xi’an AMS Center, Xi’an 710052, China.
3. Human Settlements and Civil Engineering School of Xi’an Jiaotong University, Xi’an 710049, China.
4. NSF-AMS Laboratory, Physics Building, Department of Physics, University of Arizona, Arizona 85721, USA.

Computer models suggest that the Holocene Optimum for East Asian summer monsoon precipita-
tion occurred at different times in different regions of China. This time-transgressive Holocene
Optimum would have been experienced about 3000 yr ago in southern China. To test the time-trans-
gressive Holocene Optimum model, we have closely examined high-resolution eutrophic peat/mud
sequences covering the last 18,000 years at Dahu, Jiangxi, on the southern boundary of the mid-sub-
tropical zone in China. Late Pleistocene sequences in the Dahu record indicate cool and much wetter
conditions relative to synchronous events in north-central China. The Holocene Optimum occurs in
southern China between ca. 10,000 and 6000 cal years ago, consistent with the global pattern. Con-
ditions were relatively dry and cold from 6000–4000 cal years ago. Our data also support the con-
clusion that the last deglaciation to early Holocene in the south was much wetter, resulting in the for-
mation of dense broad-leaved forests, which could have acted to moderate land temperature ~10,000
to 6000 cal years, yielding a stable early Holocene climate. After 6000 cal years, forest reduction led
to unstable land temperatures and possibly to a northerly shift of the subtropical high pressure sys-
tem. Whatever the mechanism, it resulted in decreased precipitation between 6000 and 4000 cal
years BP in southern China. The extra moisture could then be transferred to the northern areas.

#92

CHARACTERIZATION AND DATING OF SALINE GROUNDWATER IN THE DEAD SEA AREA

Naama Avrahamov1,2, Orit Sivan1, Yoseph Yechieli2, Boaz Lazar3
1. Department of Geological & Environmental Sciences, Ben Gurion University of the Negev.
2. Geological Survey of Israel, Jerusalem, Israel.
3. The Institute of Earth Sciences, The Hebrew University, Jerusalem, Israel.

This work presents a unique attempt of dating brines and determining flow rates of saline ground-
water in the arid dynamic system of the Dead Sea (DS) using radiocarbon and other chemical and
isotope measurements. Since the last mid-century, the DS level has been decreasing by about 30 m.
60 Abstracts

The continued decrease in sea level has lowered the groundwater level, shifting the fresh-saline
groundwater interface (FSI) towards the DS, so that saline groundwater is being replaced by fresh
groundwater and ancient brines are being flushed.
In order to estimate the flow rate of groundwater in this dynamic system, we have quantified the
geochemical processes of the different water bodies in the FSI zone, focusing on carbon species and
isotopes. We did this is by sampling groundwater at the DS coastal area at variable distances from
the shore. Samples were analyzed for major ions, methane (CH4), total alkalinity, dissolved inor-
ganic carbon (DIC), δ13CDIC, 14CDIC, and tritium (3H).
Most of the major ions results indicate that the dominant process at the DS coastal area is a mixing
between the modern DS water and fresh groundwater. In contrast to the conservative behavior of
most of the ions, the alkalinity, DIC, δ13CDIC, and 14CDIC indicate additional processes. Exception-
ally low values of δ13CDIC suggest that oxidation of methane occurs in this groundwater. Methane
was indeed found in many wells and is being investigated for its age. The 14C values of some saline
groundwater samples are lower than both fresh and saline end-members, suggesting the presence of
ancient brines in the lower aquifer. The high 14C values in saline groundwater near the DS shore
(similar to DS values in the 1980s) indicate that the inflows of DS brine into the aquifer still con-
tinue even though the whole groundwater system responds very fast to the DS level drop.

#93

PRELIMINARY STUDY OF PALEOCLIMATE CHANGE OF THE LAGUNA POTROKE AIKE

USING AUTHIGENIC Be ISOTOPES
K J Kim1, J H Kim1, A J T Jull2, B Zolischka3
1. Korea Institute of Geoscience and Mineral Resources, 92 Gwahang-no, Yuseong-gu, Daejeon 305-350, Korea.
2. NSF Arizona AMS Laboratory, University of Arizona 1181 E. Fourth St. Tucson, AZ 85721, USA.
3. Geomorphology and Polar Research (GEOPOLAR), Institute of Geography, University of Bremen, Celsiusstr, FVG-M,
28359 Bremen, Germany.

Atmospheric beryllium-10 concentrations from several marine sediment cores have been investi-
gated for the study of global paleoclimate change. 10Be signals from these marine cores were found
to be strongly correlated to the oxygen isotopic records of the sediment cores (Aldahan et al. 1997;
McHargue and Donahue 2005). Recent 10Be investigations for marine sediment cores for both shal-
low sea and deep sea in the region of Korea provided interesting results toward understanding
marine climatic environmental change using 9Be records (Kim et al., submitted). With the current
issue of global warming phenomenon, studies focusing on recent climatic change using beryllium
isotopes could be much more demanding for us to be urgently involved in. Especially, the study of
regional climate change can be effectively accomplished by using multiple stratigraphic proxy
records of land materials such as lake sediment, soil deposit, sand dune, speleothems, etc. We have
investigated authigenic Be isotopes as indicators of local climatic tracer for the Laguna Potrok Aike
(a maar lake) in the Pali-Aike Volcanic Field of southern Patagonia, Argentina (51°58′S 70°23′W,
116 m asl). We processed 18 samples of the lake sediment cores (PTA-02 and PTA-03) from the sur-
face to a sediment core depth of 18.64 m to obtain preliminary stratigraphic records of Be isotopes.
Our results of 9Be data are inversely correlated to the TIC(%) of the lake sediment as similar results
observed from both marine sediment core in the Pacific Ocean and dry sediment core from the Wan-
ganui Basin in New Zealand (Bourles et al. 1989; Graham et al. 1998). We are currently waiting for
both 10Be AMS and chemical flux results of our study. These data would provide much more infor-
 Abstracts 61

mation into the atmospheric paleoclimatic change of this region. Our presentation will demonstrate
authigenic Be isotope data together with other stratigraphic proxy data available for the preliminary
scientific investigation of PASADO ICDP program to unravel paleoclimate change for the period
from the Holocene to LGM of the Laguna Potrok Aike region.

#94

PALEOENVIRONMENT AND AMS RADIOCARBON DATING BASED ON INSECT FOSSILS

FROM THE ARCHAEOLOGICAL SITES IN JAPAN
Emi Okuno1, Yuichi Mori2, Toshio Nakamura3
1. Graduate School, Nagoya University, Japan.
2. Kinjo Gakuin University, Japan.
3. Center for Chronological Research, Nagoya University, Japan.

This study presents AMS 14C dating of insect fossils (remains) from archaeological sites in Japan.
We analyzed five species of insect fossils: Hydrophilus acuminatus (aquatic beetles), Onthophagus
atripennis, Craspedonotus tibialis (ground insects), Anomala rufocuprea, and Scotinophara lurida
(phytophagous insects). These insect fossils were excavated from five archaeological sites at Oka-
jima, Asahi, Nekojima, and Ooke-oki (Aichi Prefecture), and at Onigashioya (Mie Prefecture) cov-
ering the period from the Yayoi to Medieval.
S. lurida is a typical paddy pest, and H. acuminatus is an indicator of the rice paddy field. A. rufo-
cuprea is a harmful insect of the field. O. atripennis is a dung-eating insect. These insects are closely
linked to human life (often termed “village insects”). Therefore, these fossils indicate that the envi-
ronment around these sites were modified by human activity. In addition, C. tibialis is an important
species because, based on its habitant, the sites were located around sandy areas such as riverbeds
or beaches.
Insect fossils have an important role in the reconstruction of paleoenvironment compared to other
microfossil analyses. Not only do insects respond quickly to environmental changes but also they
enable us to reconstruct more the paleoenvironment in detail. Insect fossils enable understanding of
human impact to the environment; thus, they are very important when reconstructing chronologies
and characteristics of archaeological sites.

#95

ENVIRONMENTAL CHANGES OF THE ARAL SEA (CENTRAL ASIA) IN THE HOLOCENE:

MAJOR TRENDS
Sergey K Krivonogov1, Yaroslav V Kuzmin1, G S Burr2
1. Institute of Geology & Mineralogy SB RAS, Novosibirsk, Russia.
2. University of Arizona, Tucson, AZ, USA.

The fluctuations of the Aral Sea level during the Holocene closely mirror its environmental history.
The available data used to identify major transgressive and regressive phases in the Aral Sea are
ambiguous. Data concerning sea level changes compiled by Boomer et al. (2000) are inconsistent
62 Abstracts

with results obtained by others (e.g. Oberhänsli et al. 2007). Generally, our knowledge supported by
14C dates covers no more than the last 6000 yrs. According to some of the most complete sources

(e.g. Boomer et al. 2000), major regressions of the Aral Sea level are detected at ca. AD 1500–1650
(400–300 BP), AD 900–1300 (1200–600 BP), and AD 1–400 (2200–1600 BP) over the past 2000
years. Prior to that, they occurred at ca. 7000–5500 and 3800–3600 BP. Main transgressions took
place at ca. 5000–3500, 1900, and 1600–900 BP. Our data from the 8.2-m-deep core of bottom sed-
iments drilled in the south of Aral Sea near the town of Muinoq testify to deep regressions at ca.
1700–1000 BP and 5500–4500 BP. This is problematic because high and low stands of lake levels
from different interpretations overlap.
Currently, the most reliable lake-level reconstruction covers the last 600 yrs and is possible because
of archaeological finds. The Kerderi settlements and mausoleums (mazars) were discovered at the
dry bottom of the Aral Sea about 65 km off the 1960s shoreline (water depth at that time was 22 m)
(Boroffka et al. 2005, 2006). The Puljai settlement of 14th century AD situated 70 km south of
1960s Aral shoreline was later covered by water at ca. 300 BP (Reinchard et al. 2008). Thus, for this
time interval the following sequence of Aral Sea level changes is realistic: 1) the deepest Kerderi
regression in the 14th century (ca. 29 m asl); 2) transgression in the 15th–16th centuries; ca. 54 m
asl; and 3) recent catastrophic desiccation since the 1960s. The reconstruction of the Aral Sea level
history requires more work. The current study is supported by a joint CRDF-RFBR Project (Nos.
RUG1-2921-NO-07 and 08-05-91105).

REFERENCES
Boomer I. et al. 2000. Quaternary Science Reviews 19:1259–78.
Boroffka NGO et al. 2005. Mitigation Adapt. Strateg. Glob. Change 10:71–85.
Boroffka NGO et al. 2006. Geoarchaeology 21:721–34.
Oberhänsli H. et al. 2007. Irrigation and Drainage Systems 21:167–83.
Reinhardt C. et al. 2008. Geomorphology 93:302–15.

#96

LATE PLEISTOCENE VEGETATION RECONSTRUCTION OF COASTAL FOREST-GRASSLAND

MOSAIC AT SERRA DO MAR STATE PARK, SOUTHEASTERN BRAZIL
M I Francisquini1, L C R Pessenda1, P E De Oliveira2, V B Medeiros2, A A Buso-Junior1,
J R Passarini Jr.1
1. Universidade de Sao Paulo (USP). Centro de Energia Nuclear na Agricultura (CENA) - Laboratorio de Carbono 14, Sao
Paulo, Brazil.
2. Universidade Guarulhos (UnG). Laboratorio de Geociencias, Guarulhos, Brazil.

Pollen, C/N isotopes, radiocarbon dating, and elemental analyses (total organic carbon [TOC] and
total organic nitrogen) were used in samples from a peatland located in a small depression at Serra
do Mar Park (800 m asl), SE Brazil, in order to reconstruct the late Pleistocene vegetation dynamics
of the coastal mosaic forest-grassland. Between ~16,360 and ~10,000 BP, the sedimentation rate
was of about 0.016 cm/year and only elemental and isotopic analyses were performed. The highest
TOC values (14.2%) indicated significant organic matter preservation, probably associated with a
humid climate. C/N values of 33.1 and δ13C of –25.9‰ indicate the presence of C3 plants, probably
C3 grasses found in humid terrains. From ~10,000 to ~5000 BP, the sedimentation rate and TOC
changed to 0.023 cm/year and ~7.7%, respectively; the C/N of 25.7 and δ13C of –26.4‰ suggest the
dominance of C3 plants. The pollen analyses did not indicate the presence of algae, probably asso-
 Abstracts 63

ciated with a less humid climate than the previous period. TOC values (~7.2%) were constant
between ~5000 and ~2000 BP. C/N values of 23 and δ13C of –26.7‰ indicate the dominance of C3
plants; pollen analyses shown an increase of arboreal grains, probably due to an increase in the veg-
etation density. From ~2000 BP to present, the TOC increased to 10.4% and the sedimentation rate
to 0.064 cm/yr. The C/N values of ~16.4 and the significant presence of algae were associated with
a more humid climate than the previous period. Plant species such as Podocarpus, Alchornea, Sym-
plocos, and Drymis are characteristic of the cold and humid forest present in the Holocene. These
results were not in agreement with other locations in SE Brazil, where a drier climate was observed
from the early to mid Holocene. This local humid condition during the Holocene was probably
maintained by the air masses from the Atlantic Ocean.

Parallel Session 3-8: 14C Methods and Chemistry

#97

HOW TO MEASURE SMALL SAMPLES WITH A GAS ION SOURCE

L Wacker1, M Ruff1, S Fahrni2, M Nemec1,2, S Szidat2, H-A Synal1
1. Ion Beam Physics, Paul Scherrer Institute and ETH Zurich, 8093 Zurich, Switzerland.
2. Department of Chemistry and Biochemistry, University of Bern, 3012 Bern, Switzerland.

The hybrid gas ion source of the small MIniCArbonDAtingSystem at the AMS facility at ETH in
Zurich was installed 4 years ago. Since then, we have been constantly refining and extending our gas
inlet system based on a gas-tight syringe. The gas ion source at ETH can now be used for a broad
range of applications.
First installed was an ampoule cracker for measuring ultra-small samples containing only 1–30 μg
carbon. The mostly pure carbon dioxide is supplied in glass ampoules. More than 500 samples could
be measured within the last two years under routine conditions with reliable and stable results. Later,
we added an elemental analyzer and also an aerosol combustion system for on-line measurements of
samples with a carbon content in the range of 10–150 μg. In all of our constructions, we avoid using
liquid nitrogen for concentrating the CO2. The result is a simple and versatile gas handling system
with small dead volumes.
The challenges in preparation as well as in the measurements of small samples towards a limit of
only a few micrograms carbon will be discussed. With our four years of experience, we will show
what can be done with a gas ion source at present and what may be possible in the future.
64 Abstracts

#98

HYDROPYROLYSIS: IMPLICATIONS FOR RADIOCARBON PRETREATMENT AND BLACK

CARBON QUANTIFICATION
P L Ascough1, M I Bird2, W Meredith3, C E Snape3, F Brock4, T F G Higham4, R E Wood4,
C H Vane5
1. SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, UK.
2. School of Earth and Environmental Sciences, James Cook University, Cairns, Queensland, 4870, Australia.
3. Department of Chemical and Environmental Engineering, University of Nottingham, NG7 2RD, UK.
4. Oxford Radiocarbon Accelerator Unit, University of Oxford, OX1 3QY, UK.
5. British Geological Survey, Kingsley Dunham Centre, Keyworth, Nottingham, NG12 5GG, UK.

Wildfires, domestic hearths, and industrial processes such as grain drying all can result in exposure
of biomass to elevated temperatures in restricted oxygen, known as pyrolysis, producing a wide
range of materials, including soot, char, and charcoal, collectively known as pyrogenic carbon.
These substances are extensively used in global biogeochemical research, and feature heavily in the
paleoenvironmental and archaeological record, where charcoal is one of the most common materials
submitted for radiocarbon measurement. Pyrogenic carbon is neither a pure form of carbon, nor a
uniform structural entity, but comprises a range of compositions, heavily influenced by production
conditions and starting materials [1]. This is significant because it is apparent that pyrogenic mate-
rials such as charcoal can display variable resistance to environmental diagenesis. For example, dec-
adal degradation of charcoal carbon has been observed in a Zimbabwean savannah soil [2], and fluc-
tuating groundwater has resulted in loss of charcoal at an Australian archaeological site [3].
The most inert fraction of pyrogenic carbon is known as black carbon (BC), defined operationally
as having both high aromaticity and high resistance to oxidative degradation. BC degrades
extremely slowly, making isolation of BC of great interest in radiocarbon pretreatment, particularly
in more ancient samples, where contamination issues become more critical. Quantification of BC
content within pyrogenic samples is also of great importance, providing valuable information on
factors including the proportions of carbon within labile and recalcitrant environmental reservoirs.
We present results of tests using a new method for the quantification and isolation of BC, known as
hydropyrolysis (hypy), using a selection of standard soil samples and ancient charcoals, including
some from the recent BC ring trail [4, 5]. These show controlled reductive removal of non-BC
organic components, such as lignocellulosic and humic material in samples, with strong evidence
that this is effectively complete at temperatures above 500 °C. This process appears reproducible
and rapid, making hypy a promising new approach not only for BC quantification as an end in itself,
but also for 14C measurement where purified BC is the target material for dating. These two goals
are combined in many studies, for example in efforts to establish the environmental longevity of BC,
and quantification of BC production rates over extended timescales. An important additional advan-
tage of the hypy technique is that it allows for the retention of the non-BC component of a sample,
which may then be subject to further analysis and measurement.

REFERENCES
[1] Antal et al. 2003. Ind. Eng. Chem. Res. 42.
[2] Bird et al. 1999. Global Biogeochem. Cycles 13.
[3] Bird et al. 2002. Quatern. Sci. Reviews 21: doi:10.1016/S0277-3791(01)00058-0.
[4] Hammes K. et al. 2007. Global Biogeochem. Cycles 21: GB3016, doi: 10.1029/2006GB002914.
[5] Ascough PL. et al. 2008. Quatern. Geochronol. doi: 10.1016/j.quageo.2008.11.001.
 Abstracts 65

#99

A NEW PRETREATMENT METHOD FOR MOLLUSCAN SHELL USING DENSITY FRACTION-

ATION OF CARBONATES IN BROMOFORM
Jennifer A Tripp1, Christopher M Russo1, Katerina Douka2, Thomas F G Higham2
1. Department of Chemistry, University of Scranton, PA, USA.
2. Research Laboratory for Archaeology, University of Oxford, UK.

Coastal archaeological sites often have shortages of organic remains for radiocarbon dating, but
because shell middens are typically present, accurate dates of shell carbonate can provide a way to
date these sites. Because mollusks are processed soon after death and their remains (shells) are sel-
dom transported long distances, the material offers a close correlation to archaeological events.
However, accurate radiocarbon dates of molluscan shells (marine, aquatic or terrestrial) are notori-
ously difficult to obtain. This difficulty stems from two main issues: quantifying the precise marine
reservoir effect and carbonate remodeling during diagenesis. While accurate correction factors for
the reservoir effect have been published for many regions, no suitable method for effectively sepa-
rating remodeled calcium carbonate has yet been developed. We report a new method to chemically
remove remodeled shell carbonate prior to AMS dating.
Molluscan shells are composed of an organic matrix embedded with calcium carbonate crystals. In
most archaeological shells, the organic layers have disintegrated, leaving only the carbonate for dat-
ing. The exact structure of the calcified shell layers and the crystalline composition varies with the
species; while some species have prismatic layers composed exclusively aragonite, others contain a
dual calcite-aragonite or calcite only structure. Post-depositional alteration of shells typically results
in the conversion of native aragonite to calcite, often with the inclusion of allochthonous carbon into
the remodeled calcitic structures. We have developed a pretreatment procedure to separate aragonite
from calcite using density fractionation of ground carbonate in bromoform. Two sequential separa-
tions result in greater than 99% removal of calcite from reference mixtures as shown by infrared
spectroscopy and X-ray diffraction, with similar results on separations using archaeological shells.
Radiocarbon dates on the pretreated shells are expected to be more accurate than on shells not
treated using this method.

#100

OPTIMIZATION OF THE AUTOMATED GRAPHITIZATION SYSTEM AGE-1

M NÏmec1,2, L Wacker2
1. Dept. of Chemistry and Biochemistry, University of Bern, Switzerland.
2. Ion Beam Physics, ETH Zurich, Switzerland.

Graphitization of samples is a standard procedure for high-precision AMS 14C measurement. The
automated graphitization equipment (AGE) was developed at ETH Zurich to prepare fast and effi-
ciently samples for radiocarbon measurement. Samples are combusted in an elemental analyzer
(EA) and the resulting CO2 is absorbed in a zeolite trap. The CO2 is released by heating the trap and
expands to the individual reactors, where it is reduced by H2 to graphite on the surface of iron cata-
lyst. The use of liquid nitrogen is avoided. We have systematically optimized our system to get fast
graphitization reactions with low processing blanks. Special care was taken to get robust and repro-
ducible preparations.
66 Abstracts

A fast graphitization was obtained with a special pretreatment of the iron catalyst and optimized
reaction temperatures. The improvements were mainly achieved on the basis of thermodynamic data
(Baur-Glaessner diagrams). The catalyst is first oxidized by heating with air and then after evacua-
tion reduced back to iron with hydrogen. The main purpose is to shift reaction equilibrium between
iron oxides and hydrogen strongly to the iron side and remove possible carbon and other impurities.
The graphitization reactions were performed at various reaction temperatures and H2/CO2 ratios.
Resulting graphite samples were measured by AMS MICADAS system to determine the smallest
isotopic changes (d13C) at minimum of molecular fragment formation (13CH current).
A low processing blank was obtained finding tin capsules with low 14C content and minimizing the
cross-contamination from the sample combustion in the EA and from the CO2 adsorption in the zeo-
lite trap. Finally, we will demonstrate that AGE with the EA can be used for a wide range of different
sample types, from dry organic matter over dissolved organic matter to inorganic carbonates.

#101

LASER-HEATED MICROFURNACE: GAS ANALYSIS & GRAPHITE MORPHOLOGY

A M Smith, Bin Yang, Q Hua, Michael Mann
Australian Nuclear Science and Technology Organisation, Lucas Heights 2234 NSW, Australia.

We describe progress in development of a novel miniaturized laser-heated “microfurnace” aimed at

preparing ultra-small graphite samples from CO2 at the ~5 μg of carbon level (Smith et al. 2006,
2007, 2008). Recent effort has focused on automation using a LABview interface, which has permit-
ted feedback control of the catalyst temperature as the reaction proceeds and logging of reaction
parameters. Additionally, an automatic system has been developed to control the temperature of the
cold finger for trapping CO2 (–196 °C), trapping H2O (–80 °C) and releasing these gases (25 °C)
during sample transfer and during the reaction.
We have utilized a quadrupole mass spectrometer to study the gas composition during the reaction,
in order to better understand the underlying chemical reactions for such small samples and to better
estimate the overall efficiency of the process. Early results show that all CO2 is converted to CO by
reduction on the iron catalyst within a few minutes of applying laser power. The reaction pressure
stabilizes after ~20 minutes; however, some CO is not converted to graphite. The cold trap temper-
ature of –80 °C is effective at trapping H2O so there is little CH4 production.
We have trialled a number of different iron catalysts (Cerac -325, Sigma Aldrich -400 and 25 nm Fe
nanopowder) as well as Fe2O3 (reduced in situ to Fe) and have studied the graphite morphology by
scanning electron microscopy (SEM). There is a marked difference in morphology with catalyst
type; however, each graphite performs well in the cesium sputter ion source of the ANTARES AMS
facility. These developments allow us to systematically optimize the performance of the apparatus
and to develop a second generation device.

REFERENCES
AM Smith, Q Hua, A Williams, V Levchenko. Submitted. 11th International Conference on Accelerator Mass Spectrometry,
Rome, Italy, 13–19 September 2008.
AM Smith, Q Hua, VA Levchenko, XVII INQUA Congress 2007, Cairns, 28 July–3 August 2007.
AM Smith, Q Hua, AA Williams, KJ Thorpe. 2006. 19th International Radiocarbon Conference, Oxford, 3–7 April 2006.
 Abstracts 67

Parallel Session 2-9: How Accurately Can We Use Radiocarbon Dating?

#102

METHODOLOGICAL ISSUES OF DATING HISTORICAL EVENTS

Asya Engovatova1, Ganna Zaitseva2, Anatoli Sementsov2, Natalia Bourova2
1. Institute of Archaeology, Russian Academy of Sciences, Moscow, Russia.
2. Institute for the History of Material culture, Russian Academy of Sciences, St. Petersburg, Russia.
14C dating and subsequent calibration of the results allow archaeologists to obtain (with maximum
probability) precise calendar dates. This year, we had the opportunity to check the accuracy of 14C
dating through correlating it with a well-known historical event, the destruction of the ancient Rus-
sian town of Yaroslavl by the Tatar Mongols in AD 1238.
14C was used for several dozen samples of bone, wood, and charcoal from the collective sanitary
burials of AD 1238. Of special value was the possibility to use several different methods for dating
the well-known AD 1238 event. One of the burials was in a well-preserved timber well. Dendro-
chronological dating of the logs revealed that the well was constructed in the 1220s. The typology
of the archaeological artifacts from the backfill of the graves (armlets, crosses, pottery, plinth frag-
ments, etc.) allows to state that the event in question occurred certainly not later than the middle of
the 13th century.
Thus, the traditional archaeological typology of artifacts shows that the historical event, the cata-
strophic destruction of Yaroslavl, occurred between the late 1220s and the 1240s. Chronicles for the
period studied only one event in which the town and its inhabitants had perished, the destruction of
the town by the Tatar Mongols in 1238. An indirect testimony of the fact that the attacking army had
indeed been Tatar Mongols can be found in the especial brutality of the massacre: the human bones
from the collective burials show numerous mortal wounds from cutting and stabbing weapons.
Radiocarbon dating of the samples was done in two laboratories: St. Petersburg and Kiev. With the
exception of several samples, the intervals of the calibrated dates correlate well with the archaeolog-
ical dates for the artifacts and do not contradict the historical date of the event in question. Dendro-
chronological dating of the well also indicate AD 1238 as the date of the said collective burial.

#103

TAKING THE ‘PRE’ OUT OF PREHISTORY: A NARRATIVE FOR THE EARLY CENTURIES OF
THE NEOLITHIC IN BRITAIN AND IRELAND
Alex Bayliss1, Gabriel Cooney2, Frances Healy3, Alasdair Whittle3
1. English Heritage, 1 Waterhouse Square, 138-142 Holborn, London, EC1N 2ST, UK.
2. College of Arts & Celtic Studies, School of Archaeology, Newman Building, Belfield, Dublin 4, Ireland.
3. Cardiff University, Humanities Building, Colum Drive, Cardiff, CF10 3EU, UK.

Site-based Bayesian chronological models have been constructed on a routine basis in England for
well over a decade (Bayliss and Bronk Ramsey 2004). It is only in the past few years, however, that
the impact of the more precise dating which has ensued has become fully apparent, as a critical mass
of well-dated sites has begun to be established (Bayliss and Whittle 2007; Bayliss 2009).
68 Abstracts

We present the results of a project to date nearly 40 causewayed enclosures, and integrate the result-
ant generational narrative into wider histories of the early Neolithic of Britain and Ireland (Whittle
et al., forthcoming). Our analysis incorporates detailed interpretations of the stratigraphic positions,
sample taphonomy, and material associations of more than 2000 radiocarbon dates. Our dating of
the enclosures themselves has revealed the tempo of the phenomenon – the pace of its introduction,
cycles in construction, and the rhythms of its demise. We use the methodology of Blackwell and
Buck (2003) to reveal the place of enclosures within the range of things and practices that make up
the early Neolithic in Britain and Ireland. This structures the social context in which the enclosures
emerged as a developed aspect of the insular Neolithic. Our analysis not only creates an explicit,
quantified chronology of the first centuries of the Neolithic on these islands, but also provides a
sense of the tempo of the times.
George Box (1979:202) reminds us that “all models are wrong, some models are useful.” Our mod-
els, and the data on which they are based, are imperfect and of uneven quality. Nonetheless, we feel
that the scale of difference between the refined, quantitative chronologies that we present and the
impressionistic, fuzzy nature of previous frameworks cannot be exaggerated. Nor can the difference
in the type of (pre)history that we can practice as a result. We can trace sequences, where previously
the unfoldings of past lives were compressed into one, we can compare contemporary phenomena
to reveal the diversity of choices made by people in the past, and we can begin to unpick the varying
tempos and scales of past change, where previously we were forced to dwell largely in the long term.
It is clear that our chronologies will develop over the coming decades, but this paper gives a taste of
the new type of (pre)history that is to come.

REFERENCES
Bayliss A. 2009. Rolling out revolution: using radiocarbon dating in archaeology. Radiocarbon 50(1):123–47.
Bayliss A, Bronk Ramsey C. 2004. Pragmatic Bayesians: a decade integrating radiocarbon dates into chronological models.
In: Tools for constructing chronologies: tools for crossing disciplinary boundaries (eds C E Buck and A R Millard). Lon-
don: Springer. p 25–41.
Bayliss A, Whittle A (eds). 2007. Histories of the dead: building chronologies for five southern British long barrows. Cam-
bridge Archaeological Journal 17(suppl).
Blackwell P, Buck CE. 2003. The Late Glacial human reoccupation of north-western Europe: new approaches to space-time
modelling, Antiquity 77:232–9.
Box GEP. 1979 Robustness in scientific model building. In: Robustness in statistics (eds R L Launder and G N Wilkinson).
New York: Academic Press. p 201–36.
Whittle A, Healy F, Bayliss A. Forthcoming. Gathering time: dating the early Neolithic enclosures of southern Britain and
Ireland. Oxford: Oxbow.

#104

RADIOCARBON DATING FOR SUB-CENTURY CHRONOLOGICAL DIFFERENCES: THE ANA-

LYTICAL AND ARCHAEOLOGICAL INSIGHTS
E Boaretto1, I Sharon2, A Gilboa3, E M Scott4, A J T Jull5, J van der Plicht6, J Heinemeier7
1. Department of Land of Israel Studies and Archaeology, Bar Ilan University; and Radiocarbon and Cosmogenic Isotopes
Laboratory, Weizmann Institute of Science, Israel.
2. Institute of Archaeology, Hebrew University, Israel.
3. Zinman Institute for Archaeology, Haifa University, Israel.
4. Department of Statistics, University of Glasgow, UK.
5. NSF Arizona AMS laboratory, Arizona University, Tucson, Arizona, USA.
 Abstracts 69

6. Center for Isotope Research, Groningen University, Groningen, the Netherlands.

7. AMS 14C Dating Centre, Department of Physics and Astronomy, Aarhus University, Aarhus, Denmark.

The chronology of the Iron Age in Israel has been one of the high-priority research issues in biblical
archaeology. Different approaches have been adopted for understanding sub-century chronological
issues. We have addressed this problem as a large-scale intercomparison between different groups.
The integration of these methodologies and mutual feedback in order to improve and determine the
precision and accuracy of the results has been fundamental. Hundreds of samples for radiocarbon
have been measured, and an analytical protocol has been developed to detect bias, with strict control
on the archaeological context, and most important, full collaboration of the archaeologists for deter-
mining the good contexts. This has resulted in comparable radiocarbon results within errors.
Although differences remain in the interpretation, the radiocarbon results obtained represent an
important step forward to understand how to approach a chronological problem in involving a large
geographical area and a long span of time.

Parallel Session 3-9: 14C Methods and Chemistry II

#105

DEVELOPMENT OF AN AUTOMATED REDUCTION SYSTEM FOR AMS SAMPLE PREPARA-

TION
Wan Hong, Jung Hun Park, Hyung Joo Woo, Jun Kon Kim, Han Woo Choi, Gi Dong Kim
Korea Institute of Geoscience and Mineral Resources (KIGAM), 92 Gwahakno, Yuseong-gu, Daejeon 305-350, Korea.

KIGAM introduced an AMS system at the end of 2007. The main purpose of this machine is geo-
logical and environmental research using radiocarbon dating and Be measurement, and sample
preparation facilities have been installed for this purpose.
A reduction system to convert carbon dioxide gas to graphite has been developed. Twenty four lines
are provided on the system to treat 24 samples at once. An elemental analyzer (EA) with an auto-
matic sampler is employed for combustion of organic matter. This EA is directly connected to the
reduction system so that CO2 gas can be sent to the reduction chamber automatically. Both the
reduction system and EA are controlled by a personal computer. All the control and signal lines have
been interfaced using data acquisition modules and remote control modules of National Instruments.
The reduction software also has been developed using DAQ and MS Visual Studio.
Operation of valves, furnaces, water traps is controlled by the PC. The temperature and pressure
during the reduction process are monitored and stored in the PC. This is a very efficient method to
maintain the same reaction conditions of 24 reduction chambers. The detailed characteristics of the
reduction system will be presented at the conference.
70 Abstracts

#106

SIMPLE PRETREATMENT METHOD OF IRON AND CARBONATE SAMPLES

Junghun Park, Wan Hong, H J Woo, Han Woo Choi, Jun Kon Kim, Gi Dong Kim
Korea Institute of Geoscience and Mineral Resources, 92 Gwahakno, Yuseong-gu, Daejeon 305-350, Korea.

An AMS system was installed at Korea Institute of Geoscience and Mineral Resources (KIGAM)
and chemical treatment methods of various kinds of samples have been established for radiocarbon
dating, e.g. wood and other organic samples, soil and sediment, bone, pottery, and air samples. In the
case of iron and carbonate samples, simple chemical treatment methods were tested.
In the case of iron samples, after physical cleaning without chemical treatment, a simple pretreat-
ment method using an elemental analyzer (EA) was attempted. In the case of carbonate samples,
after cleaning and drying carbonate samples at 150–200 °C overnight without special chemical
treatment, a simple pretreatment method using an EA was also attempted. Overnight drying of car-
bonate samples removed modern CO2 contamination, and the background level decreased to that of
a sample pretreated with phosphoric acid.

#107

INITIAL MEASUREMENTS OF RADIOCARBON IN ATMOSPHERIC FORMALDEHYDE AT NAR-

RAGANSETT, RHODE ISLAND, USA
Haiwei Shen1, Ann P McNichol2, Li Xu2, John Merrill1, Brian G Heikes1
1. Center for Atmospheric Chemistry Studies, Graduate School of Oceanography, University of Rhode Island, Narragansett,
Rhode Island, USA.
2. National Ocean Sciences Accelerator Mass Spectrometry Facility, Department of Geology and Geophysics, Woods Hole
Oceanographic Institution, Woods Hole, Massachusetts, USA.

Formaldehyde is a key intermediate in tropospheric photochemistry. Its atmospheric sources involve

a complex mixture of biogenic and anthropogenic volatile organic compounds (VOCs) and its pho-
tolysis products impact odd-hydrogen radicals and ozone (O3) chemistry. The relative contribution
of fossil VOCs and biogenic VOCs to formaldehyde, and subsequently to ozone and oxidant chem-
istry is expected to vary seasonally and spatially due to VOC speciation, emission patterns, and reac-
tivity. Implementation of regional strategies to control O3 in the northeastern United States has
focused on non-methane anthropogenic VOCs and the efficacy of these strategies depends upon the
relative mix of reactive biogenic VOC emissions and anthropogenic emissions. The radiocarbon,
14C, content of formaldehyde is useful in assessing the relative contributions of fossil and biogenic

VOCs to formaldehyde, hence to O3. We report a compound specific radiocarbon analysis (CSRA)
method for formaldehyde based upon gaseous formaldehyde collection, derivatization to thiazoli-
dine, preparative capillary gas chromatography separation, and AMS analysis. Ambient measure-
ments from Narragansett, RI, made in winter and summer 2007 are presented. On 11 of 13 samples,
we find 80 to >95% of the collected formaldehyde is of fossil origin and, contrary to our initial
hypothesis, we see no seasonal shift in proportion. The remaining 2 samples, one each from winter
and summer, are 30–40% modern carbon. The measurements are interpreted considering fossil-bio-
genic source attribution and local transport conditions and contrasted with prior measurements from
Nova Scotia. Further, CSRA measurement of acetaldehyde is feasible with the method.
 POSTER SESSION

AMS

P-1

IMPROVED PERFORMANCE OF THE CAMS/LLNL HIGH-INTENSITY Cs-SPUTTER NEGATIVE

ION SOURCE: ION TRAJECTORY MODELING AND MODIFIED ELECTRODE GEOMETRIES
Thomas A Brown and Thomas P Guilderson
Center for Accelerator Mass Spectrometry, LLNL, 7000 East Avenue, Livermore, CA 94550, USA.

The ionizer/sample region of the CAMS/LLNL high-intensity Cs-sputter negative source has been
modeled, using the gird-mesh based modeling program NEDLab™, with the aim of improving neg-
ative ion output while not decreasing ionization efficiency or deleteriously affecting ion source sta-
bility. The program has several features that are useful in modeling the interior of such ion sources
including: 1) space charges effects on electric field distributions and ion trajectories can be modeled
through a convergent iterative process involving successive calculations of Poisson equation deter-
mined electric fields and spatial distributions of charge resulting from the trajectories of emitted ions
moving in the electric fields; 2) the program allows ion emission from surfaces to be space charge-
limited following Child’s Law; 3) the program allows sets of ions to be emitted with specified ther-
mal energy distributions and randomized emission angles.
Based on the results of the modeling effort, modified versions of two components of the interior
geometry of the ion source were fabricated. Initial operation of the source with these components
indicates that the modified geometry allows an increase in 12C ion production of ~50%. Further
experiments to determine whether the modified geometry deleteriously affects negative ion produc-
tion efficiency and to determine the emmittance of the ion beam under different source parameters
are planned.
This work was performed under the auspices of the US Department of Energy by Lawrence Liver-
more National Laboratory under Contract DE-AC52-07NA27344.

P-2

MEASURING SUB-MICRON-SIZE FRACTIONATED PARTICULATE MATTER ON ALUMINUM

IMPACTOR FILTERS
Bruce A Buchholz1, Paula Zermeño1, Hyun-Min Hwang2, Thomas M Young2,
Thomas P Guilderson1
1. Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, P.O Box 808, Livermore, CA
94551, USA.
2. Dept. of Civil and Environmental Engineering, University of California, One Shields Avenue, Davis, CA 95616, USA.

Sub-micron-sized airborne particulate matter is not collected well on regular quartz or glass fiber fil-
ter papers due to disturbed flow rates when impactor samplers are used. We used a micro-orifice uni-
form deposit impactor (MOUDI) to size fractionate particulate matter (PM) into six size fractions

71
72 Abstracts

and deposit it on specially designed high-purity thin aluminum disks. The MOUDI separated PM
into fractions 56–100 nm, 100–180 nm, 180–320 nm, 320–560 nm, 560–1000 nm, and 1000–1800
nm. Since MOUDI have low flow rates, it takes several days to collect sufficient carbon on 47-mm
foil disks. The small carbon mass (20–200 μg C) and large aluminum substrate (~25 mg Al) presents
several challenges to production of graphite targets. The Al foil consumes large amounts of oxygen
as it is heated and tends to melt into quartz combustion tubes causing gas leaks. We describe sample
processing techniques to reliably produce graphitic targets for 14C AMS analysis of PM deposited
on Al impact foils.
This work performed in part under the auspices of the US Department of Energy by Lawrence Liv-
ermore National Laboratory under Contract DE-AC52-07NA27344.

P-3

14C
AMS MEASUREMENTS AT SUERC: IMPROVING QA DATA FROM THE 5MV TANDEM AMS
AND 250KV SSAMS
P Naysmith1, G T Cook1, S P H T Freeman1, E M Scott2, R Anderson1, S Xu1, E Dunbar1,
G K P Muir1, A Dougans1, K Wilcken1, C Schnabel1, N Russell1, P L Ascough1, C Maden1
1. SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, Scotland, UK.
2. Statistics Department, University of Glasgow, Glasgow G12 8QQ, Scotland, UK.

Up until 2003, SUERC had procured its AMS measurements at the University of Arizona. In 2003,
a National Electrostatics Corporation 5MV tandem accelerator mass spectrometer was installed at
SUERC, providing the radiocarbon laboratory with 14C measurements to 4–5‰ repeatability. In
2007, a 250kV single-stage accelerator mass spectrometer (SSAMS) was added to provide addi-
tional 14C capability as the 5MV instrument was increasingly being used for measuring other cos-
mogenic isotopes. However, the SSAMS is now the preferred system for 14C analysis. Changes to
the technology and to our operations are evident in our copious quality assurance data: typically, we
now use the 134-position MC-SNICS source, which is filled to capacity. The source is notionally
divided into 13 batches of 10 samples, each containing 7 unknowns, 1 Belfast Cellulose standard (as
used in FIRI as Sample I), and either a Barley Mash standard (as used in TIRI as Sample A) or a
background standard (either an interglacial wood sample or a geological age carbonate, depending
on the form of the unknowns). Measurement of standards shows that spectrometer running without
the complication of on-line δ13C evaluation is a good operational compromise. Currently, 3‰ 14C/
13C measurements are routinely achieved for samples up to nearly three half-lives old by consistent

sample preparation and an automated data acquisition algorithm with sample random access for
measurement repeats. Although the measurements are made to around 3‰, the routine quoted pre-
cision for samples <2 half-lives old is currently ± 30 to 35 years at 1 σ, the error being limited by
the standard deviation on the standards within the batch. Background and known age standard data
are presented for the period 2003–2008 for the 5MV system and 2007–2008 for the SSAMS to dem-
onstrate the improvements in performance of these two instruments over time, with a brief overview
of the modifications that have been made to each.
 Abstracts 73

P-4

THE KECK CARBON CYCLE AMS LABORATORY, UNIVERSITY OF CALIFORNIA IRVINE:

STATUS REPORT
Robert K Beverly, Denis Tauz, Will Beaumont, Shenil Patel, Kaelyn M Ormsby,
Guaciara M Santos, John R Southon
Earth System Science Dept., University of California, Irvine CA 92612, USA.

The Keck Carbon Cycle AMS facility was established in 2001–2002 and was set up to use carbon
isotopic techniques, primarily AMS, to advance understanding of the carbon cycle and its linkages
with climate. The AMS spectrometer is a National Electrostatics 0.5MV 1.5SDH-1 AMS system
with a single 40-sample MC-SNICS ion source. After an initial series of technical upgrades (Sou-
thon et al. 2004, 2007; Southon and Santos 2004, 2007), 0.2–0.3% precision is routinely achieved
for samples with >0.7 mg of carbon (Santos et al. 2007b). The spectrometer has been relatively trou-
ble-free, for a system that runs 24/7 and is reaching 60,000 targets measured, but, after 7 years of
operation, it is starting to show its age and consequently some significant problems have occurred.
We summarize recent upgrades and some of the lessons learned during the last 3 years of spectrom-
eter operation.
Regarding sample preparation, the KCCAMS prep laboratory (that provides 60% of all samples
measured at the facility) has been upgraded to handle a larger number of samples. This lab routinely
produces graphite samples down to 0.015 mg C for outside submitters and internal research as small
as 0.001 mg C, with close to 100% yield (Santos et al. 2007a,b). The main recent modifications
include the construction of a third 12-headed graphitization line, digital monitoring of reaction ves-
sel pressures during graphitization and catalyst preconditioning, additional pretreatment of the Fe
catalyst, and transition from pressing samples with a hammer to a modified pellet press. In addition,
storage issues with combusted and graphitized samples, as well as carbonate sample processing, are
discussed.

REFERENCES
Santos et al. 2007a. Ultra small-mass 14C-AMS sample preparation and analysis at the KCCAMS Facility. Nuclear Instru-
ments and Methods in Physics Research B 259:293–302.
Santos et al. 2007b. AMS 14C sample preparation at the KCCAMS/UCI Facility: status report and performance of small sam-
ples. Radiocarbon 49(2):255–269.
Southon and Santos 2004. Ion source development at KCCAMS, University of California, Irvine. Radiocarbon 46(1):33–39.
Southon and Santos 2007. Life with MC-SNICS Part II: recent ion source development at the Keck Carbon Cycle AMS Fa-
cility. Nuclear Instruments and Methods in Physics Research B 259:88–93.
Southon et al. 2004. The Keck Carbon Cycle AMS laboratory, University of California, Irvine: initial operation and a back-
ground surprise. Radiocarbon 46(1):41–49.
Southon et al. 2007. Cs feed tests and emittance measurements on a modified MC-SNICS ion source for radiocarbon AMS.
Radiocarbon 49(2):301–305.
74 Abstracts

P-5

ULTRA-SMALL (<30 μg C) AMS 14C MEASUREMENTS AT THE CAMS FACILITY

Amy C Englebrecht1 and Thomas P Guilderson2
1. Department of Earth and Planetary Science, University of California, Berkeley, CA 94720, USA.
2. Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA.

Measurements of the natural abundance of radiocarbon are widely used, but the development of pro-
tocols for routine measurements of ultra small carbon samples is critical for many applications.
Here, we have evaluated the precision and system efficiency for analyses of samples <30 μg C at
CAMS. We made several modifications to our standard graphite preparation and AMS procedures,
including optimizing the catalyst:C ratio, decreasing the cathode voltage, and varying the position of
the cathode to account for a change in the cesium focal point at lower cathode voltages. Together, the
modifications resulted in 5‰ counting statistics on individual samples down to 20 μg C. The stan-
dard deviation of repeat analyses was better than 1%. Total system efficiency for samples of these
sizes was 17% (sputtering in the ion source of ~34%), and is comparable to full-sized samples using
much higher cathode voltages. Further, protocol modifications did not negatively impact sample
throughput time.

P-6

XCAMS: EXTENDING THE POTENTIAL OF LOW VOLTAGE, LOW DETECTION LIMIT 14C AMS
TOWARDS 10Be AND 26Al WITH A PLANNED FACILITY AT NEW ZEALAND’S NATIONAL ISO-
TOPE CENTRE
A Zondervan1, F Bruhn1, Gregory A Norton2, Richard L Kitchen2, Thilo M Hauser2
National Isotope Centre, GNS Science, Lower Hutt, New Zealand.
National Electrostatics Corporation, Middleton, Wisconsin, USA.

We present the layout and beam optics of a new low-voltage AMS system for 14C, 10Be, and 26Al,
scheduled for commissioning in early 2010 at the NIC (National Isotope Centre, GNS Science, New
Zealand). The new system is part of an ongoing program to increase the science and commercial
capabilities of the NIC. It is based on NEC’s compact carbon AMS (CAMS, 0.5MV Pelletron) sys-
tem and on the findings by Mueller et al. [NIM B266 (2008) 2207–12] on how to reject scattered 9Be
from mass 26 injection. A post-analysis Si-nitride window is added to create an energy separation
between 10Be and 10B. Rejection is expected to be complete with a 45-degree magnet followed by
double slits in the rare-isotope beamline. For optimum rare-isotope transmission, two electrostatic
quadrupole triplets are added on the high-energy side of the spectrometer.
 Abstracts 75

P-7

FROM DATING HISTORY TO DECODING PALEOCLIMATE: THE CONTRIBUTION OF AMS AND

CARBON STABLE ANALYSIS
L Calcagnile1, G Fiorentino2, G Quarta1, V Caracuta2, M D’Elia1
1. CEDAD (Centre for Dating and Diagnostics), Dept. of Engineering of Innovation, University of Salento, Lecce, Italy.
2. Laboratory of Archeobotany and Palaeoecology, University of Salento, Lecce, Italy.

Analytical methods based on the determination of carbon isotopic ratios have become a fundamental
tool in several research fields including archaeological and climatological sciences. In fact, besides
the well-established AMS radiocarbon dating method for obtaining chronological information, the
measurement of carbon stable isotope ratios of plant remains can supply important information
about past climatic conditions. We report on the potentialities of a combined approach in which
plant remains, selected from natural as well as anthropogenic deposits, are submitted for carbon sta-
ble isotope and 14C AMS analyses in order to reconstruct past climatic fluctuations. Methodological
issues associated with the measurement of δ13C term on-line by AMS will be discussed and the
results of a comparative IRMS-AMS exercise presented.
The results obtained in the analysis of plant remains recovered from five different archaeological
sites will be presented: Grotta della Madonna in Praia a Mare (SW Italy), Ebla and Qatna (NW
Syria), Tas šilg (Malta), and Faragola (SE Italy). The studied sites, all located within the Mediterra-
nean basin, correspond to different periods during the last 8000 years and give the possibility to
define climate fluctuations occurred during the Holocene.

P-8

AMS RADIOCARBON DATING FOR THE IROSIN IGNIMBRITE AND CO-IGNIMBRITE ASH-
FALL, SOUTHERN LUZON, PHLIPPINES
M H T Mirabueno1, M Okuno2, T Nakamura3, T Danhara4, E P Laguerta5, C G Newhall6,
T Kobayashi7
1. Philippine Institute of Volcanology and Seismology (PHIVOLCS), C.P. Garcia Avenue, University of the Philippines, Dil-
iman, Quezon City, Philippines.
2. Dept. of Earth System Science, Faculty of Science, Fukuoka University, Fukuoka 814-0180, Japan.
3. Center for Chronological Research, Nagoya University, Nagoya 464-8602, Japan.
4. Kyoto Fission-Track Co. LTD, Kyoto 603-8832, Japan.
5. Mayon Volcano-Ligñon Hill Observatory, Philippine Institute of Volcanology and Seismology (PHIVOLCS), Ligñon Hill,
Legazpi City, Philippines.
6. Formerly U.S. Geological Survey, USA.
7. Dept. of Earth and Environmental Science, Faculty of Science, Kagoshima University, Kagoshima 890-0065, Japan.

The eruption of dacitic to rhyolitic pyroclastic flows, Irosin ignimbrite, resulted in the formation of
the Irosin caldera in Bicol Peninsula, Philippines. The caldera, together with the active Bulusan Vol-
cano and associated older volcanic centers in various stages of erosion, comprise the Bulusan Vol-
canic Complex (BVC). This paper presents not only the AMS 14C dates for the Irosin ignimbrite dis-
tributed in the province of Sorsogon, but also describes for the first time the co-ignimbrite ash-fall
at Inascan Scoria Cone on the western slope of Mayon Volcano. The calendar year of 41 cal kyr BP
obtained for the eruption contributes to the study of the Irosin caldera and to the database of wide-
spread tephra deposits in the Philippines.
76 Abstracts

P-9

A HIGH-PERFORMANCE 14C ACCELERATOR MASS SPECTROMETRY SYSTEM

M L Roberts1, J R Burton1, K L Elder1, B E Longworth1, C P McIntyre1, K F von Reden1,
B X Han2, B E Rosenheim3, W J Jenkins1, E Galutschek1, A P McNichol1
1. Woods Hole Oceanographic Institution, Department of Geology and Geophysics, Woods Hole, MA 02543, USA.
2. Present address: Research Accelerator Division, SNS, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA.
3. Present address: Department of Earth and Environmental Sciences, Tulane University, New Orleans, LA 70118, USA.

A new and unique 14C accelerator mass spectrometry facility has been constructed at the Woods
Hole Oceanographic Institution. The defining characteristic of the new system is its large-gap opti-
cal elements that provide a larger-than-standard beam acceptance. Such a system is ideally suited for
high-throughput, high-precision measurements of 14C. Details and performance of the new system
are presented.

P-10

DEVELOPMENT OF A GAS CHROMATOGRAPH-COMBUSTION SYSTEM FOR 14C AMS

Cameron P McIntyre1, Sean P Sylva2, Mark L Roberts1
1. National Ocean Sciences Accelerator Mass Spectrometry Facility, Dept. of Geology and Geophysics, Woods Hole Ocean-
ographic Institution, Woods Hole, Massachusetts 02543, USA.
2. Dept. of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543,
USA.

A novel gas chromatograph-combustion interface has been constructed for the microwave plasma
gas ion source 14C AMS system under development at the Woods Hole Oceanographic Institution.
It is capable of separating and combusting up to 10 μg of compound and transferring the resulting
CO2 into less than 1 mL/min of argon gas, which is the preferred gas for operation. The system is
fully on-line and does not require cryogenic trapping of the products prior to analysis. It represents
a significant step forward technically and has the potential to be compatible with a wide range of
AMS systems.
We are continuing to characterize and refine the system with respect to chromatographic perfor-
mance and isotopic fidelity. Peak resolution is maximized by selection of column properties, sample
size, and gas flow rates. Extra column effects on peak broadening are being minimized by optimiza-
tion of plumbing geometry. Efficiency of sample delivery to the source is controlled by an open
split. Isotopic fidelity is being investigated by offline trapping of the CO2 and analysis via conven-
tional graphite/AMS techniques. We outline the design of the system and show results from analysis
of standard compounds and real world samples. The system is shown to perform satisfactorily and
meets the requirements of the gas ion source for natural abundance 14C GC-AMS measurements.
 Abstracts 77

P-11

A NEW GAS-ACCEPTING NEGATIVE ION SOURCE FOR 14C DETECTION

E Galutschek1, J Wills2, B Han3, C McIntyre1, K von Reden1, W Jenkins1, R Schneider1,
M L Roberts1
1. Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA.
2. Chalk River Laboratories, Chalk River, Ontario K0J 1J0, Canada.
3. Oak Ridge National Laboratory, Oak Ridge TN 37831, USA.

A new gas-accepting negative ion source for 14C detection designed in collaboration with Atomic
Energy of Canada, Chalk River Laboratories, has been built for the continuous-flow accelerator mass
spectrometry (CFAMS) system at the National Ocean Sciences Accelerator Mass Spectrometry Facil-
ity (NOSAMS). A continuous-flow of CO2 gas is injected into a conventional microwave ion source
operating at 2.45 GHz. The total extracted positive ion current (typically several mA) then passes
through a magnesium vapor cell where negative ions are formed in charge-exchange collisions.
A prototype of this ion source typically accepted a gas flow of 250–300 μl CO2 per second and
showed negligible memory effect. Operationally, this new source is expected to outperform the pro-
totype previously tested at NOSAMS. Presently, the new ion source is interfaced to the new CFAMS
system. Characteristics, first results, as well as planned future modifications will be presented.

ARCHAEOLOGY
P-12

ANOMALOUS 14C AGES OBTAINED ON 612 BCE HUMAN SKELETONS FROM ANCIENT NIN-
EVEH: ALTERNATIVE EXPLANATIONS
R E Taylor1 and John Southon2
1. Department of Anthropology, University of California, Riverside; Cotsen Institute of Archaeology, University of Califor-
nia, Los Angeles; Keck Carbon Cycle Accelerator Mass Spectrometry Laboratory, University of California, Irvine, USA.
2. Keck Carbon Cycle Accelerator Mass Spectrometry Laboratory, University of California, Irvine, USA.

Radiocarbon determinations have been obtained on a suite of nine human bone samples recovered
from a major destruction layer historically dated to 612 BCE at the ancient Assyrian city of Nineveh
in northern Mesopotamia (modern Iraq). The samples were obtained during excavations carried out
under the direction of Professor David Stonach (UC Berkeley) at the Halzi Gate area at Nineveh. All
of the bones measured derived from human skeletons all killed at the same time during a defense of
an assault on the city gate. The calibration of these 14C age determinations places the equivalent cal-
endar/solar ages of these bones in the early 8th century BCE. The calibrated 14C values appear to
yield ages almost 2 centuries too old. The archaeological and historical data documenting 612 BCE
as the year of a single destruction event appears to be very strong and thus we assume a known-age
event to ±1 solar years. We consider data and arguments associated with several alternative expla-
nations for the anomalous 14C ages including (i) failure of the pretreatment method employed to
remove sufficient exogenous organics from the bone samples, (ii) a bone dietary reservoir age
effect, resulting from fish in the diet of the contemporary population, and (iii) a regional atmo-
spheric Δ14C offset. One purpose of presenting this data set is to solicit comparisons with 14C values
obtained by other investigators from other known-age historically informed archaeological sites and
contexts for this time period in this region.
78 Abstracts

P-13

ARCHAEOLOGICAL RADIOCARBON DATES FOR STUDYING THE POPULATION HISTORY IN

EASTERN FENNOSCANDIA
M Oinonen1, P Pesonen2,3, M Tallavaara3
1. Finnish Museum of Natural History, Dating Laboratory, University of Helsinki, Finland.
2. National Board of Antiquities, Department of Archaeology, Finland.
3. Institute for Cultural Research, Department of Archaeology, University of Helsinki, Finland.

There is increasing interest worldwide to investigate prehistoric population history based on distri-
butions of radiocarbon dates (e.g. Rick 1987; Gamble et al. 2005; Fidel and Kuzmin 2007; Shennan
and Edinborough 2007; Smith et al. 2008). Following a similar approach, we have been working on
this question in the area of eastern Fennoscandia (Tallavaara et al. 2009). Keeping in mind the critics
presented (Surovell and Brantingham 2007), we will scrutinize the radiocarbon data gathered from
eastern Fennoscandian contexts, especially its suitability for studies of population history.
The collected database consists of >2500 individual radiocarbon dates forming a timeline for the
period from the earliest colonization ca. 9000 BC to the modern era. A large fraction of data is due
to charcoal finds. Other sample material contains mostly wood, charred crust, bone, burned bone,
and birch tar. The bone and wood samples, particularly, suffer from deterioration effects along with
the studied long time span and, therefore, deserve special attention. In addition, the possible role of
research priorities, reservoir and own age effects are considered. We will elaborate the effects of
these different materials on the shape of the summed probability distribution of dates that is used as
a basis for archaeological interpretations and also for spatiotemporal studies of the population his-
tory (Onkamo et al. 2008).

REFERENCES
Fiedel SJ, Kuzmin YV. 2007. Radiocarbon 49(2):741–56.
Gamble C et al. 2005. Cambridge Archaeological Journal 15(2):193–223.
Onkamo P et al. 2008. http://www.helsinki.fi/bioscience/argeopop/index.htm.
Rick JW. 1987. American Antiquity 52(1):55–73.
Smith MA et al. 2008. The Holocene 18(3):389–401.
Shennan S, Edinborough K. 2007. Journal of Archaeological Science 34:1339–45.
Surovell TA, Brantingham PJ. 2007. Journal of Archaeological Science 34:1868–77.
Tallavaara M, Pesonen P, Oinonen M. 2009. In preparation.

P-14

ORIGIN OF MUMMIES OF PACHACAMAC, PERU: USING RADIOCARBON DATING AND

DIETARY ANALYSIS
Mai Takigami1, Izumi Shimada2, Rafeal Segura2, Minoru Yoneda1
1. Graduate School of Frontier Sciences, University of Tokyo, Japan.
2. Department of Anthropology, Southern Illinois University, Carbondale, IL 62901, USA.

Pachacamac is an Andean city located 40 km southeast from Lima, Peru. It is believed that Pachaca-
mac functioned as a large religious center in the middle region of Andean civilization before the
Spanish invasion in AD 1532. Andean civilization was characterized by population movement.
Pachacamac was one of the important destinations for religious pilgrimage. Fifty-two mummies
 Abstracts 79

were excavated at the Pachacamac site in 2005. One mummy wore a red wood mask, but others did
not have any outstanding characteristics.
We dated the mummies using 14C and carbon and nitrogen isotope analyses in order to ascertain
their origin. The mummies spanned a large range of 14C dating from AD 900 to 1500, which indi-
cates that the tomb was used for a prolonged period. Furthermore, a large variability of dietary habits
in the mummies was observed by carbon and nitrogen isotope analyses of hair, showing that all the
mummies had seasonal variation and a high ratio of maize consumption. However, the marine ratios
of food were distinct, meaning that the mummies came from various regions and ages.

P-15

UPLAND OLIVE DOMESTICATION IN THE CHALCOLITHIC PERIOD: NEW 14C DETERMINA-

TIONS FROM EL KHAWARIJ (AJLUN), JORDAN
Jaimie Lovell1, John Meadows, Geraldine Jacobsen
1. Council for British Research in the Levant Kenyon Institute (CBRL) PO Box 19283 Jerusalem, East Jerusalem 91192,
Israel.

This poster reports on 10 new accelerator mass spectrometry (AMS) dates from the Chalcolithic
period occupation (late 5th millennium cal BC) at el Khawarij, part of the University of Sydney’s
Wadi Rayyan Archaeological Project (WRAP). The new AMS dates from el Khawarij were delib-
erately obtained from carbonized olive pits in order to date the exploitation of olive at el Khawarij,
a late prehistoric settlement believed to have been reliant on olive production. The results reveal a
much longer span of occupation than hitherto suspected, including two dates that may fall later than
3900 cal BC (particularly OZI 221, 3950–3530 cal BC). These later dates are in line with dates from
other upland sites in the region, strengthening suggestions that Chalcolithic settlement persisted for
longer in better watered upland areas (Lovell 2002). Further, an early date from a sample in a rock-
cut installation in Area A suggests a much earlier date for occupation at the site, implying that
upland olive cultivation may have commenced before 4700–4450 cal BC.

REFERENCE
Lovell JL. 2002. Shifting subsistence patterns: some ideas about the end of the Chalcolithic in the southern Levant. Paléorient
28/1:89–102.

P-16

RADIOCARBON DATING THE “VILLA OF AUGUST” AT SOMMA VESUVIANA, ITALY

Minoru Yoneda1, Emilia Allevato2, Hiroyoshi Matsuzaki1, Cohei Sugiyama1,
Satoshi Matsuyama1, Katsuhiro Iwaki1, Gaetano Di Pasquale2, Carmine Lubritto2,
Filippo Terrasi2, Toshitsugu Fujii1, Masanori Aoyagi3
1. University of Tokyo, Chiba/Tokyo, Japan.
2. Università Federico II, Napoli, Italy.
3. National Museum of Western Art, Tokyo, Japan.

Age determination of the “Villa of August” ruin at Somma Vesviana, Italy, was conducted by AMS
radiocarbon dating on charcoal and mortar, to determine the age of its construction. Our previous
80 Abstracts

study showed that the eruption of Mt. Vesuvius at AD 472 buried this site, although it is well known
that the eruption at AD 79 damaged the southern foot of the mountain including Pompeii. This villa
has been named the “Villa of August” based on paleography and its location. However, no direct
information on the founder and the age of this villa has been found in the series of excavations by
Japanese and Italian archaeologists. One problem is the lack of suitable sample for dating the initial
human occupation at the site, which was intensively used by wine makers after the villa fell into
ruin. Few archaeological materials for cultural chronology (pottery, epigraph, etc.) and few organic
matter samples originating from construction events are available. Furthermore, no wooden material
was found as a part of the original architecture. Hence, we collected a series of charcoals buried in
a plastered wall. Although the origin of these charcoal samples is not essentially related to the con-
struction, the accordance of their dates will show the shared date of these walls, such as their reno-
vation or the villa’s construction. The plaster neighboring the charcoal samples was analyzed to see
the relation of the plaster to the dated sample. The plaster consists of calcium hydrate and atmo-
spheric carbon dioxide. If it did not contain carbon from limestone and showed little diagenetic
effect and isotopic exchange, radiocarbon dating may suggest the age of plaster. 14C determinations
of charcoal and plaster were compared with the architectural structure in order to determine the
period of its construction or renovation, which must have been conducted in the active period of this
villa.

P-17

VÖRS-MÁRIAASSZONYSZIGET - A MULTIPERIOD ARCHAEOLOGICAL SITE IN SW HUNGARY

Éva Svingor1, Katalin T Bíró2, Zsófia Medzihradszky3, Mihály Molnár1, Zsuzsanna M Virág4,
Nándor Kalicz5, István Futó1
1. Institute of Nuclear Research of the HAS (ATOMKI), Debrecen, Hungary.
2. Hungarian National Museum, Budapest, Hungary.
3. Hungarian Natural History Museum, Dept. of Botany, Budapest, Hungary.
4. Budapest History Museum, Budapest, Hungary.
5. Archaeological Institute of the HAS, Budapest, Hungary.

Vörs-Máriaasszonysziget (Transdanubia, Hungary) is a multiperiodic archaeological site with

archaeological finds ranging from the Early Neolithic to Medieval periods. The site is located on a
low sandy peninsula protruding an ancient embayment of Lake Balaton. The site has special impor-
tance for several points of view: (1) Vörs is known as one of the northernmost extensions of the
StarËevo culture of southeastern origin, and at the site human remains, representing the oldest dated
inhabitants of Transdanubia, were found in two graves. (2) The site represents a very special envi-
ronment supplying new data on the beginning of agriculture in the Carpathian Basin. (3) The cli-
matic changes and varying sediment discharge to the lake changed the environment of the site,
which is traceable in the archaeological finds.
The early Neolithic period of the settlement was radiocarbon dated to the middle of the 6th millen-
nium BC. Sporadic settlement traces in various periods of the Copper Age (4th millennium BC) give
evidence of human presence between 4500 and 2800 BC. The second most significant period of
habitation is represented by the Early Bronze Age (around 2000 BC), when the settlement area was
surrounded with a circular ditch. At this time, the forests show signs of human activity.
 Abstracts 81

P-18

DATING LANDNÁM IN THE FAROE ISLANDS

M J Church1, SV Arge2, P L Ascough3, J M Bond4, G T Cook3, S J Dockrill4, A J Dugmore5,
E Dunbar3, R Edvardsson6, T H McGovern6
1. Dept. of Archaeology, Durham University, Durham, DH1 3LE, UK.
2. Føroya Forminnissavn, Hoyvik, Postbox 1155, FR-110 Tórshavn, Føroyar, Denmark.
3. Scottish Universities Environmental Research Centre, Scottish Enterprise Technology Park, Rankine Avenue, East Kil-
bride G75 OQF, UK.
4. Dept. of Archaeological Sciences, Division of Archaeological, Geographical and Environmental Sciences, University of
Bradford, West Yorkshire, BD7 1DP, UK.
5. Institute of Geography, School of GeoScience, University of Edinburgh, Edinburgh EH8 9XP, UK.
6. Anthropology Department, Hunter College CUNY, 695 Park Avenue, New York, NY 10021, USA.

Over 40 AMS radiocarbon dates have been obtained from two archaeological sites on the island of
Sandoy in the Faroes, to assess the timing and duration of the first human settlement, known as land-
nám (“land take”) in Old Norse. The two sites, Undir Junkarinsfløtti and Á Sondum, consist of
Viking to Medieval settlement mounds, with occupation sequences over 3 m deep revealed by
coastal erosion. A variety of single-entity materials (barley grains, cow bones, pig bones, limpet
shells) from a range of archaeological contexts (middens, floor surfaces, hearth deposits) were used
to construct site chronologies. The dates are discussed within three key research themes; 1) the tim-
ing of human settlement in the Faroes, 2) the length of occupation and the timing of abandonment at
the three sites, and 3) recurrent dating issues on Faroese archaeological sites, including the survival
of archaeological remains from the earliest settlement, ecofact taphonomy, and marine radiocarbon
reservoir effects. The findings are then placed within the wider context of the Norse colonization of
the North Atlantic islands of Atlantic Scotland, Faroes, Iceland, and Greenland. The sites were exca-
vated as part of an international research project investigating the settlement of the Faroes (“The
Heart of the Atlantic Project”) and the radiocarbon dates were provided through the “Landscapes
circa landnám” project funded by the Leverhulme Trust.

P-19

PROBLEMS IN DATING ARTIFACT ASSEMBLAGES FROM OPEN-AIR SITES IN WESTERN

AUSTRALIA: MULKA’S CAVE IN CONTEXT
Alana M Rossi1 and R Esmée Webb2
1. Kurongkurl Katitjin, Edith Cowan University, Mount Lawley, WA 6050, Australia.
2. School of Natural Sciences, Edith Cowan University, Joondalup, WA 6027, Australia.

Interpreting the Aboriginal archaeological record poses problems not necessarily encountered in
less arid parts of the world, particularly if artifact sequences changed over time more obviously or
dramatically than they did in Australia. There is frequently little change in the color and/or texture
of the sediment excavated from even quite deeply stratified Aboriginal sites. Why is unclear, but it
leads Australian archaeologists to excavate in arbitrary spits, concealing rather than revealing those
delicate depositional changes that do occur. Hence, the stratigraphy of some sites only becomes
clear after excavation is complete, affecting their interpretation. The areas excavated on Western
Australian Aboriginal sites are also usually small, often only a meter square test pit is dug, exacer-
bating the visual difficulties already mentioned and limiting the types of answers excavation can
82 Abstracts

yield. The types and shapes of Aboriginal stone artifacts also changed little over time, making Aus-
tralian archaeologists very dependent on the results of radiometric assay to provide a chronological
framework for the sites they study. Given these limitations, the problems that can arise are illustrated
by Rossi’s research at Mulka’s Cave, a profusely decorated Aboriginal site situated on the western
edge of the Western Desert, in southwestern Australia. Rossi excavated two 0.5-m square pits into
the area where her Aboriginal informants said they camped at the site when children. She obtained
three 14C dates from 0.4 m of deposit in CA1 and two from 0.3 m of deposit in CA2. The interpre-
tative problems posed by these dates, which are stratigraphically incoherent, caused us to search the
archaeological literature for other Western Australian Aboriginal sites where artifact assemblages
deposited in the open air had been dated radiometrically. It appears that the dates from such
sequences are frequently incoherent. The reasons for their incoherence are explored in this poster.

P-20

NEW DATING EVIDENCE FOR NORTH SEA TRADE BETWEEN ENGLAND, SCOTLAND, AND
NORWAY IN THE 11TH CENTURY AD
D W Hall1, G T Cook2, G K P Muir2, W D Hamilton3
1. SUAT Ltd 55 South Methven St., Perth PH1 5NX, Scotland, UK.
2. SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, Scotland, UK.
3. School of Archaeology & Ancient History, University of Leicester, University Road, Leicester LE1 7RH, England, UK.

This study follows on from previous research at Perth, Scotland, in which we dated carbonized food
residues removed from the external surface of rim sherds of cooking pots of London Sandy Shelly
Ware pottery. The 15 residues that were dated produced 14C ages between 910 ± 35 and 1085 ± 40
BP. Application of a Bayesian approach to the data demonstrated that this pottery type first appeared
in Perth around cal AD 910–1020 and had disappeared by cal AD 1020–1140. This, together with
supporting 14C measurements from associated materials such as leather artifacts from the same
secure contexts, demonstrated that this type of pottery was in use at Perth at least 100 years earlier
than the accepted chronology for this fabric type. Subsequent dating has been carried out on similar
residues from the same fabric obtained from the Billingsgate excavations in London (where the
chronology of this fabric type was first defined) and the Bryggen excavations in Bergen, Norway.
The London and Bergen measurements gave age ranges of 905 ± 35 to 1115 ± 35 BP and 920 ± 35
to 1055 ± 35 BP, respectively, which are almost identical to the range for the Perth samples. This
provides important new information for the trading of fish around the North Sea and the nature and
origin of the inhabitants of early settlements in both Perth and Bergen. The method of dating, the
types of material dated, and the archaeological conclusions arising from the work will be discussed.
 Abstracts 83

EURASIAN ARCHAEOLOGY

P-21

THE NEOLITHIZATION OF SOUTHEASTERN EUROPE: NEW EVIDENCE FROM THE EAST

Pavel M Dolukhanov1, Anvar M Shukurov2, Ganna I Zaitseva3
1. School of Historical Studies, Newcastle University, UK.
2. School of Mathematics and Statistics, Newcastle University, UK.
3. Institute for History of Material Culture, Russian Academy of Sciences, St. Petersburg, Russia.

According to radiocarbon evidence, the earliest manifestations of agriculture in southeastern Europe

are acknowledgeable at ca. 8.6–7.5 kyr cal BC (Franchthi Cave). The next stages of agricultural
expansion include Proto-Sesklo (6.5–6.0 kyr cal BC), Sesklo (6.0–5.3 kyr cal BC), and, further
north, Karanovo I-II (6.1–5.8 kyr cal BC); Karanovo III (5.4–5.1 kyr cal BC), Karanovo IV (5.3–4.8
kyr cal BC), StarËevo-Körös-Cri∫ (5.9–5.5 kyr cal BC), VinËa (5.5–4.0 kyr cal BC), and Linear Pot-
tery Culture (5154 ± 62 BC) in central Europe. Newly available statistically viable series of radio-
carbon dates acknowledge the early signals of pottery-making in the North Caspian area and, the
Middle-Lower Volga at ca. 8–7 kyr and the Lower Don at ca. 7 kyr. Early pottery sites in the Ukraine
and Moldova (Surian, Dniepr-Donetsian, and Bug-Dniestrian) lie in the timespan between 6.2 and
5.0 kyr. This evidence strongly suggests an independent emergence of pottery-making tradition in
the East, and its gradual spread westwards. The later sites apparently overlap with early farming
entities (StarËevo-Körös-Cri∫ and Linear Pottery), which is perceptible in archaeological materials.
This evidence conforms the previously formulated model (Davison et al. 2008), which suggests that
the spread of the Neolithic involved at least two waves propagating from distinct centers, the west-
ern one stemming from the Near East, and the eastern one, spreading through the east via the “steppe
corridor,” resulting in the establishment of the “eastern version” of the Neolithic in Europe. Remark-
ably, in all studied areas, the early Neolithic coincided with the optimal environments of the Atlantic
period, which included a notable increase of rainfall.

P-22

CHRONOLOGY OF THE OBI-RAKHMAT GROTTO (UZBEKISTAN): RESULTS AND PROBLEMS

OF THE PALEOLITHIC KEY-SITE IN CENTRAL ASIA
Andrei I Krivoshapkin1, Yaroslav V Kuzmin2, A J Timothy Jull3
1. Institute of Archaeology & Ethnography SB RAS, Novosibirsk, Russia.
2. Institute of Geology & Mineralogy SB RAS, Novosibirsk, Russia.
3. University of Arizona, Tucson, AZ, USA.

The Obi-Rakhmat Grotto (41°34′8.8″N, 70°08′0.3″E; 1250 m asl) is located ca. 100 km northeast of
the city of Tashkent (Republic of Uzbekistan), and is one of the key Paleolithic sites in Central Asia.
Archaeological excavations begun in 1962 have revealed 22 geological strata containing 40 litho-
logical substrata and 36 archaeological horizons. Obi-Rakhmat lithic assemblages display many
similar features with both late Middle Paleolithic blade industries and early Upper Paleolithic com-
plexes in Southwest Asia and the Siberian Altai Mountains, suggesting a process of gradual transi-
tion from the Middle to the Upper Paleolithic occurred in western Central Asia. New hominid
remains from Obi-Rakhmat Grotto discovered in 2003 mark the first additions of human fossils
84 Abstracts

from Uzbekistan in over 65 years (after discovery of the Teshik-Tash child) and express a mixture
of archaic and modern traits. Different chronometric methods, 14C, OSL, U-series, and ESR, were
applied in order to obtain information about the age of the Middle and Upper Paleolithic stone
assemblages and hominid remains. It turned out that only 14C dates are reliable in terms of corre-
spondence to general framework of the Paleolithic of Central Asia and neighboring regions (Near
and Middle East, southern Siberia, and Mongolia). After deleting outliers, the site can be dated from
ca. 36,100 BP (layer 7) to ca. 48,800 BP (layer 14.1). The U-series dating results are unreliable due
to high U content and presence of detritus, which contaminates sediments (travertine). The OSL dat-
ing gave uniform ages for all cultural successions (ca. 10 m of deposits), and can testify only for
very rapid sedimentation rate. Results of ESR dating depend greatly on the choice of uptake model.
Dates calculated for early uptake model generally fit 14C data (although not exactly). The linear
uptake model (Skinner et al. 2007) makes sediments very old (ca. 55,000-90,000 years ago), and this
is not in a good correspondence with archaeological data.

REFERENCE
Skinner AR, Blackwell BAB, Mian A. et al. 2007. Radiation Measurements 42:1237–42.

P-23

THE RESEARCH OF ENVIRONMENTAL ARCHAEOLOGY AND 14C CHRONOLOGY IN DAN-

JIANG, SHAANXI PROVINCE, CHINA
Yizhi Zhu1, Peng Cheng1, Yu Huagui1, Yang Yachang2, Kang Zhihai1, A J T Jull3
1. State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences,
Xi’an 710075, China.
2. Shaanxi institute of Archaeology, Xi’an 710054, China.
3. NSF Arizona AMS Facility, University of Arizona, Tucson, AZ 85721, USA.

The Danjiang River upstream region lies in the Shang-Dan faulted basin of the Qinling orogenic belt
in north China (33°51′N, 109°58′E). The annual average temperature of this region is 12 °C, annual
average rainfall is 750 mm, and the elevation is about 600–800 m. Our research focuses on the com-
bination of archaeological culture relics and the corresponding sequence of Holocene loess-paleo-
sol. There are continuous sequences of archaeological cultures relics since the Holocene, namely,
the Lao Guantai culture (or the Old Machine culture), Yang Shao culture (or the Painted Pottery cul-
ture, Banpo type), the Long Shan culture (or the Black Pottery culture), and the Xia, Shang, Zhou
cultures. Beside the relics, a 2- to 3-m-thick Holocene loess-paleosol sequence is found in which the
So paleosol contains 2 layers of complex soil. We sampled 1–2 cm intervals for high-resolution anal-
yses. Charcoal from relics and organic matter from loess-paleosol samples were supplied for 14C
dating. In order to obtain more precise ages of different cultures, we used Bayesian analysis to cali-
brate the 14C ages. The preliminary 14C chronological sequence of the cultural development and
environmental change since Holocene was reconstructed as follows:
The bottom of Holocene paleosol: 10440~8000 BP. The climate changed from cool to warm. The
Lao Guantai culture: 8000~6700 BP, the climate was warm and wet. Yang Shao culture: 6700~4900
BP, Early stage, the climate was warm and wet; Late stage, the climate fluctuated. The Long Shan
culture, Xia Dynasty, Shang Dynasty: 4900~3100 BP, the climate changed from warm and wet to
warm and dry; Zhou Dynasty: 3100~2200 BP, the climate changed from warm and wet to cool and
dry, recent loess began deposition.
 Abstracts 85

P-24

ON THE EXECUTION AGES OF GENJI NARRATIVE SCROLL AND BAN DAINAGON NARRA-
TIVE SCROLL BY RADIOCARBON DATING OF KOHITSUGIRE CALLIGRAPHIES ATTRIBUTED
TO ASUKAI MASATSUNE
Hirotaka Oda1 and Kazuomi Ikeda2
1. Center for Chronological Research, Nagoya University, Japan.
2. Faculty of Letters, Chuo University, Japan.

Genji-monogatari emaki is a narrative scroll depicting Murasaki Shikibu’s novel Tale of Genji by
colorful paintings and graceful calligraphy. A part of the calligraphy in the scroll is attributed to
Asukai Masatsune (1170–1221). The calligraphy in Ban Dainagon narrative scroll dealing with the
Otenmon incident is also attributed to him. They are two of the four oldest examples among extant
narrative scrolls and were designated as the national treasures of Japan. Many ancient manuscripts
written in the Heian-Kamakura period (9th–early 14th century) had been divided into leaves in order
to appreciate their handwriting as hanging scrolls. Such ancient paper sheets with elegant calligra-
phy are called kohitsugire. Old manuscripts in the Heian-Kamamkura period are rarely discovered;
therefore, kohitsugire can be significant materials for historical studies. Two kohitsugire called
Imaki gire and Kinginkirihaku wakanroeisyu gire are attributed to Asukai Masatsune according to
the kiwamefuda certificates, and also have identical handwriting as the Genji and Ban Dainagon nar-
rative scrolls. Paleographical study, however, proposed a different theory—that they are the genuine
handwriting of Fujiwara no Norinaga (1109–1180?) and not by Asukai Masatune. This would mean
that the execution of the Genji and Ban Dainagon narrative scrolls ascends to the 12th century.
Therefore, in this study, we measured the 14C ages of Imaki gire and Kinginkirihaku wakanroeisyu
gire by AMS. The results indicated that they were written from the middle 11th to the late 12th cen-
tury. It includes the period when Fujiwara no Norinaga had flourished as a calligraphist and is too
early to be Asukai Masatsune’s handwriting. 14C dating supported the paleographical theory and
narrowed the execution ages of the Geiji and Ban Dainagon narrative scrolls to the late 12th century.

P-25

A CASE STUDY OF THE ORIGIN AND AGE OF AGRICULTURE IN NORTHERN CHINA

Cheng-Bang An
CAEP, Key Laboratory of West China's Environmental System (Ministry of Education), Lanzhou University, Lanzhou
730000, P R China.

Study on the origin of agriculture research in the world and northern China shows that northern
China is the only area where it does not seem to unfold, in situ, from local hunter-gatherers in 10
instances known worldwide in which agriculture evolved independently. Agriculture evolution from
domesticated seeds is not clear devoid of systematic agriculture survey. So published radiocarbon
dates of archaeological sites and materials in northern China were collected, and a large scale survey
was carried out since 2001 in order to answer these questions. Here are some of our findings.
1. The periods 10–8, 7.4–6.6, 4.8–4.5, and 3.8–3.5 kyr BP are the low value zones based on the
cumulative probability of 1364 radiocarbon dates in northern China. The periods 10–8 and 7.4–6.6
86 Abstracts

kyr BP are right on the transition from the Paleolithic to Neolithic and Early Neolithic to Late
Neolithic, respectively, and both periods witnessed a “culture gap” in northern China. So no contin-
uous archaeological record for origin of agriculture was ever discovered in north China.
2. The agriculture survey indicates that hunter-gathering was dominated by the low ubiquity of half-
domesticated Panicum miliaceum during 8–7.3 kyr BP, but developed agriculture a quite higher
ubiquity of domesticated Panicum miliaceum after 6.4 kyr BP. The replacement of hunter-gathering
by agriculture may take place in the transition from Dadiwan 1 to Late Banpo, Yangshao culture.
The first continuous archaeological record exposed at the Dadiwan site makes it possible to detect
the climate change for the origin of early agriculture and the transition from the Paleolithic to
Neolithic in northern China. Environmental proxies reconstructed by magnetic susceptibility and
CaCO3 suggest that the periods 60–25 kyr BP and 12–7 kyr BP are moister, while the periods 25–
12 kyr BP and 7.2–6.4 kyr BP were much drier. The period from 7.2–6.4 kyr BP is very important
for the origin of agriculture in northern China.

P-26

APPLICATION OF RADIOCARBON DATING TO THE RESEARCH OF PREHISTORIC DISAS-

TERS AND THE EFFECTS ON HUMAN MIGRATION IN GUANTING BASIN, QINGHAI PROV-
INCE, CHINA
Guanghui Dong, Ma Minmin, Chen Fahu
CAEP, MOE Key Laboratory of West China’s Environmental System, Lanzhou University, Lanzhou 730000, China.

The Lajia site (35°48′N, 102°42′E), located in the second terrace of the Yellow River, Guanting
Basin, Qinghai Province, China, was ruined by disasters including floods, earthquakes, and moun-
tainous torrents during the Qijia cultural period (4300–3900 BP) (Yang et al. 2003). To study the
prehistoric disasters and the effects on human migration, we investigated the stratum of another
Qijia cultural site, Qingquanhantai site (35°52′N, 102°50′E), in the third terrace of the Yellow River,
3 km northeast from the Lajia site. We collected 8 charcoals and a layer of ash from an ash pit and
cultural layer for radiocarbon dating, and compared these dating dates with those of Lajia site. As
shown by our results, the ages of Qingquanhantai site are between 3992–3517 cal yr BP, while the
14 AMS radiocarbon dates of Lajia site are mainly between 4400–3900 cal yr BP (14C Laboratory,
the Institute of Archaeology, CASS, 2003, 2005). The archaeological records of both sites show
similar characteristics. These results suggest that the prehistoric disasters in Guanting Basin hap-
pened during 4000–3900 cal yr BP, which may have been the result of the human migration from
lower land such as the Lajia site (1793 m asl) to the higher land such as Qingquanhantai site (1832
m asl). The Qijia culture existed in Quanting Basin until 3600–3500 cal yr BP.
 Abstracts 87

P-27

CHRONOLOGICAL STUDY OF THE OCCUPATION OF THE GONUR DEPE BRONZE AGE SITE
(TURKMENISTAN)
Ganna I Zaitseva1, Nadezhda A Dubova2, Anatoliy A Sementsov1, Nataliya Bourova1,
Högne Jungner3
1. Institute for History of Material Culture, Russian Academy of Sciences, St. Petersburg, Russia.
2. Institute of Ethnology and Anthropology Russian Academy of Sciences, Moscow, Russia.
3. Helsinki University, Finland

The very famous Gonur-Depe site belonging to the Bronze Age is located in the ancient delta of the
Murgab River. This site is represented by a place-temple and ritual complexes, indicating that Gonur
Depe was probably an administrative center of the ancient Margiana country. It is very important to
determine the chronological stages of the occupation of this place using both archaeological and
radiocarbon dates. There are very representative series of the 14C determinations produced by dif-
ferent radiocarbon laboratories. Materials for dating were charcoal, animal and human bones, and
wood from constructions. Samples from the Palace, Kremlin (North Gonur), and Temenos (South
Gonur), from the north, east, west and south areas out of Kremlin wall, from the Main and Royal
Gonur necropolises and from Area 16 (ritual burial of animals) were dated. Essential differences
between the results of different laboratories were not observed. The radiocarbon determinations
establish that this site existed during the period 2500–1500 BC. The most intensive exploration of it
can be connected with the eve of 2000 BC. Around 1500 BC, the use of this site stopped completely.
The work was possible thanks to the financial support from the Russian Foundation for Basic
research (project 07-06-00062a).

P-28

MEASURING THE AGE OF AN ANCIENT POTTERY WORKSHOP IN SOUTHWEST CHINA

X D Ruan1, W M Wu1, Y J Guan1, H J Wang1, Jiang Shan2, He Ming2, Z M Xiong3, K X Liu4
1. College of Physical Science and Engineering Technology, Guangxi University, Nanning 530004, China.
2. China Institute of Atomic Energy, P.0. Box 275(50), Beijing 102413, China.
3. Guangxi Provincial Institute of Cultural Relics and Archaeology, Nanning 530023, China.
4. Institute of Heavy Ion Physics Key Laboratory of Heavy Ion Physics, Ministry of Education, Peking University, Beijing
100571, China.

Hepu is an ancient area located in southwest China with more than 2200 years of history. An ancient
pottery workshop, a site of original pottery production, was excavated in Hepu County 3 years ago.
In this workshop, all the processes for producing pottery can be observed very well in the excavated
location. More than 50 integrated relics were found in the workshop.
During the period of the Han Dynasty, Hepu was an important harbor. It was one of three initial har-
bors connecting the Silk Road to the sea. The excavated pottery workshop is powerful proof to sup-
port the notion of Hepu’s importance in that time period.
In order to determine the exact age of the workshop, samples of bamboo and wood were taken in the
excavated workshop, and the 14C results were measured in the Peking University AMS lab. The cal-
ibrated age range (1 σ) of the bamboo and the wood were determined to be 200–90 BC and 240–160
BC, respectively.
88 Abstracts

P-29

AMS RADIOCARBON DATING OF THE TANE IV TEPHRA DISTRIBUTED ON TANEGASHIMA

ISLAND, SW JAPAN
M Okuno1, H Naruo2, M Torii3, T Kobayashi4
1. Dept. of Earth System Science, Faculty of Science, Fukuoka University, Fukuoka 814-0180, Japan.
2. Takeokadai High School, Kagoshima 890-0022, Japan.
3. Kumamoto Gakuen University, Kumamoto 862-8680, Japan.
4. Dept. of Earth and Environmental Sciences, Faculty of Science, Kagoshima University, Kagoshima 890-0065, Japan.

Four local tephra, Tane I, II, III and IV in ascending order, are distributed between two widespread
tephras, Kikai-Tozurahara (K-Tz: 95 kyr) and Aira-Tn (AT: 29 cal kyr BP), on Tanegashima Island,
20 km off the south end of Kyushu Island (Okuno and Kobayashi 1994). The Yokomine archaeolog-
ical site located in the southern part of the island is a Paleolithic ruin interbedded between the Tane
IV and Tane III tephras. The 14C dates of charcoal fragments were previously reported as >30,260
BP (GaK-16775), >31,080 BP (GaK-16776), >28,110 BP (GaK-16777) in this site (Minamitane
Town Education Board 1993). To determine the date of eruption of the Tane IV tephra, we con-
ducted 14C dating of the paleosol just below the AT and Tane IV tephras, and charcoal fragments
between the Tane IV and Tane III tephras, using a NEC 15SDH-2 AMS system at the Japan Atomic
Energy Agency Tono Geoscience Center (Xu et al. 2000). The obtained 14C dates are 24,310 ± 140
BP (JNC-7816), 28,670 ± 180 BP (JNC-7815), and 32,150 ± 230 BP (JNC-7817) in descending
order. These dates are consistent with the stratigraphic relation, and can be calibrated to 29 cal kyr
BP, 34 cal kyr BP, and 37.5 cal kyr BP, respectively, using the Fairbanks calibration program (Fair-
banks et al. 2005). Thus, it seems that the age of the Tane IV tephra is 34 cal kyr BP. The charcoal
date (37.5 cal kyr BP) not only constrains the Tane IV tephra, but also serves as a possible date for
the archaeological site.

P-30

RADIOCARBON CHRONOLOGY FOR EARLY CAVES OF MOGAO GROTTOES, DUNHUANG,

CHINA
Qinglin Guo1, Hiromi Takabayashi2, Toshio Nakamura3, Chen Gangquan1, Ken Okada2,
Boming Su1, Hiroshi Nishimoto4
1. Institute of Conservation, Dunhuang Academy, Mogaoku, Dunhuang, Gansu Province 736200, China.
2. Japan Center for International Cooperation in Conservation, National Research Institute for Cultural Properties, 14-43
Ueno Park, Taito-ku, Tokyo, 110-8713, Japan.
3. Center for Chronological Research, Nagoya University, Chikusa, Nagoya 464-8602, Japan.
4. Graduate School for Environmental Studies, Nagoya University, Chikusa, Nagoya 464-8602, Japan.

The Mogao Grottoes site at Dunhuang is one of the largest stone cave temples in China, and pos-
sesses 735 caves with Buddhist mural paintings. As a scientific investigation on chronology of early
caves of Mogao Grottoes, 14C dates were measured by AMS on plant remains collected from four
early caves, 268, 272, 275, and 285. The three caves (268, 272, and 275) are regarded as earliest
existent caves of Mogao Grottoes by archaeological analysis. The other cave, 285, possesses
inscriptions on the north wall mentioning the oldest dates of the Chinese Mogao era.
 Abstracts 89

Pieces of the plant materials were collected as samples for AMS 14C dating at Nagoya University,
from a plaster layer of mural paintings and core material of statues. In the earliest caves, 11 samples
were collected from the murals and statues initially produced. Two samples from cave 275 showed
calibrated 14C ages of around cal AD 380–430 in the 1-σ range of probability density distribution.
The other samples resulted in a time interval of cal AD 400–550. The calibrated 14C ages obtained
for the samples taken from initially painted murals and the statues in cave 285 are consistent with the
date given by the inscription remaining on the north wall.

P-31

MEASUREMENT OF RADIONUCLIDES IN ANCIENT RELICS OBTAINED FROM MATSUSAKI

SITE AND HIROHATA SHELL MOUND ON PACIFIC SITE, JAPAN
T Ohta1, T Kubota1, Y Saito2, S Fukutani1, T Fujii1, A Ando3, T Nakano2, E Nakata4, Y Mahara1
1. Research Reactor Institute, Kyoto University, Japan.
2. Research Institute for Humanity and Nature, Japan.
3. College of Natural Sciences, Pusan National University, Japan.
4. Central Research Institute of Electric Power Industry, Japan.

The production of salt from other than seawater is dependent on resources such as rock salt and salt
lakes. In ancient Japan, which has no rock salt or salt lakes, the source of almost all salt is seawater.
Matsusaki site (Aichi prefecture) and Hirohata shell mound (Ibaragi prefecture) are famous salt-pro-
duction sites on the Pacific coast in Japan. Hirohata shell mound and Matsusaki site are from the
Jomon period (BC) and the period from Kofun to Kamakura (AD), respectively. The age between
Matsusaki site and Hirohata shell mound are different by more than 1000 years. In the 2 salt-produc-
tion sites, a white substance was discovered from the salt manufacturing earthenware and a wooden
tubing vessel. In our study, we measured 14C, 228Ra/226Ra, 234U/238U, and 87Sr/86Sr, and δ13C in the
substances that adhered to the wooden tubing vessel for Matsusaki site and to earthenware for Hiro-
hata shell mound to compare the characters of the substances obtained from 2 salt-production sites.
As for two relics, the preservation situation is contrasted as follows: 1) the relics of Hirohata shell
mound had been left on the ground for several decades, and 2) the relics for Matsusaki site had piled
up thick in the log and had been buried in sands on the coast. As the relics for Hirohata shell mound
were in the environment that was dug up, and exposed directly to rainwater and undertook the pol-
lution of modern 14C after the nuclear test, the 14C age was thought to be younger than the actual age.
Yet, the 14C age obtained from relics in Matsusaki was nearly same as the reported age, suggesting
that the environments and preservation situations affected the 14C in the substances.

P-32

15TH TO 17TH CENTURY JAR BURIALS IN THE CARDAMOM MOUNTAINS, CAMBODIA:

PRELIMINARY RESULTS
Nancy Beavan Athfield1, John Miksic2, K Rethy Chhem3, Kyle Latinus4, Louise Shewan5,
Dougald O’Reilly5
1. Rafter Radiocarbon, GNS Science, New Zealand.
90 Abstracts

2. National University of Singapore, Singapore.

3. University of Western Ontario, Canada.
4. Pannasastra University of Cambodia, Phnom Penh, Cambodia.
5. University of Sydney, Australia.

In March 2003, an initial investigation was made of secondary burials of human bone packed into
stoneware jars at four rock ledge sites in the Cardamom Mountains, Kingdom of Cambodia. Radio-
carbon dating and stable isotope analysis have been completed on three bone samples (one rib bone
and two separate skulls) representing three individuals from one stoneware jar at Site 4. A Ward &
Wilson T' test (Ward and Wilson 1978) indicates that all three dates are not significantly different (df
2, T' = 1.7). If the assumption is made that the individuals all died and their remains were placed in
the jar at the same time, then the dates can be combined (Combine function, OXCAL v3.10; Bronk
Ramsey 2001, 2005) to give a combined calibrated radiocarbon age of 374 ± 18 years BP. Due to the
wiggle in the calibration curve at this time, the calendar age ranges diverge into two possible periods
of AD 1440–1530 plus AD 1570–1630. The associated 15th century Chinese and Sisatchanalai
ceramics in the site support the conclusion that the radiocarbon ages give an accurate range for the
deposition of the objects in the overhangs where they were found. Stable isotope evidence also sug-
gests the radiocarbon ages are not affected by marine influences. We also report EDXRF analysis of
glass beads found within the jars. There are outstanding questions about the funerary practice, the
selection of these remote sites, and the cultural affinity of the remains, as the sites are unique in the
Cambodian cultural-historical context.

REFERENCES
Bronk Ramsey C. 2001. Development of the radiocarbon program OxCal. Radiocarbon 43(2A):355–63.
Bronk Ramsey C. 2005 OxCal Program v3.10. Oxford: Oxford Radiocarbon Accelerator Unit.
Ward GK, Wilson SR. 1978. Procedures for comparing and combining radiocarbon age determinations: a critique. Archae-
ometry 20:19–31.

BONE CHEMISTRY
P-33

EXAMINING ASCETIC ACID TREATMENT OF BONE CARBONATE FOR RADIOCARBON DAT-

ING: PRELIMINARY RESULTS
N Beavan Athfield1, K Sutton1, B Trompetter1, B Barry1, R Soong1, L Shewan2, D O’Reilly2
1. National Isotope Centre, GNS Science, New Zealand.
2. University of Sydney, Australia.

Trials of acetic acid treatment methods to eliminate exogenous CaCO3 and obtain reliable radiocar-
bon ages on the bioapatite fraction of bone (Haynes 1968; Krueger 1991) can result in bone ages that
sometimes too young (Tamers and Pearson 1965; Taylor 1994). It has been suggested the problem
relates to burial environment conditions; for example, bone carbonate ages from arid climates have
produced ages in agreement with collagen or charcoal dates (Saliège et al. 1995; Person et al. 1998).
But why are discordant ages usually younger? We tested paired samples of unburnt bone carbonate
and tooth protein (dentine) in human mandibles from Cambodian archaeological sites. We present
 Abstracts 91

preliminary results from experiments using a modified Lanting et al. (2001) method, with variations
on treatment suggested by Garvie-Lok et al. (2004) to examine changes in the composition of pre-
and post-chemical treated bone. Qualitative X-ray defraction (XRD) determined approximate rela-
tive abundance of calcium carbonate vs. hydroxyapatite, particle induced X-ray emission (PIXE)
determined the elemental makeup of bone samples, and Rutherford backscattering spectrometry
(RBS) provided quantitative in situ elemental profiles.

REFERENCES
Garvie-Lok et al. 2004. Journal of Archaeological Science 31:763–76.
Haynes V. 1968. Science 161(3842):687–88.
Krueger HW. 1991. Journal of Archaeological Science 18:355–61.
Lanting JN et al. 2001. Radiocarbon 43(2A):249–54.
Person A et al. 1998. Pré-actes du 3ème Congrès International 14C et Archéologie, Lyon 1998. p 77–8.
Saliège et al. 1995. Journal of Archaeological Science 22:301–12.
Tamers MA, Pearson FJ 1965. Nature 208:1053-–5.
Taylor RE. 1994. Radiocarbon dating of bone using accelerator mass spectrometry. In: Bonnichsen R, Steele DG (eds.).
Method and Theory for Investigating the Peopling of the Americas. Corvallis: Center for the Study of the First Americans.
p 27–44.

P-34

REFINING THE PRETREATMENT BACKGROUND CORRECTION OF COLLAGEN EXTRACTED

BY ULTRAFILTRATION AT ORAU
R E Wood, C Bronk Ramsey, T F G Higham
RLAHA, University of Oxford.

Accurate and precise calculation of the amount and age of carbon added to samples during labora-
tory processing is critical for the production of reliable radiocarbon dates. This is particularly perti-
nent when dating beyond 30 kyr BP as contamination is usually found to be modern in age. The
lengthy pretreatment required for the effective removal of contamination incorporated from the dep-
osition environment of bone makes this sample type especially susceptible to laboratory-derived
contamination prior to combustion and graphitization. We present a refinement of the pretreatment
correction for collagen extracted with the ultrafiltration protocol (Ramsey et al. 2004) constructed
from repeat measurements of bone of infinite age.

REFERENCE
Bronk Ramsey C, Higham T, Bowles A, Hedges R. 2004. Improvements to the pretreatment of bone at Oxford. Radiocarbon
46(1):155–63.
92 Abstracts

P-35

EFFECT OF TEMPERATURE ON THE δ13C VALUES OF CARBON EXTRACTED FROM CRE-

MATED BONE SAMPLES
P Naysmith, G T Cook, E Dunbar, S Xu
SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, Scotland, UK.

At the last conference we presented the results of an intercomparison study on 14C dating of cre-
mated bone (Naysmith et al. 2007), which demonstrated that the laboratories taking part obtained
consistent ages using the method developed by Lanting et al. (2001). This method involves the
extraction of the small carbonate component contained within the hydroxyapatite-based inorganic
fraction, which is locked into the bone during cremation at temperatures above 600 °C. This has now
become a standard method for analyzing cremated bone samples, and during 2008 the SUERC
Radiocarbon Laboratory analyzed 86 cremated bone samples from various sites across the UK.
However, when analyzing the data we noted a large range in the δ13C values (–16.6‰ to –28.3‰),
obviously much wider than is observed when analyzing bones from inhumations. Large δ13C varia-
tions were also observed during the cremated bone intercomparison study (Naysmith et al. 2007).
What needs to be established is how much of the variation in the δ13C is due to (1) the diet of the per-
son or animal, (2) the cremation temperature of the pyre, and (3) variations in pretreatment of the
samples. From the above, the question then arises of whether it is possible to derive any dietary
information from cremated bone δ13C values. We cremated two bone samples (one human with a
terrestrial diet and one dolphin: these represent the light and heavy ends of the δ13C range for bone,
respectively) at varying combustion temperatures and then extracted the carbon using identical pre-
treatment schemes and subsequently undertook both δ13C and 14C analyses. This poster will present
the results obtained from these studies.

REFERENCES
Lanting JN, Aerts-Bijma AT, van der Plicht J. 2001. Dating cremated bone. Radiocarbon 43:249–54.
Naysmith P, Scott EM, Cook GT, Heinemeier J, van der Plicht J, van Strydonck M, Bronk Ramsey C, Grootes PM, Freeman
SPHT. 2007. A cremated bone intercomparison study. Radiocarbon 49(2):403–8.

P-36

A NONDESTRUCTIVE PRESCREENING METHOD FOR BONE COLLAGEN CONTENT USING

MICRO-COMPUTED TOMOGRAPHY
Jennifer A Tripp1, Maria E Squire2, Julie Hamilton3, Robert E M Hedges3
1. Department of Chemistry, University of Scranton, Pennsylvania, USA.
2. Department of Biology, University of Scranton, Pennsylvania, USA.
3. Research Laboratory for Archaeology, University of Oxford, UK.

Isolation of bone collagen for radiocarbon dating is a labor-intensive and time-consuming process
that sometimes results in unacceptably low protein recovery. A suitable nondestructive prescreening
method would enable collagen recovery to be predicted, allowing for bones with low collagen con-
tent to be excluded before the costly pretreatment process begins. Micro-computed tomography
(microCT), a nondestructive technique that uses X-rays to produce high-resolution (30 μm) three-
dimensional images of materials such as bone, may provide a suitable prescreening option for bones
of questionable preservation.
 Abstracts 93

Preliminary studies involving sheep and cattle long bones of variable preservation from Etton, Cam-
bridgeshire, UK, suggest that bone volume fraction calculations from microCT images of cortical
bone correlate well with measured collagen recovery using normal chemical pretreatment processes.
While high-resolution images offer a more detailed insight into bone morphology, we have found
that low-resolution images, which take approximately one-quarter of the time to acquire (e.g. 4
hours as opposed to 16), also offer suitable results and correlations with collagen recovery. Because
the bone volume fraction measurements can be spatially resolved, this technique also offers the
promise of prescreening bones prior to sampling to determine the most suitable location within the
bone for high collagen recovery. We will demonstrate through images and image analysis the prom-
ise of microCT as a prescreening method for collagen recovery in archaeological bone.

P-37

RADIOCARBON AGE OF BONE FRACTIONS FROM ARMENIAN PREHISTORIC SITES

Alexander Cherkinsky1 and Christine Chataigner2
1. University of Georgia, Athens, USA.
2. CNRS., Lyon, France.

Prehistoric cultures in Armenia are still poorly known because of the lack of traditionally dating
material (e.g. wood and charcoal) and most of the bone artifacts are poor preserved. AMS radiocar-
bon dates are invaluable in constructing an accurate chronological frame. The radiocarbon dating of
bone has been problematic since the early days of 14C dating due to frequent discrepancy between
bones and associated charcoal or wood dates and/or between different fractions isolated from a sin-
gle bone. Recently published studies show that these problems still exist. Five archaeological sites
have been dated on the bone material with the different degree of preservation. The bone samples
have been collected from sites representing the Middle/Upper Paleolithic, Chalcolithic, and Early
Bronze periods. Most of the bone samples are poorly preserved. We describe the separation tech-
nique for the extraction of both collagen and bioapatite fractions. In many cases where bone had
very low organic material content, the collagen fractions yielded a rejuvenated age, although the
ages of bioapatite fractions were found to be in good agreement with archaeological artifacts. In
cases where bone was well preserved, both fractions exhibited ages in good agreement with arti-
facts. The proper pretreatment of bone samples permits the separation of diagenetic, secondary car-
bonates from bioapatite carbonates if their structure has not been degraded completely and exposed
to isotope exchange in the environmental.The accuracy of radiocarbon dating of bone material
always depends on its degree of preservation, and each case should be carefully evaluated to deter-
mine which fraction is less contaminated in order to date the burial event.

P-38

A NEW METHOD FOR CORRECTING THE MARINE RESERVOIR EFFECT ON RADIOCARBON

AGE OF HUMAN REMAINS BASED ON NITROGEN ISOTOPIC COMPOSITIONS OF INDIVIDUAL
AMINO ACIDS IN COLLAGEN
Yuichi Naito1, Noah V Honch2, Yoshito Chikaraishi3, Naohiko Ohkouchi3, Minoru Yoneda1
1. Graduate School of Frontier Sciences, University of Tokyo, Japan
94 Abstracts

2. Research Laboratory for Archaeology and the History of Art, University of Oxford, UK.
3. Institute of Research on Earth Evolution Biogeosciences, Japan Agency for Marine-Earth Science and Technology, Japan.

Compound-specific nitrogen isotopic analysis of individual amino acids from bone collagen was
applied to archaeological materials as a novel tool for investigating marine protein intake by ancient
humans. It has been shown recently that the nitrogen isotopic composition of some amino acids
from biological samples is useful for estimating their trophic levels in aquatic environments. In this
study, we applied this approach to archaeological human and faunal samples from some Japanese
archaeological sites in order to evaluate relative importance of marine protein consumed by the indi-
vidual humans. Consequently, we established an evaluation method using glutamic acid and pheny-
lalanine for the proportion of marine protein in diets, and estimated the percentage of marine protein
from the total dietary protein. Importantly, it is possible to perform this estimation by analyzing
human remains and only a single animal sample from each marine and terrestrial ecosystem, without
assuming contentious figures of bioenrichment from diet to collagen and average values with each
nutritional group with larger variability from an isotopic point of view. This method can supply reli-
able correction for omnivorous animals including humans.

CALIBRATION

P-39

SCANDINAVIAN MODELS: RADIOCARBON DATES FROM SCANDINAVIAN PASSAGE

GRAVES AND WHAT THEY TELL US ABOUT THE NEOLITHIC SOCIETY
Bettina Schulz Paulsson
Graduate School Human Development in Landscapes, University of Kiel, Germany.

Approximately 2700 radiocarbon results are today available out of European Megalithic contexts.
The interpretation of these radiocarbon dates is often difficult. It is not easy to connect many of these
dates out of their archaeological context to the construction or to the burial phase of these graves.
Scandinavia is a special megalith region, since it is the only place in Europe with radiocarbon dates
directly connectable to the construction of the passage graves, considering the estimable bone pres-
ervation and new dating sequences. Some 188 radiocarbon results are now available out of Scandi-
navian passage graves. The dates from birch bark as filling material between dry walls are allowing
to build up a sequence for the construction phase of these graves in Denmark from the 35th century
on. In Sweden, new sequences of radiocarbon dates from Öland and Falbygden suggest that these
graves were built in the first half of the 35th century. With Bayesian models, it is possible to untan-
gle the nuances of the different burial periods and to get an approach to the organization of the Scan-
dinavian Middle Neolithic society. This poster presents the sequences in an interpretive Bayesian
statistical framework with alternative archaeological interpretations.
 Abstracts 95

P-40

A WIGGLE-MATCHED DATE FOR THE COPPER AGE CEMETERY AT MANERBA DEL GARDA,
NORTHERN ITALY
Lawrence H Barfield1, Sturt W Manning2, Thomas F G Higham3, Erio Valzolgher4
1. Institute of Archaeology and Antiquity, Arts Building, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK.
2. Malcolm and Carolyn Wiener Laboratory for Aegean and Near Eastern Dendrochronology, Cornell Tree-Ring Laboratory,
B-48 Goldwin Smith Hall, Cornell University, Ithaca, NY 14853-3201, USA.
3. Oxford Radiocarbon Accelerator Unit, Research Laboratory for Archaeology and the History of Art, Dyson Perrins Build-
ing, South Parks Road, University of Oxford, Oxford OX1 3QY, UK.
4. Ricerche Archeologiche snc/Archäologische Untersuchungen OHG, via Guglielmo Marconi/Marconi Strasse 8, I-39042
Bressanone/Brixen (Bolzano/Bozen), Italy.

Excavated between 1976 and 1994 by Birmingham University, the cemetery at Manerba del Garda
represents one of the richest sources of evidence for 3rd millennium BC collective burial in Europe.
This cemetery occupies a natural rockshelter, the Riparo Valtenesi, situated at the base of a high cliff
on the western shore of Lake Garda, Brescia, Italy. The archaeological research conducted at the site
revealed the existence of wooden collective burial chambers and several other ritual features dating
to the Copper Age. In order to further refine the absolute dating of burial activity at the rockshelter,
a charred construction oak timber has been sampled from burial chamber 133 for dendro radiocar-
bon wiggle-matching. The results from DWM analysis (the first of its kind for the Italian Copper
Age as a whole), along with a series of radiocarbon measurements previously obtained for this cem-
etery, will be discussed within the broader framework of the north Italian Copper Age chronology.

P-41

DATING A LATE NEOLITHIC ENCLOSURE AT MARNE BARRACKS, NORTH YORKSHIRE,

ENGLAND
Andrew Millard, Andy Platell, Duncan Hale
Department of Archaeology, Durham University, UK.

Open area excavation conducted in advance of development on the former airfield at Marne Bar-
racks, Catterick, North Yorkshire, England, in 2004 identified a late Neolithic palisaded enclosure.
The enclosure consisted of two concentric sub-circular palisades, and projections from the exposed
structure indicate that the total area was about 2.75 ha. Each palisade was formed from a series of
closely-spaced radially-aligned slots separated from each other by gaps of ~0.1 m. On excavation
each slot resolved into two radially-aligned postholes at depth, the posts being ~1 m apart from cen-
ter to center. Thus, each palisade consisted of a double circle of posts. Most of those on the western
side of the monument had been sufficiently carbonized for the individual posts to be identifiable.
The Marne Barracks enclosure exhibits significant differences to other known examples of this
monument type and in particular the “paired post” arrangement of a double circuit of posts in each
palisade is unparalleled in any other known example.
Dating is crucial to place this monument in its context, and allow comparisons with other sites in the
region. Twenty-one radiocarbon samples were collected from the posts and post-holes. This paper
will discuss the Bayesian models used to refine the dates from the basis of less-than-ideal samples
and the process leading to selection of the best model, which places the date of the enclosure at
2530–2310 cal BC.
96 Abstracts

P-42

WHAT DOES IT MEAN TO SUM THE PROBABILITIES OF DATES?

Andrew Millard
Department of Archaeology, Durham University, UK.

In a number of contexts, but notably when dealing with calibrated radiocarbon dates, archaeologists
and other paleoscientists sometimes use the sum of probability distributions as a summary of the
spread of the dates of a particular temporal unit, whether a stratum, a site, or an entire archaeological
culture. Discussions at the recent Fifth Radiocarbon and Archaeology conference (Zürich, March
2008) highlighted problems with using this approach. Despite their extensive use in the literature,
there is no paper from a statistician which recommends their use, nor any account of whether a date
range (whether interquartile, 68% or 95%) obtained in this way can be shown to have any archaeo-
logical meaning: this has simply been assumed. This presentation will evaluate what might be the
meaning of such sums of probability distributions and whether they provide meaningful archaeolog-
ical information.
Two approaches are explored: (1) an empirical exploration (via simulation) of the summed distribu-
tions produced in various simple chronological scenarios, (2) inferences from statistical theory. The
outcomes of these explorations show that while for a large set of dates a summed probability distri-
bution may provide a useful visual summary of the distribution through time, it does not provide any
quantified summary with a simple archaeological meaning. Bayesian phase models, on the other
hand, work very well to obtain the results that archaeologists have been trying to obtain using
summed probability distributions and are therefore recommended as the tool of choice for determin-
ing the beginning, end, and duration of a period from radiocarbon dates.

P-43

A SERIAL PROBLEM: DATING THE INTRODUCTION OF CEREALS TO SCOTLAND AND

NORTHERN ENGLAND
Seren Griffiths1 and Peter Marshall2
1. Cardiff University, Humanities Building, Cardiff, Wales, CF10 3EU, UK.
2. English Heritage, UK.

The timing of the introduction of cereals to Scotland and northern England, and the associated
changes in human relationships with the land, are of critical importance for our understanding of
Neolithic lifeways. Radiocarbon dates obtained by AMS dating of single cereal grains interpreted
within a Bayesian framework allow robust estimates for the first use of cereal to be estimated. Such
an approach contrasts with the traditional “visual inspection” of radiocarbon dates (e.g. Brown
2007). Visual inspection fails to deal with statistical scatter, or the effects of the shape of the calibra-
tion curve, and can lead to inaccurate interpretation of the radiocarbon evidence.
The results presented here utilize a growing corpus of radiocarbon measurements obtained from
direct dating of Neolithic cereals during developer funded archaeological work in northern Britain.
We explore the effects that different “prior information” about the adoption of agriculture (Zvelebil
1986) has on the estimates for when the use of cereals first occurred. Finally, we try to place our
 Abstracts 97

results within a the growing body of research exploring the strands of change that took place in Brit-
ain in the 4th millennium cal BC.

REFERENCES
Brown A. 2007. Dating the onset of cereal cultivation in Britain and Ireland: the evidence from charred cereal grains. Antiquity
81:1042–52.
Zvelebil M. 1986. Mesolithic prelude and Neolithic revolution. In: Zvelebil M (ed.). Hunters in Transition: Mesolithic soci-
eties of Temperated Eurasia and Their Transition to Farming. Cambridge: Cambridge University Press.

P-44

LAKE SUIGETSU – 2006: A NOVEL APPROACH TO VARVE COUNTING; COMBINING THIN

SECTION MICROSCOPY AND MICROXRF AND RADIOGRAPHY
Gordon Schlolaut1, Michael Marshall2, Achim Brauer1, Takeshi Nakagawa3, Christopher Bronk
Ramsey4, Henry Lamb2, Yusuke Yokoyama5, Suigetsu 2006 Project Members
1. German Research Center for Geoscience (GFZ), Section 5.2: Climate Dynamics and Landscape Evolution, Telegrafen-
berg, D-14473 Potsdam, Germany.
2. Institute of Geography and Earth Sciences, Aberystwyth University, SY23 3DB, UK.
3. Department of Geography, University of Newcastle upon Tyne, Newcastle upon Tyne, United Kingdom.
4. Research Laboratory for Archaeology & the History of Art, University of Oxford, UK.
5. Department of Earth and Planetary Sciences, University of Tokyo, Japan.

A sediment core taken previously from Lake Suigetsu from one bore hole remains one of the most
comprehensive terrestrial radiocarbon records (Kitagawa and van der Plicht 2000), but the sediment
profile was not entirely continuous. In 2006, four parallel cores were recovered from Lake Suigetsu
and combined to a new master profile (SG06) ensuring complete recovery of the entire 73.5-m
sequence. As well as being extremely rich in terrestrial leaf fossils, providing a unique, truly atmo-
spheric, record of radiocarbon for the last 10–60 kyr BP, the Lake Suigetsu sediment is also annually
laminated for much of its depth. This project aims to contribute to the international terrestrial radio-
carbon calibration model, extending it to 50,000 years using the new SG06 Lake Suigetsu sediment
record.
Along with the new program of AMS radiocarbon measurement (Staff et al., this conference), varve
counting is being carried out using two different methods: thin section microscopy at the German
Research Center for Geoscience (GFZ), Potsdam and high-resolution X-ray fluorescence and radi-
ography at Aberystwyth University.
As part of this new approach, the two different counting methods are initially carried out completely
independently. A detailed comparison is then made of count differences produced by the two meth-
ods, beginning at meter scale and eventually down to the sub-mm scale. Differences are accounted
for and a varve count synthesis, using results from both methods, is established for difficult sections
of the profile. This substantially reduces internal error resulting in a greater degree of accuracy than
previously possible. Counting is being carried out on overlapping double-L core sections (Naka-
gawa 2007) from the composite core as well as from parallel cores, in order to bridge problematic
intervals in the composite cores. This allows us to further quantify and reduce internal error.
The two methods also provide complementary data on varve composition and micro-facies changes,
therefore providing crucial information about past processes in the lake and its catchment. This
98 Abstracts

includes the identification of detrital event layers (e.g. turbidites, tephras) with both methods, allow-
ing their precise removal prior to age-depth model reconstruction. This poster presents some prelim-
inary results utilizing this novel duel-method approach to varve counting from the Late Glacial-
Interglacial transition. The counts from both methods display a high degree of similarity depending
on the degree of varve preservation, and are in general good agreement with the 14C dates calibrated
using IntCal04.

REFERENCES
Kitagawa H, van der Plicht J. 2000. Atmospheric radiocarbon calibration beyond 11,900 cal BP from Lake Suigetsu laminated
sediments. Radiocarbon 42(3).
Nakagawa T. 2007. Double-L channel: an amazingly non-destructive method of continuous sub-sampling from sediment
cores. Quaternary International 167–168:298.

P-45

HIGH-RESOLUTION INTER-BASIN CHRONOLOGY OF HOLOCENE PALEOSEISMIC EVENTS

AT THE DEAD SEA BASIN BY A BAYESIAN RADIOCARBON DEPOSITION MODEL
E Kagan1,2, M Stein2, F Neumann3,4, A Agnon1, C Bronk Ramsey5
1. Institute of Earth Sciences, Hebrew University of Jerusalem, Givat Ram, Jerusalem, 91904, Israel.
2. Geological Survey of Israel, 30 Malkhe Israel, Jerusalem, 95501, Israel.
3. Bernard Price Institute for Paleontology, University of Witwatersrand, Johannesburg, South Africa.
4. Steinmann Institute for Geology, Mineralogy, Palaeontology, University of Bonn, Germany.
5. Research Laboratory for Archaeology & the History of Art, University of Oxford, UK.

High-resolution, continuous, and precise chronology is required for conducting inter-basin compar-
ison of paleoseismic sedimentary records. We applied the Bayesian statistical method of the OxCal
v4 program (Bronk Ramsey 2008) to AMS radiocarbon ages from samples collected at two late
Holocene Dead Sea sections (the Ein Feshkha Nature Reserve and the Ze’elim gully, located in the
northwestern and southwestern corners of the northern Dead Sea basin, respectively). In both sec-
tions, a large number of breccia layers (interpreted as seismites) were identified and dated by 14C
analyses of terrestrial organic debris. For the near-shore Ze’elim record, 6 14C ages were obtained,
lying in the range of ~1000 BC to AD 1500, allowing the determination of 10 seismite ages using
an age-depth model. Historical earthquakes were assigned to each of the seismites and fit within
their 1-σ age range. At Ein Feshkha, 15 14C ages were obtained, lying in the range of ~900 BC to
~AD 1300. We constructed an OxCal P_sequence age-depth model that incorporates 6 “anchor
points,” seismites that were correlated to prominent and well-documented historical earthquakes.
Fifty-two seismite ages from Ein Feshkha were interpolated from the OxCal models based on their
depths. Historical earthquakes were then assigned to each of the seismites and fit within their 1- or
2-σ age ranges. The resulting inter-basin paleoseismite chronology reveals the timing of seismites
that “pass through” the entire Dead Sea basin and the timing of seismic events that were recorded in
the northern or southern parts of the basin only. Our focus now is on the characterization of signifi-
cant individual earthquakes that are expressed by seismites in the entire basin.
 Abstracts 99

P-46

A FIRST TELL NORTH OF THE ALPS? RESULTS FROM A MULTILAYERED SETTLEMENT

MOUND AT NIEDERRÖBLINGEN (GERMANY)
C Lubos1, S Dreibrodt1, S Friederich2, M-J Nadeau3, P M Grootes3, O Nelle1, M Fuchs4,
H Meller2, H-R Bork1
1. Ecology-Centre, University of Kiel, Germany.
2. State Office for Archaeology, Halle (S), Germany.
3. Leibniz-Laboratory for Radiometric Dating and Stable Isotope Research, University of Kiel, Germany.
4. Department of Geomorphology, University of Bayreuth, Germany.

What appears to be the first multilayered settlement mount north of the Alps was discovered at
Niederröblingen (Sachsen-Anhalt, Germany). Radiocarbon results and archaeological and geoar-
chaeological methods indicate that it was occupied for a period of almost 6000 years with at least
one occupation gap.
The site is located in a small alluvial plain in the southeastern foreland of the Harz Mountains, ca.
80 km west of the city of Halle. The almost radial mound structure has a diameter of ~500 m and a
sequence of at least 13 distinguishable settlement layers. Its maximum thickness reached about 2 m.
The chronology is based on the regional pottery typology, 122 radiocarbon ages, and 2 OSL ages. A
settlement history of almost 6000 years starting in the Early Neolithic (LBK) and ending in the pre-
Roman Iron age is illustrated by the findings. A settlement intermission from about 4600–3000 BC
is reflected by a lack of mid to late Neolithic finds and the creation of a humic horizon above the first
Neolithic settlement layers.
We will discuss the settlement chronology based on a sequence Bayesian model. Fluxes of nutrient
and other elements between the settlement and its surrounding landscape, observed within the dated
settlement layers and supplemented by information from archaeological (artifacts, bones) as well as
geoarchaeological (geophysical, geochemical, mineralogical, paleoecological data) investigations
will also be presented.
Due to our results, we propose the use of the term “tell,” which is neither requested by the geograph-
ical distribution of a respective site nor by the dimensions (height) of a settlement structure, but
rather by its genesis.

P-47

APPLICATION OF BAYESIAN RADIOCARBON DATA ANALYSIS FOR THE CHRONOSTRATI-

GRAPHIC SEQUENCE OF MADONNA CAVE (CALABRIA, SOUTHERN ITALY)
L Calcagnile1, V Tinè2, G Quarta1, M D’Elia1, G Fiorentino3, F Scarciglia4, G Robustelli4,
M Abate4, M F La Russa5, A Pezzino5
1. CEDAD (Centre for Dating and Diagnostics), Dept. of Engineering of Innovation, University of Salento, Lecce, Italy.
2. Soprintendenza Speciale al Museo Nazionale Preistorico Etnografico “L. Pigorini,” Rome, Italy.
3. Laboratory of Archeobotany and Palaeoecology, University of Salento, Lecce, Italy.
4. Dipartimento di Scienze della Terra, Università della Calabria, Arcavacata di Rende (Cs), Italy.
5. Dipartimento di Scienza Geologiche, Università di Catania, Italy.

The Santuario della Madonna Cave (Praia a Mare, Cosenza, Italy), along the northwestern coast of
Calabria (southern Italy), represents part of a karstic system, developed in Meso-Cenozoic dolo-
100 Abstracts

mite-marly limestones and reshaped by the sea during the Quaternary. The cave has been the object
of systematic archaeological excavations by the University of Rome since the 1960s and revealed an
impressive stratigraphy with occupation phases spanning from the late Paleolithic to the advanced
phases of the Middle Bronze Age.
Recently, a new digging area has been opened and short-lived vegetal remains, selected from each
of the cultural layers identified on the basis of associated diagnostic archaeological material, were
submitted for AMS radiocarbon dating. The aim of the study was to define an accurate chronology
of the different cultural phases (spanning from the Mesolithic to the Bronze Age) and to explore the
potentialities resulting from the application of advanced statistical tools for the analysis of radiocar-
bon data in such a context. The selection of short-lived samples and the application of Bayesian-
based data analysis resulted in the possibility to obtain a highly accurate absolute chronological
framework for the different cultural horizons, supplying an important contribution to the problem of
the absolute chronology for the Italian prehistory. The possibility to use stable carbon isotopes for
the reconstruction of past climatic fluctuations has been also explored and preliminary results will
be presented.

P-48

THE PRODUCTION OF AN AMS 14C SUCROSE STANDARD FOR THE CALIBRATION AND
QUALITY CONTROL OF HIGH-ACTIVITY MEASUREMENTS
Marie Sydoff1,2 and Kristina Stenström2
1. Dept. of Clinical Sciences, Medical Radiation Physics, Malmo University Hospital, SE-205 02 Malmö, Sweden.
2. Lund University, Dept. of Physics, Division of Nuclear Physics, PO Box 118, SE-221 00 Lund, Sweden.

A number of standards with different properties are used for 14C measurements with AMS. They are
used for calibration and quality control of the AMS system. When analyzing biological samples
from e.g. microdosing studies, the activity content can be up to a 100 times the IAEA-C6 standard.
IAEA has eight different reference materials (IAEA-C1 – C8) of which the IAEA-C6 sucrose stan-
dard is the most active standard on the market with a 14C value of 150.61 pMC. Thus, there is a need
for a standard material with higher activity content than IAEA-C6 for making accurate AMS mea-
surements of this type of samples. This poster describes the procedure of making a secondary stan-
dard with an activity content of about 7 times the IAEA-C6 standard, i.e. 1025 pMC. The material
chosen has to be chemically inert and the activity must be homogeneously divided in the whole
material. The material chosen for the new standard is sucrose, since it is a non-toxic, inexpensive
and organic material, which after combustion forms only carbon dioxide (CO2) and water (H2O). It
is also is a well-known standard material.
 Abstracts 101

P-49

MODERN RADIOCARBON LEVELS FOR NORTHEASTERN MEXICO DERIVED FROM TREE

RINGS – A COMPARISON WITH NORTHERN HEMISPHERE ZONES 2 AND 3 CURVES
Laura E Beramendi-Orosco1, Galia Gonzalez-Hernandez2, Jose Villanueva-Diaz3, Francisco J
Santos-Arevalo4, Isabel Gomez-Martinez4, Edith Cienfuegos1, Pedro Morales1, Jaime Urrutia-
Fucugauchi2
1. Instituto de Geologia, Universidad Nacional Autonoma de Mexico, Ciudad Universitaria, Mexico DF 04510, Mexico.
2. Instituto de Geofisica, Universidad Nacional Autonoma de Mexico, Ciudad Universitaria, Mexico DF 04510, Mexico.
3. Instituto Nacional de Investigaciones Forestales y Agropecuarias, CENID-RASPA, Gomez Palacio, Durango 35140,
Mexico.
4. Centro Nacional de Aceleradores, Seville, 41092, Spain.

Radiocarbon levels for the period 1940–2004 have been obtained by AMS and LSC analysis of tree
rings in order to study the radiocarbon variation for northeastern Mexico, and make comparisons
with the Northern Hemisphere zones 2 and 3. Two tree-ring sequences of Pseudotsuga menziesii,
sampled in a site isolated from urban or industrial centers and active volcanoes (26.18ºN, 106.3ºW,
2900 m asl), were dendrochronologically dated through a master tree-ring chronology. Latewood
samples of annual rings for the period 1955–1965 and years 1950, 1970, 1975, 1980, 1990, 2000 and
2004 were analyzed by AMS at CNA. Samples analyzed by LSC at UNAM corresponded to whole
wood of 4 consecutive rings, mixing both trees, in order to obtain a sample size suitable for analysis.
Results obtained show a similar profile to that reported for the Northern Hemisphere, with signifi-
cant correlation coefficients r = 0.989 and r = 0.994, for zone 2 and 3, respectively. The maximum
peak is centered at 1964.5 with a F14C value of 1.7252 ± 0.009‰, a pre-bomb F14C value of
0.9839 ± 0.003‰ for 1950, and a F14C value of 1.0791 ± 0.005‰ for 2004. The values obtained for
the period 1955–1970 are lower than zone 2 values by an average of 30.4‰, and higher than zone 3
by an average of 25.9‰. For the period 1970–2004, the values obtained are higher than the NH
curve, with a mean difference of 23‰, suggesting that the fossil fuel CO2 emissions for the sampling
site are lower than for the sites compiled in the curve. The fact that the F14C values found in this
study are closer to zone 3 values is relevant as Mexico is located in zone 2 as proposed by Hua and
Barbetti (2004) in the compilation of post-bomb radiocarbon data.

P-50

14 C
WIGGLE-MATCHING OF THE HARUNA FUTATSUDAKE PUMICE (HR-FP), CENTRAL
JAPAN
M Okuno1, T Oikawa2, Y Hoshino3, S Yatsuzuka1, N Geshi2, T Nakamura4
1. Department of Earth System Science, Faculty of Science, Fukuoka University, Fukuoka 814-0180, Japan.
2. AIST, Geological Survey of Japan, Tsukuba305-8567, Japan.
3. Botanical Gardens, Tohoku University, Sendai 980-8576, Japan.
4. Center for Chronological Research, Nagoya University, Nagoya 464-8602, Japan.

The Haruna Volcano located in the central part of Japan erupted twice in the 6th century (Arai 1962).
The first eruption occurred in the early part of the century and produced a tephra sequence named
the Haruna Futatsudake Ash (Hr-FA). The Haruna Futatsudake Pumice (Hr-FP) is a product of the
second eruption that occurred in the middle of the century. Both tephras provide a chronological
102 Abstracts

framework to archaeological sites in the Kofun period. In this period, there are no historical records
of eruptions in Japan. However, these ages of eruptions were previously determined by pottery chro-
nology (Sakaguchi 1986). On the other hand, Nakamura et al. (2008) obtained calendar age of cal
AD 485–504 (2 σ) for the Hr-FA eruption, using 14C wiggle-matching of three charred wood trunks.
In order to determine the age of the Hr-FP eruption by a HVEE Tandetron AMS system at Nagoya
University (Nakamura et al. 2000), we collected a charred wood sample from a trunk within a pyro-
clastic flow deposit on the eastern flank of the volcano. The obtained 14C dates are calibrated to cal-
endar dates with computer program CALIB 5.0 (Stuiver and Reimer 1993). We present the results
of precise age determination for the Hr-FP eruption by 14C wiggle-matching of the charred trunk.

P-51

14C WIGGLE-MATCHING OF THE 10TH CENTURY ERUPTION OF BAITOUSHAN VOLCANO

(CHINA/NORTH KOREA)
S Yatsuzuka1, M Okuno1, T Nakamura2, K Kimura3, Y Setoma4, T Miyamoto5, K H Kim6,
H Moriwaki7, T Nagase8, X Jin9, B L Jin10, T Takahashi11, H Taniguchi5
1. Department of Earth System Science, Faculty of Science, Fukuoka University, Fukuoka 814-0180, Japan.
2. Center for Chronological Research, Nagoya University, Nagoya 464-8602, Japan.
3. Faculty of Symbiotic System Science, Fukushima University, Fukushima 960-1296, Japan.
4. The Kitakyushu Chamber of Commerce and Industry, Kitakyushu 802-8522, Japan.
5. Center for Northeast Asian Studies, Tohoku University, Sendai 980-8576, Japan.
6. Department of Earth Science Education, Ewha Womans University, Seoul 120-750, South Korea.
7. Section of Physical Geography, Faculty of Law, Economics and Humanities, Kagoshima University, Kagoshima 890-
0065, Japan.
8. The Tohoku University Museum, Tohoku University, Sendai 980-8578, Japan.
9. College of Geo-exploration Science and Technology, Jilin University, Changchun 130-026, China.
10. Qiaonguang Science and Technology Research Institute Sichuan Province, 133001, China.
11. Wood works “Yui”, Iwate 028-7301, Japan.

The Baitoushan Volcano located on the border between China and North Korea, erupted in the 10th
century. The ash-fall generated during the eruption (B-Tm) reached northeastern Japan (Machida et
al. 1981). Machida (1992) proposed that the collapse of the Bo-hai Kingdom (AD 926), which occu-
pied the areas around the volcano at that time, might have been caused by the eruption. However, the
age of the eruption is still highly debated. The eruption age was previously reported as AD 938 by
non-glacial varve chronology (Fukusawa et al. 1998) and 935+8/–5 cal AD by 14C wiggle-matching
(Nakamura et al. 2007). This paper presents the results of precise age determination for the 10th cen-
tury eruption by 14C wiggle-matching using multi-samples with 10- or 2-ring intervals of charred
wood trunks. Two charred wood trunks, Samples A and B collected from pyroclastic flow deposit on
the northeast flank of the Baitoushan, were used to determine the age of eruption by a HVEE Tan-
detron AMS system at Nagoya University (Nakamura et al. 2000). The obtained 14C dates were cal-
ibrated to calendar dates with computer program CALIB 5.0 (Stuiver and Reimer 1993). Our wig-
gle-matching provides a confident eruption age of cal AD 953 for Sample A and cal AD 942 for
Sample B.
 Abstracts 103

P-52

ANATOLIAN TREE RINGS FOR THE EARLY IRON AGE FROM THE KAMAN-KALEHÖYÜK
SITE, TURKEY: RECONSIDERATION OF ANATOLIAN REGIONAL 14C OFFSETS

Takayuki Omori1 and Toshio Nakamura2

1. Graduate School of Environmental Studies, Nagoya University, Japan.
2. Center for Chronological Research, Nagoya University, Japan.

In the Anatolian area, regional 14C offsets have been discussed by Kromer et al.(2001), Manning et
al. (2001), and Kuniholm et al. (1996). Kromer et al. (2001) suggested that Anatolian short-term off-
sets exist from 800 to 750 BC and are caused by the colder and wetter climate associated with solar
activity, based on the Anatolian 14C data set constructed by the Aegean Dendrochronology Project.
An Early Iron Age (typically 11th to 8th century BC) covered the term during which many archae-
ological events occurred, for example, the end of Phrygian Kingdom, but little archaeological infor-
mation exists. Although we need to build a high-resolution chronology for the Early Iron Age using
14C dating, the question whether we can apply the Anatolian 14C data set to age calibration or not

remains unanswered. This is because the 14C offsets have been discussed with insufficient data for
the 9th–8th century BC.
We present a developed Anatolian 14C data set including additional tree-ring samples. Tree trunks
having about 100 annual rings were collected from the Early Iron Age stratum at the Kaman-Kale-
höyük site, Turkey, and were dendrochronologically dated to the 10th to 8th century BC. AMS 14C
measurements have been carried out for the quinquennial or decadal tree rings. We discuss the Ana-
tolian regional 14C offsets in comparison with our results and global 14C changes.

P-53

14 C
AGE MEASUREMENTS OF SINGLE-YEAR TREE RINGS FOR TWO OLD WOODS AT
ABOUT 22,000 AND ABOUT 46,000 14C YR BP
Hirohisa Sakurai1, Taiichi Sato2, Kayo Suzuki2, Yui Takahishi2, Hiroyuki Matsuzaki3,
Yoko Sunohara3
1. Dept. of Physics, Faculty of Science, Yamagata University, 1-4-12 Kojirakawa-machi, Yamagata City, Yamagata 990-
8560, Japan.
2. Graduate School of Science and Engineering, Yamagata University, 1-4-12 Kojirakawa-machi, Yamagata City, Yamagata
990-8560, Japan.
3. Dept. of Nuclear Engineering and Management, School of Engineering, University of Tokyo, 2-11-16 Yayoi, Bunkyo-ku,
Tokyo 113-0032, Japan.
14C concentrations at 25,000 yr BP were much higher, approximately 400‰, than at present, and

they are rapidly decreased to 0‰ from 40,000 to 45,000 yr BP, indicating a long-term variation in
the geomagnetic field (Chiu et al. 2007). However, it is important to know the variations in cosmic
rays for these eras, because 14C in the atmosphere is produced by cosmic rays coming to Earth. In
particular, the 14C concentrations of single-yr tree rings indicate solar modulations of cosmic rays,
such as the 11-yr solar cycle, and hence old tree rings are a powerful tool to detect the past variations
in cosmic rays.
104 Abstracts

In order to investigate characteristics of single-year measurements for very old wood, we have mea-
sured the AMS 14C ages of tree rings for two old wood samples, samples of Kaminoyama and Zao
that were dug out from Yamagata City, Japan (39°N). For the Kaminoyama sample, three sequential
measurements were carried out for 33 consecutive single-year tree rings for one sequence. For the
Zao sample, a sequential measurement was carried out for 6 consecutive single-year tree rings.
The average 14C age of the Kaminoya sample was 22,237 ± 140 14C yr BP and the fluctuation was
within ~400 14C yr. The 14C age of single-year tree rings for the Zao sample was 46,375 14C yr BP
with an error of 600 14C yr. We will describe the accuracy of 14C age measurements for single-year
tree rings of very old wood samples with different ages.

P-54

BOMB 14C DATA PROFILES FROM 1950 TO 2000 IN JAPANESE TREE RINGS FROM THE
PACIFIC OCEAN SIDE AND THE JAPAN SEA SIDE
Hirohisa Sakurai1, Kayo Suzuki2, Yui Takahashi2, Taiichi Sato2
1. Dept. of Physics, Faculty of Science, Yamagata University, 1-4-12 Kojirakawa-machi, Yamagata City, Yamagata 990-
8560, Japan.
2. Graduate School of Science and Engineering, Yamagata University, 1-4-12 Kojirakawa-machi, Yamagata City, Yamagata
990-8560, Japan.

The profiles of atmospheric 14C concentrations due to atmospheric nuclear detonations were estab-
lished in both the observation of CO2 and the tree-ring record. Recently, Hua and Barbetti indicated
that the tree rings growing in summer at Japan Island are influenced by southern winds from the
Pacific Ocean, using atomic bomb data of 14C concentrations in the Northern Hemisphere and the
Southern Hemisphere (Journal of Geophysical Research 112: D19120). In Japan Island, the direc-
tions of the window are approximately opposite between the summer and the winter by the seasonal
positions of Intertropical Convergence Zone (ITCZ).
In order to check this effect, we have measured the 14C concentrations of single-year tree rings
between 1950 and 1998 for cedars located at three areas on Japan Island. They are Akita cedar
(39°N), Kinpou cedar (38°N), and Murou cedar (34°N). The Akita and Kinpou cedars were growing
on the Japan Sea side of the island and the Murou cedar on the Pacific Ocean side. For 1964, all three
cedars show a peak in 14C concentrations. Δ14C values were 728.2‰, 766.8‰, and 783.5‰ for the
Akita, Kinpou, and Murou cedars, respectively, implying little latitude effect for the difference
between the Pacific Ocean and the Japan Sea sides.
Moreover, exchange rates between the atmosphere and the surface ocean due to the carbon cycle are
estimated by applying a three-box model for the decreasing curve of the temporal profiles of the data
sets. We will compare our data sets and the published data sets of bomb profiles for the CO2 and the
tree rings, focusing on the influence of the summer season and the decreasing profiles after 1964.
 Abstracts 105

P-55

RADIOCARBON DATES OF JAPANESE TREE RINGS FOR 1060 BC–AD 400

Hiromasa Ozaki1,6, Minoru Sakamoto1, Mineo Imamura1, Hiroyuki Matsuzaki2,
Toshio Nakamura3, Koichi Kobayashi4, Shigeru Itoh4, Etsuko Niu4, Takumi Mitsutani5
1. National Museum of Japanese History, The National Institutes for Humanities, Japan.
2. Department of Nuclear Engineering and Management, The University of Tokyo, Japan.
3. Center for Chronological Research, Nagoya University, Japan.
4. AMS Dating Facility, Paleo Labo Co., Ltd., Japan.
5. National Research Institute for Cultural Properties, Nara, Japan.
6. Present address; AMS Dating Facility, Paleo Labo Co., Ltd., Japan.

Radiocarbon calibration has recently become one of the most important parts in radiocarbon dating
for archaeological applications. Universal calibration data sets like IntCal04 have been widely used
for the Northern Hemisphere assuming the absence of regional offsets. Since IntCal04 has been
based on data from European and North American wood, possible offsets from IntCal04 may result
in significant deviations from real dates when applied to Japanese archaeological samples. Thus, we
have studied radiocarbon dates of dendrodated Japanese tree rings.
AMS radiocarbon dates of Japanese tree rings are presented for every 5 years from 1060 BC to AD
400 (1460 years). The period ranges from the latest stage of the Jomon period, to the early stage of
the Kofun period in Japanese pre- and proto-history, respectively, with an intervening Yayoi period,
which is characterized by the start and spread of wet rice agriculture in the Japanese Archipelago.
Generally, the obtained dates agreed well with IntCal04. However, there are some periods when the
dates differ significantly from IntCal04. For example, from the 1st to 3rd century AD, systematical
differences up to more than 50 14C years from IntCal04 are observed in two independent Japanese
tree-ring samples. Such large deviations would lead to invalid calibrated dates by an order of 100
calendar years. Significant differences are also found in a short period at around 680 BC, but these
differences are not systematic, but rather randomly fluctuate in magnitude.
We will discuss the deviations in terms of regional effects and their possible consequences for the
calibration.

P-56

EXTENDING TREE-RING CHRONOLOGIES AND SETTLEMENT HISTORY OF HIGH ASIA BY

RADIOCARBON DATING OF WOOD SAMPLES FROM HISTORIC MONASTERIES AND TEM-
PLES IN TIBET AND NEPAL
Andreas Scharf1, Achim Bräuning2, Wolfgang Kretschmer1, Katrin Leichmann1,Stefanie Gierl1,
Iris Burchardt2, André Alexander3
1. Erlangen AMS Laboratory, University Erlangen-Nuremberg, Germany.
2. Institute for Geography, University Erlangen-Nuremberg, Germany.
3. Tibet Heritage Fund.

In the last 15 years, a network of 50 tree-ring chronologies of different species of living conifers
from the Tibetan Plateau and western Nepal was compiled, covering a range of 300 to 1500 years.
Now, in four regions of High Asia valuable wood samples from historic monasteries, temples, and
106 Abstracts

secular buildings could be collected. Some could be dendrodated, but many could not, indicating a
higher age than the existing local chronologies. In addition to the dating of wood samples from these
historic monuments, 14C AMS dating with wiggle-matching gives the opportunity to extend the
range of the regional tree-ring chronologies and to get highly resolved annual information about past
climatic variability in Tibet and the Himalayas. Wood samples of this project from Dolpo (western
Nepal) and Lhasa (central Tibet) were radiocarbon dated at the Erlangen AMS Laboratory, using α-
cellulose extraction, which was recently established in Erlangen. The first AMS results indicate that
the existing chronologies can be actually extended.

P-57

IS THE CONSENSUS VALUE OF ANU SUCROSE (C-6) TOO HIGH?

Xiaomei Xu, Matthew S Khosh, Kevin C Druffel-Rodriguez, Susan E Trumbore, John R Southon
Keck AMS Facility, Earth System Science Department, University of California, Irvine, CA 92697, USA.

Primary and secondary standards are essential in radiocarbon analyses for the purpose of reporting
and comparing data among laboratories, as well as for internal laboratory data quality control. ANU
sucrose (C-6) is one of the IAEA certified radiocarbon standards with a consensus value of 150.61 ±
0.11 pMC (standard error, n = 22). Using OXI as our primary standard, we routinely use ANU as one
of our secondary standards, usually one sample in each 40-sample wheel. All of our measurements
of ANU sucrose (n = 351) over the last seven years result in an average value of 150.16 ± 0.46 pMC
(1σ standard deviation). After applying the same outlier tests and statistical analysis methods as
used for IAEA reference standards (Rozanski et al. 1992), the weighted average value is 150.16 ±
0.02 pMC (standard error, n = 294), which is significantly lower than the consensus value (T test
with unequal variance; p = 0.023). The results are similar without the removal of outliers. In con-
trast, our measurements of OXII over the same time period, which we use as an additional secondary
standard, result in an average value of 133.98 ± 0.41 pMC (n = 602, 1σ standard deviation, no out-
liers removed), and a weighted average value of 133.96 ± 0.01 pMC (n = 524, standard error, outli-
ers removed), both of which are consistent with the OXII consensus value of 134.07 ± 0.19 pMC.
Our measurements of other secondary standards such as TIRI B and FIRI J are also consistent with
the corresponding consensus values. Since our lab values for OXII and other secondary standards
agree with the consensus values, our comparatively larger number of ANU sucrose measurements
seem to imply that the consensus value of ANU sucrose might be too high by ~0.34 pMC, or that the
ANU sucrose standard is not homogeneous. Given that some labs routinely use ANU sucrose as a
primary standard, our results suggest that revisiting the consensus value of ANU sucrose may be
necessary.

REFERENCE
Rozanski K, Stichler W, Gonfiantini R, Scott EM, Beukens RP, Kromer B, van der Plicht J. 1992. The IAEA 14C Intercom-
parison Exercise. Radiocarbon 34(3):506–19.
 Abstracts 107

P-58

RADIOCARBON CALIBRATION FROM 22 TO 30 KYR BP USING 238U-232Th DATING OF A STA-

LAGMITE FROM CAVE 2-9, SAN LA PROVINCE, VIETNAM
Susana E Gonzalez1, J Warren Beck1, Chuan-Chou Shen2, Nguyen Chi Dung2, Yue-Gau Chen2,
Lin Yin2, Hong-Wei Chiang2, Doan Dinh Lam3
1. NSF-Arizona AMS Laboratory, University of Arizona, Tucson, AZ, USA.
2. Department of Geosciences, National Taiwan University, Taiwan, R.O.C.
3. Institute of Geological Sciences, Vietnamese Academy of Science and Technology, Vietnam.

In an effort towards extending the radiocarbon calibration curve beyond 26 kyr BP, we have made
combined AMS 14C and ICPMS 238U-232Th measurements on a 50.8-cm-length stalagmite recov-
ered from cave 2.9 in San La Province, northwestern Vietnam. 230Th dating was used to derive a cal-
endar age/depth model using a piece-wise polynomial spine technique. U-series measurements were
made using a Finnigan-MAT Neptune multi-collector inductively coupled plasma mass spectrome-
ter (MC-ICP-MS) with average 2σ errors ~3.4‰ located at the Department of Geosciences,
National Taiwan University, Taipei. These results indicate this stalagmite grew between 31.2 and
22.6 kyr BP. AMS 14C measurements were made at the NSF-Arizona AMS facility at the University
of Arizona. Samples prepared for AMS 14C measurement were handled with new extraction proce-
dures designed to minimize processing blank. The dead carbon fraction (DCF) was determined by
comparing speleothem results with IntCal04 for the period 22.6–26 cal kyr BP. A comparison
between the speleothem record and other existing records will be discussed.

P-59

WIGGLE-MATCH DATING OF WOODEN SAMPLES FROM IRON AGE SITES IN NORTHERN

ITALY
G Quarta1, M I Pezzo2, S Marconi2, U Tecchiati3, M D’Elia1, L Calcagnile1
1. CEDAD (Centre for Dating and Diagnostics), Dept. of Engineering of Innovation, University of Salento, Lecce, Italy.
2. Museo Civico di Rovereto, Dendrochronoly Laboratory, Rovereto, Italy.
3. Soprintendenza Provinciale ai Beni Culturali di Bolzano, Ufficio Beni Archeologici, Bolzano, Italy.

Archaeological excavations carried out in the sites of Lajen (Bolzano, northern Italy) and Stufles-
Oberegger (Bressanone, northern Italy) brought to light very well preserved wooden samples in cul-
tural layers archaeologically dated to the Iron Age. From the two sites, different wooden samples
were recovered in a preservation status that allowed the clear identification of the tree species and
of the ring structure.
In particular among the different wooden samples, two were selected for dendrochronology and
radiocarbon analyses: from Lajen a beam with an unbroken sequence of 158 rings and from Stufles-
Oberegger a combusted trunk with a sequence of 217 rings. Both samples were identified as belong-
ing to the Larix Decidua species. From each sequence, different single-ring samples were selected
and submitted for AMS radiocarbon dating analysis at CEDAD. Conventional radiocarbon ages
were then calibrated to calendar ages by using the IntCal04 atmospheric data set, while the statistical
constraints resulting from the defined ring sequence were used to develop a “wiggle-matching”
approach by making use of the Bayesian analysis routines available in OxCal. The applied wiggle-
108 Abstracts

matching approach resulted in a significant reduction of the statistical uncertainties associated with
the calibrated time ranges and allowed to supply an important contribution for the chronological
definition of the different occupation phases of the studied sites.
A detailed comparison between dendrochronology and radiocarbon results will be also presented.

P-60

CHRONOSTRATIGRAPHIC SUBDIVISION OF THE LATE GLACIAL AND THE HOLOCENE IN

ALASKA
Danuta J MichczyÒska1 and Irka Hajdas2
1. GADAM Centre of Excellence, Institute of Physics, Silesian University of Technology, Gliwice, Poland.
2. ETH Zurich, Laboratory for Ion Beam Physics, Radiocarbon Laboratory, 8093 Zurich, Switzerland.

The first step of our “data mining” work was collection of the radiocarbon data for samples from the
Alaska region. We constructed database using radiocarbon date lists published by different radiocar-
bon laboratories (mainly in the journal Radiocarbon). The next step was careful analysis of col-
lected dates and excluding dates suspected of contamination by younger or older organic matter.
Such a fact could be stated, for instance, on the base of the inconsistency of radiocarbon age and
stratigraphy or palynology. Finally, we calibrated all chosen dates and constructed probability den-
sity functions (PDFs) for whole set, and also for subsets created for different climatic regions.
Analysis of the shape of PDF was the subject of previous studies (MichczyÒska et al. 2007). In our
analysis we take into account the distinct tendency to collect samples from selected horizons. It is a
general rule to take samples for radiocarbon dating from places of visible sedimentation changes or
changes in palynological diagram. Therefore, the culminations of the PDF represent periods of envi-
ronmental changes and could be helpful in identifying the chronostratigraphic boundaries on the cal-
endar timescale.

REFERENCE
MichczyÒska DJ, MichczyÒski A., Pazdur A. 2007. Frequency distribution of radiocarbon dates as a tool for reconstructing
environmental changes. Radiocarbon 49(2):799–806.

P-61

COSMOGENIC ISOTOPES, LONG-TERM SOLAR ACTIVITY, AND 20TH CENTURY GLOBAL

WARMING
V A Dergachev1, O M Raspopov2, G A Zaitseva3
1. Ioffe Physico-Technical Institute of RAS, St. Petersburg, Russia.
2. St.Petersburg Filial (Branch) of Pushkov Institute of Terrestrial Magnetism, Ionosphere, and Radiowave Propagation of
RAS, St. Petersburg, Russia.
3. The Institute for the History of Material Culture RAS, St. Petersburg, Russia.

Identification of the causes of past climate change requires detailed knowledge of one of the most
important natural factors—solar influence on climate. The Sun has an obvious influence on climate
 Abstracts 109

since its radiation is the main energy source for upper Earth’s atmosphere. Information on solar
activity during the last millennium is provided by data from direct and historical solar observations
as well as from the best-known proxy for changes in solar activity, the cosmogenic isotopes 14C and
10Be. All data show that solar activity has varied rhythmically with the main periods of approxi-

mately 11(22) yr, 90 yr, and 200 yr during the last 1000 years. This means that solar fluctuations
were involved in causing widespread climatic changes, like the Little Ice Age (AD 1450–1850) that
followed the Medieval Warm Period (AD 900–-1400). Direct solar satellite observations of solar
irradiation during the last 28 years show that solar output is in a phase in sunspot number changes.
From sunspot and cosmogenic isotope data, it follows that solar activity in this period was relatively
high. It is necessary to take into account both periods of high and low solar variability. Some studies
clearly suggest that solar irradiation has varied on timescales longer than the well-known 11-year
solar cycle.
The results of analysis of patterns of climatic characteristics and natural influences for the past cen-
turies are presented. It is expected that a recent warming could emerge from the superimposed influ-
ence of the 90- and 210-yr solar cycles on the climate system, which is characterized by nonlinear
dynamics. Any human-induced effect on climate will be superimposed on the background “noise”
of natural climate variability, which results both from internal fluctuations and from external causes.

FRESHWATER AND GROUNDWATER

P-62

SPATIAL DISTRIBUTION OF Δ14C VALUES OF SURFACE SEDIMENTS OFF SARU RIVER IN

HOKKAIDO, JAPAN
S Nagao1, T Irino2, T Aramaki3, K Ikehara4, M Uchida3, Y Shibata3
1. LLRL, KU-INET, Kanazawa University, Japan.
2. Faculty of Environmental Earth Science, Hokkaido University, Japan.
3. AIST, Geological Survey of Japan, Japan.
4. National Institute for Environmental Studies, Japan.

A variety of geochemical approaches have been employed to define the mixing ratio of marine and
terrestrial organic matter. The simultaneous use of Δ14C and δ13C values adds a second dimension
to isotopic studies of carbon cycling in surface aquatic environments. This study discusses the effects
of flooding on the spatial distribution of organic matter on the shelf region on the basis of the Δ14C
and δ13C values. Transport of organic matter due to flooding is important for the dynamics and flux
of organic matter released from terrestrial environment. The surface sediments off the Saru River,
southern part of Hokkaido, Japan, were collected from 20 sites with a grab sampler in August 2007,
one year after the severe flood in 2006. Suspended particles of the Saru River were concentrated with
a single-bowl continuous-flow centrifuge in August 2007 and 2008. 14C measurements were per-
formed by accelerator mass spectrometry at the National Institute for Environmental Studies.
The Δ14C values of organic matter in surface sediments (water depth of 10–39 m) off the Saru River
range from –665 to –77‰. The silt and clay sediments have Δ14C values of –240 to –77‰, but the
sandy sediments ranges from –665 to –388‰. The Δ14C values of particulate organic carbon in the
110 Abstracts

Saru River are –292 to –247‰ at normal flow conditions. The sediments with fine particles depos-
ited by floods have organic carbon with a mixture of young and older organic carbon. Therefore,
these results indicate that surface soil with younger organic matter is deposited by flooding on the
sediment off the Saru River and/or freshly produced organic matter in sea surface is deposited after
the flooding.

P-63

APPARENT RADIOCARBON AGE IN LAKE SUGAN, NORTHWEST CHINA

Ai-feng Zhou1, Zong-li Wang1, Fa-hu Chen1, Mei-lin Yang2, Ming-rui Qiang1, Jia-wu Zhang1
1. CAEP, Key Laboratory of West China’s Environmental System (Ministry of Education), Lanzhou University, Lanzhou
730000, P.R China.
2. Appraisal Center of Environmental Engineering of Yunnan Province, Kunming 650032, P.R. China.

Reliable chronologies are vital for most paleolimnological studies. However, due to the dead carbon
effect (radiocarbon reservoir effect) existing in many lakes, a convincing chronological framework
is difficult to achieve. Furthermore, the dead carbon effect often leads to wrong climatic interpreta-
tions in paleoenvironmental research. Especially in the arid northwest of China, saline lake water
often suffers from hardwater effects. To eliminate this effect, many efforts were made to determine
the exact reservoir effect, such as by comparing the radiocarbon chronology with other chronologies
established by OSL, U-series, etc. Usually, however, different chronological tools also take in the
uncertainties. As a consequence, it is often difficult to place the high-resolution depositional records
within a chronological framework of comparable quality, making it impossible to resolve the envi-
ronmental questions presented by this period.
In this paper, the original study on the determination of apparent 14C age for the Lake Sugan is pre-
sented. Aquatic seeds, bulk organics, terrestrial plants, and carbonate samples were collected in
2000s, and the mean apparent 14C age of carbonate in Lake Sugan is ca. 1150 yr. This values may be
used to correct the former 14C data obtained from this region and also will be of benefit to future
research.

P-64

14CDATING OF BURIED TREES FROM THE LANDSLIDE-DEPOSIT-FORMED, LANDSLIDE-

DAMMED LAKE AT TOYAMA RIVER, NAGANO PREFECTURE, CENTRAL JAPAN
Akiko S Goto1, Takeshi Muramatsu2, Yoshiji Teraoka3
1. Center for Chronological Research, Nagoya University, Japan.
2. Iida City Museum, Japan.
3. Natural Association of Ina Valley, Japan.

Huge landslides, triggered by strong earthquakes or torrential rains, often yield natural dams and
lakes, and bring serious damage to human life. Landslide-dammed lakes are frequently formed
when active faults cause big earthquakes in Japan. Therefore, in this research, we tried to get infor-
 Abstracts 111

mation about ancient earthquakes from the survey of landslide-dammed lakes, especially the dam
parts, and their timing. Clarifying the timing of landslides is difficult work because we have to esti-
mate it by wood samples were related to the landslide: the samples indicate various tree types and
conditions. We inferred that the 14C dating method is suitable for clarifying the timing by dating bur-
ied trees in such situations, though the method cannot decide the exact event time.
Evidence of a landslide-dammed lake was found on the Toyama River, central Japan. The timing of
the formation was estimated at AD 714 in previous research by dendrochronological dating of a bur-
ied Japanese cypress. The timing of the lake formation showed the same time as the occurrence of
huge earthquake in ancient documents. In this research, we identified avalanche deposits containing
some wood fragments by geological survey, and presumed that their relation with the landslide dam
formed in AD 714. To determine when the landslide dam was formed, the branches and wood frag-
ments from the avalanche deposit were analyzed by 14C dating, together with some stocks kept
standing on the stream side, which was covered by the landslide before its erosion. Periods of reli-
able samples showed a date around AD 714. The samples were twigs, a callus, and a stock with its
bark, probably indicating their time of death. We will mainly present 14C dating results of the sam-
ples from deposits, and propose the landslide-dammed lake’s situation, its size, and dam location,
triggered by an earthquake in AD 714.

P-65

AN INVESTIGATION INTO THE HARDWATER EFFECT AND ITS INFLUENCE ON RELIABLE

ESTUARINE RADIOCARBON DETERMINATIONS FROM AOTEA HARBOUR, NEW ZEALAND
Helen McKinnon1, Alan Hogg1, Fiona Petchey2, Thomas Higham2
1. Radiocarbon Dating Laboratory, School of Science and Engineering, University of Waikato, Private Bag 3105, Hamilton
3240, New Zealand.
2. Oxford Radiocarbon Accelerator Unit, Research Laboratory for Archaeology and the History of Art, Dyson Perrins Build-
ing, University of Oxford, Oxford OX1 3QY, UK.

The term “hardwater effect” refers to the addition of 14C-depleted bicarbonate from calcareous rock
formations such as limestone, siltstones, or mudstones in estuarine, lacustrine, and riverine envi-
rons. These hardwaters dilute the affected radiocarbon reservoir, and render radiocarbon dates of
shellfish artificially old.
A systematic investigation at Aotea, a harbor on the northwest coast of New Zealand with a calcar-
eous hinterland drained by various freshwater streams, was undertaken to determine if a hardwater
effect could be identified in shell from the estuary. Two methods were used in the analysis. First, 14C
determinations on modern shellfish (living at the time of collection) from within Aotea Harbour
were compared to results from modern shellfish from an ocean beach. A detectable hardwater effect
should result in a measurable difference in the radiocarbon results between these two areas. The sec-
ond method involved measuring the difference in age between stratigraphically paired samples of
marine shell and charcoal from archaeological sites located around the harbor to determine the offset
between the two.
No detectable hardwater effect was measured in either the modern or archaeological specimens.
This research indicates that three critical environmental factors are required to result in a measurable
hardwater effect: a source of 14C-depleted carbon, a significant freshwater input (indicated by low
112 Abstracts

salinity values for water in the estuary), and a long residence time of water to allow uptake of the
14C-depleted carbon.

P-66

SPATIAL VARIABILITY OF 14C, δ13C, AND δ18O IN GROUNDWATER OF SLOVAKIA

P P Povinec1, P K Aggarwal2, L J Araguás-Araguás2, R Breier1, L Lúèan1, A Šivo1,
M Richtáriková1
1. Comenius University, Faculty of Mathematics, Physics and Informatics, SK-84248 Bratislava, Slovakia.
2. International Atomic Energy Agency, Isotope Hydrology Laboratory, A-1400 Vienna, Austria.

With the development of geostatistical methods of data treatment based on the Geographical Infor-
mation System (GIS), it is possible to map the spatial variability of the isotopic composition of water
in selected regions of the world. In such a complex system, it is possible to trace the origin and path-
ways of different water masses on the basis of the developed isotopic maps, covering temporal and
spatial distribution of isotopes and hydrochemical data. Such an integrated attempt will gather new
information on temporal and spatial variability of groundwater, on its dynamics, on anthropogenic
and climatic impacts, and on its vulnerability. Investigations of spatial variability of 14C, δ13C, and
δ18O carried out till now on groundwater from Slovakia suggest large isotopic heterogeneity with
variable 14C content between 5 and 90 pMC. A vertical distribution of 14C in groundwater of the
éitný ostrov confirms that the majority of groundwater has 14C content below 20 pMC, except in a
small area in the northwest, where we see younger waters at the surface and subsurface (around
500 m) with 14C values around 30 pMC. An interesting feature is the existence of old waters with
14C values below 10 pMC, which are surprisingly seen in subsurface layers down to 200 m water

depths. This new research approach, carried out in the framework of IAEA’s Coordinated Research
Project “Geostatistical analysis of spatial isotope variability to map the sources of water for hydrol-
ogy studies,” will improve the capability and efficiency in using isotopic tools for deeper evaluation,
more rigorous assessment, and more efficient management of water resources in the region.

P-67

CARBON RESERVOIR EFFECT OF FRESHWATER IN LAKE PUMAYUM CO ON THE SOUTH-

EASTERN TIBETAN PLATEAU
Toshio Nakamura1, Takahiro Watanabe2, Tetsuya Matsunaka3, Mitsugu Nishimura3,
Liping Zhou4, Tomoko Ohta1, Masayo Minami1
1. Center for Chronological Research, Nagoya University, Chikusa, Nagoya 464-8602, Japan.
2. Graduate School of Science, Tohoku University, Japan.
3. Graduate School of marine Sciences & Technology, Tokai University, Japan.
4. Institute of Tibetan Plateau Research, Chinese Academy of Sciences, China.

Sediments on the bottom of large, long-lived lakes are quite important for analysis of paleoenviron-
ment for the Quaternary or older periods. To establish a precise chronology of the younger part of
the cored sediments from the lakes, 14C dating on terrestrial plant remains collected from the sedi-
 Abstracts 113

ments is most important. However, terrestrial plant remains are rarely obtained, and sometimes
water plants or a mixture of terrestrial and water plants are used for dating. In the case of an olig-
otrophic lake, such as Lake Baikal, total organic carbon (TOC) can only be used for 14C dating.
When using water plant fragments or TOC, we should have knowledge of the carbon reservoir effect
of freshwater in the lakes, to get reasonable age of sedimentation.
Four-meter-long bottom sediments of Lake Pumayum Co, located at the altitude of 5000 m on the
southeastern Tibetan Plateau, were collected for analysis of paleoenvironmental changes over the
last glacial to Holocene periods. During the core collection, we collected lake water, river water,
aquatic plants growing in the lake, and terrestrial plants around the lake that can be potential source
materials to be contained in the lake sediments.
We measured concentrations and carbon isotopic ratios of dissolved inorganic carbon (DIC) in water
samples, with the 14C AMS system at Nagoya University, and compared the results with those of
aquatic and terrestrial plants. 14C concentrations of lake-water DIC, aquatic and terrestrial plants
were all quite consistent with one another, suggesting that a carbon reservoir effect of lake water was
almost negligible at Lake Pumayum Co. Thus, we can use any plant fragments that were produced
from either aquatic plants or terrestrial plants for 14C dating. In fact, we have obtained a reasonable
chronology by using micro-plant fossils separated from the cored sediments there.

P-68

ASSESSING THE POTENTIAL FOR RADIOCARBON DATING OF COSMOID SCALES OF AUS-

TRALIAN LUNGFISH (NEOCERATODUS FORSTERI)
Kelly James1, Stewart Fallon1, Tom Espinoza2, Andrew McDougall2,4, Craig Broadfoot3
1. Australian National University, Canberra ACT, Australia.
2. Department of Natural Resources and Water, Bundaberg, QLD, Australia.
3. Department of Primary Industries and Fisheries, Bundaberg, QLD, Australia.
4. Current address: PO Box 1544, Broome, WA 6725, Australia.

We present a novel application for radiocarbon dating by aging four scales from a single large adult
lungfish (Neoceratodus forsteri) from the Burnett River, Queensland, Australia. This relict fish spe-
cies lacks a solid crystalline otolith, which precludes the use of the otolith annuli as a reliable age
indicator. Previous attempts to reliably age this fish using a number of techniques have not been suc-
cessful. We report on ages obtained from the dense lamellar bone of the cosmoid scale, once isolated
from the cosmine and vitrodentine layers that thicken and subsequently obscure the 14C signal. The
atmospheric radiocarbon “bomb” signature is apparent although dampened, possibly as an artifact
of sampling, or the overprint of sedimentary carbon consumed in conjunction with benthic organ-
isms. Using the inflection of the bomb curve on the largest scale an extension rate of ~288 μm/year
can be inferred between 1955 to 1966, with this rate slowing to as little as ~64 μm/year from 1966
until present. This accounts for only ~10% of the total growth from the primordia (origin of growth)
to the outer edge, inferring a minimum age of ~55 years. High-resolution sampling was performed
to correlate scale reading and extension rates and provide ecologists with a non-lethal means of
assessing population dynamics. This technique may have further application on younger fish to
determine cohorts hatched at the time of inflection in the 14C signal. The information gleaned from
this study will aid in assessing the population structure and therefore management of this vulnerable
species.
114 Abstracts

MARINE STUDIES

P-69

RADIOCARBON AGE SURVEYS OF SOUTHERN OCEAN DEEP-SEA CORALS

Andrea Burke, Laura F Robinson, Ann P McNichol, Dana S Gerlach, William J Jenkins
Woods Hole Oceanographic Institution, Woods Hole, MA, USA.

Deep-sea corals provide a unique archive of the past ocean: they have a global distribution, live to
depths greater than 5000 m, and can be dated using U-series nuclides. It is time consuming, how-
ever, to date large numbers of corals to determine which specimens have appropriate ages for paleo-
climate reconstructions. Here, we update the sealed-tube zinc reduction method of Xu et al. (2007),
and evaluate its use to make age surveys of deep-sea corals. Samples are combusted with an elemen-
tal analyzer and resulting ages are within error of those obtained by hydrolysis methods. Variations
in sample mass and different cleaning procedures have no effect on the Fm. Replicate samples from
a single coral yielded a standard deviation of 21 years (n = 8). If we assume a δ13C = 0‰ in order to
improve efficiency, the standard deviation increases to 78 years. We have dated 104 corals, a subset
of more than 6000 specimens of Flabellum, Balanophyllia and Desmophyllum collected from the
Drake Passage (50°S–70°S, 120–1700 m water depth). We convert them to calendar age using the
Marine04 curve assuming a regional reservoir correction of 774 years. This conversion is the largest
source of error because past Southern Ocean radiocarbon values are largely unknown. The corals
range in age from modern to more than 60 kyr, and include fossils from the Younger Dryas, Bølling-
Allerød, and Last Glacial Maximum. The age distributions of fossil corals differ between locations.
At Burdwood Bank on the Argentine shelf, 60% of samples were alive within the last 800 years. In
contrast, corals from seamounts in the central Drake Passage were more prevalent in the early
Holocene, with 50% alive ~ 10 kyr. Coupling U-series dating techniques with radiocarbon analyses
on these specimens will provide better estimates of the history of Southern Ocean radiocarbon.

P-70

REFINING THE BOMB RADIOCARBON PROFILE IN THE NORTH PACIFIC OCEAN USING FISH
OTOLITHS
Kristen M Munk and Jodi Neil
Alaska Department of Fish and Game, Box 115526, Juneau, AK 99811, USA.

We suspect notable differences among the bomb radiocarbon profiles ranging from southeast Alaska
to the Aleutians. For example, the post-peak radiocarbon trend for the North Pacific Ocean Pacific
halibut chronology differs markedly from the southeast Alaska yelloweye rockfish chronology. Dif-
ferences may compromise the application of a radiocarbon reference curve from one area, to the val-
idation of a species collected from another area. We are in process of refining the bomb radiocarbon
record from southeast Alaska to the Aleutians on three fronts. Otolith cores for one Pacific rockfish
species collected from 4 locales have been submitted for 14C analysis. Black rockfish (Sebastes mel-
anops) was selected because it is considered easy to age and where we assume variability will be due
to 14C variability and not age reading inaccuracy. We will compare the year markers of initial rise
and initial decline and the mean 14C values of pre-bomb and peak-bomb to assess regional differ-
 Abstracts 115

ences. Second, we are defining the post-bomb 14C profile from 1977 to 2007 using otoliths from fish
of known-age or nearly known-age. Surface marine 14C reached pre-bomb levels in southeast
Alaska in 2005; however, 14C values again spiked in 2006 and 2007. We speculate that this spike
may be due to the proximity of nearshore sample locations to 14C reservoirs; therefore, the seasonal
and increased melting from nearby glaciers. And third, we will extend the existing southeast Alaska
yelloweye rockfish chronology (1940 to 1990) from 1991 to 1998, and the southeast Alaska quill-
back rockfish radiocarbon profile (1950 to 1985) from 1986 to 1995, using otoliths of estimated age.
All 14C profile refinements (locale-specific, and extensions using known-age and young but
unknown-age fish) will increase understanding of 14C variability and be used to address age estima-
tion questions.

P-71

MARINE RESERVOIR AGES IN NORTHERN SENEGAL AND MAURITANIA COASTAL WATERS

Maurice Ndeye
Laboratoire carbone 14, Institut Fondamental d’Afrique Noire (IFAN), Université Cheikh Anta Diop, Dakar, Sénégal.

In order to estimate the modern reservoir age of seawater (R) and the local offset from the global
marine radiocarbon curve (ΔR) for coastal sites of Senegal and Mauritania, we analyzed pre-bomb
mollusk shells collected between AD 1837 and 1945. In total, 27 shell samples were measured, 19
from Senegal and 8 from Mauritania. The result in Senegal for the weighted mean of R is 511 ± 50
BP and ΔR is 176 ± 15 BP; for Mauritania R is 421 ± 15 BP and ΔR is 71 ± 13 BP. While these val-
ues indicate a significant difference from the global mean value of R for Senegal, the R value for
coastal Mauritania is close to the average ocean value R of 400 yr (Stuiver and Braziunas 1993).

P-72

VARIATIONS OF RADIOCARBON RESERVOIR AGES OF BLACK SEA WATERS AND SEDI-

MENTARY ORGANIC CARBON DURING ANOXIC PERIODS: INFLUENCE OF PHOTOSYN-
THETIC VERSUS CHEMOAUTOTROPHIC PRODUCTION
Michel Fontugne1, François Guichard1, Ilham Bentaleb1,2, Claudia Strechie3,
Nadine Tisnérat-Laborde1, Gilles Lericolais4
1. Laboratoire des Sciences du Climat et de l’Environnement (UMR1572 CNRS/CEA/UVSQ) Domaine du CNRS, 91198
Gif sur Yvette Cedex, France.
2. Institut des Sciences de l’Evolution, Montpellier (I.S.E.-M.) UMR- UMII - CNRS (UMR 5554), Université de Montpellier
II, CC 064 Paléo-environnements et Paléoclimats, 34095 Montpellier, Cedex 05, France.
3. GeoEcoMar, National Institute for Marine Geology and Geo-ecology, GeoEcoMar Str., D.Onciul nr. 23-25, 024053,
Bucharest, Romania.
4. IFREMER, Centre de Brest, DCB/GM - BP 70, F-29280 Plouzane cedex, France.

Radiocarbon activity of dissolved inorganic carbon has been measured in the northwestern Black
Sea. Both continental shelf and open-sea profiles show that surface waters are in equilibrium with
the atmosphere. The observed distribution of 14C activity shows a poor contribution of the deep
depleted radiocarbon CO2 to the photic zone. Such a distribution of 14C within the water column is
116 Abstracts

unable to explain the aging of sedimentary organic matter and reservoir age greater than 500 yrs. A
contribution of production by chemoautotrophic bacteria feeding on 14C-depleted methane at the
boundary of oxic and anoxic zones is a realistic hypothesis. A contribution estimated to less than
15% of the photosynthetic primary production is enough to explain a 14 reservoir age greater than
1300 yrs.

P-73

A RADIOCARBON METHOD FOR DETERMINING DEEP-SEA SPONGE GROWTH RATES

Rebecca Norman1, Stewart J Fallon1, Michael J Ellwood1, Michelle Kelly2
1. Research School of Earth Sciences, Australian National University, Australia.
2. National Centre for Aquatic Biodiversity & Biosecurity, National Institute of Water & Atmospheric Research (NIWA) Ltd,
Auckland, New Zealand.

A potentially simple and reliable method of dating siliceous deep-sea sponges using 14C AMS is
tested. Currently, techniques for estimating sponge growth rates include 32Si dating of siliceous spi-
cules and growth-band counting. The use of the 32Si dating technique to age sponges requires a large
amount of material (~5 g) and months to analyze. Growth-band aging requires sponges to have
defined growth bands. There are, however, large inaccuracies associated with this technique: errors
are on the order of ±10% and it relies on other indicators to confirm the age represented by a single
band (annual, seasonal, etc.). Here, we explored the use of carbon trapped with siliceous sponge spi-
cules as means of aging spicules at the time of formation. Sponges incorporate low levels (0.05%)
of proteinaceous material into their spicules as a way of directing spicule formation. Here, we liber-
ated the carbon associated with this proteinaceous material and dated it using the 14C AMS tech-
nique. As our results show, the 14C AMS enables fast, reliable analysis of a very small amount of
material (<1 mg C), which enables small samples of sponge to be analyzed. This has potential appli-
cation to spicules left in the fossil record.

P-74

MARINE RESERVOIR VARIABILITY IN THE KIMBERLEY REGION, WESTERN AUSTRALIA

Sue O’Connor1, Stewart J Fallon2, Sean Ulm3, Anthony Barham1, Ian Loch4
1. Archaeology and Natural History, RSPAS, Australian National University, Australia.
2. Research School of Earth Sciences, Australian National University, Australia.
3. Aboriginal and Torres Strait Islander Studies Unit, University of Queensland, Australia.
4. Australian Museum, Australia.

New ΔR values are presented for the Kimberley region of western Australia. Recent dating of shell/
charcoal “paired” samples collected from stratigraphically associated archaeological contexts has
highlighted the uncertainty surrounding the appropriate correction for marine reservoir effect for
this region and the need for further work. Current published estimates of ΔR for northwest Australia
are few (n = 6) and the date of live collection of the dated shells known only within 50 years (Bow-
man 1985). This paper describes the results of our recent attempt to constrain the variability of ΔR
 Abstracts 117

for this region by carrying out a dating program using known-age pre-AD 1950 shell samples from
the Australian Museum collection.

REFERENCE
Bowman GM. 1985. Oceanic reservoir correction for marine radiocarbon dates from northwestern Australia. Australian Ar-
chaeology 20:58–67.

P-75

DEVELOPING ROBUST CHRONOLOGIES OF ABORIGINAL OCCUPATION FROM SHELL MID-

DENS ALONG THE KIMBERLEY COAST OF AUSTRALIA
Stewart J Fallon1, Sue O’Connor2, Anthony Barham2
1. Research School of Earth Sciences, Australian National University, Australia.
2. Archaeology and Natural History, RSPAS, Australian National University, Australia.

In many parts of the Australian coast exposures of midden shell are ubiquitous and extensive. The
extent to which these midden “exposures” represent single or multiple occupation events, and
whether these occurred over days, decades or millennia is largely unknown. These same landscapes
are often actively eroding as a result of natural events such as cyclones and increased 4WD vehicle
access. This vanishing record of Indigenous occupation is very poorly dated. There has been much
discussion about the number of dates required to provide a robust chronology for such landscapes.
While multiple intrasite dating is desirable, in reality the coastal archaeological record for most
areas falls well short of this, being based on a few dated shells. There are also a wide range of shell
species that have been used for radiocarbon dating purposes and the comparability of dates obtained
on different species is currently poorly understood.
Here we present an occupational history of a small sampled area of the Kimberley coast near Cape
Leveque, based on multiple species shell dates from individual middens. Forty midden sites have
been identified and shell material collected, to date sixteen sites have had at least 3 shells dated from
each with a few sites having 5 or more shell dates. Shell species used include: Pinctada sp. (pearl
shell), Barbatia sp., Melo amphora (baler shell), Terebralia sp. (mud whelk), Geloina coaxans (mud
clam), oyster and Turbo sp. At one site, six individual shells from an oyster shell cluster were radio-
carbon dated; all of the ages are within the radiocarbon age range of 740-790. Calibrated age ranges
for the sampled middens range from 300-2700 BP. The implications of this study for the reliability
of regional chronologies built on small numbers of dates are discussed.

P-76

OVER 500 14C YR OF MARINE RESERVOIR AGE VARIATION IN A MESODESMA DONACIUM

SHELL FROM SOUTHERN PERU
Kevin B Jones1, Gregory W L Hodgins2, Miguel F Etayo-Cadavid3, C Fred T Andrus3,
Daniel H Sandweiss4
1. Department of Geosciences, University of Arizona, USA.
118 Abstracts

2. NSF-Arizona AMS Facility, Tucson, AZ, USA.

3. Department of Geological Sciences, University of Alabama, Tuscaloosa, AL, USA.
4. Climate Change Institute, University of Maine, Orono, ME, USA.

Mollusk shells from archaeological middens provide brief (<5 yr per shell) records of past marine
conditions. Marine radiocarbon reservoir age, R, is recorded in shell carbonate at the time of precip-
itation, and can be calculated from the radiocarbon age of shell carbonate when shell antiquity is
known. Marine upwelling changes (proportion of super- vs. sub-thermocline water) result in R
changes, so R can be used as a proxy for the source of upwelling water. We report 12 R determina-
tions using small (~2 mg) samples from a Mesodesma donacium (surf clam) shell from Site QJ-280,
Quebrada Jaguay, southern Peru: the R range is 530 14C yr; R from individual aragonite samples
ranges from 200 ± 110 to 730 ± 170 14C yr.
The archaeological stratum that contained the shell, a semi-subterranean house floor, dates to
7630 ± 30 14C yr BP and was deposited rapidly based on field observations and radiocarbon ages of
bracketing strata. Based on shell size and δ18O profile, we estimate the shell grew for ~5 yr. This
range in dates, then, suggests seasonal or interannual R variations of several hundred 14C yr in south-
ern Peru during the early Holocene. In Peru, M. donacium precipitates more shell carbonate during
winter than during summer because it prefers cold conditions. Due to this seasonal growth variation
and seasonal changes in upwelling (and thus R), radiocarbon samples from this shell may underes-
timate the R range during its life. A model of the radiocarbon sampling of a mollusk shell that grew
seasonally suggests that the actual range in R during the life of this M. donacium may have been
~700 14C yr BP. Because of intrashell R variation, caution is needed when using marine material
from a variable upwelling environment for radiocarbon chronometry.

P-77

PRE-BOMB MARINE RESERVOIR AGES IN THE NORTHWESTERN PACIFIC

Kunio Yoshida1, Tatsuaki Hara1, Dai Kunikita1, Yumiko Miyazaki1, Takenori Sasaki1,
Minoru Yoneda2, Hiroyuki Matsuzaki3
1. The University Museum, University of Tokyo, Japan.
2. Graduate School of Frontier Science, University of Tokyo, Kashiwa Choba, Japan.
3. School of Engineering, University of Tokyo, Japan.

The Japanese archipelago has many ruins of shell mounds, many of which are from the Jomon
period. At present, charred materials are often used to determine the age of a midden by the AMS
method, but as for middens excavated in the past, only shells have been preserved. In order to deter-
mine the age of the midden by measuring the age of shell, it is necessary to know the local marine
reservoir effect on the location. For the northern and southern coastal areas of the Japanese archipel-
ago, values have been reported (Yoneda et al. 2007). At Sakhalin and Hokkaido Island in the north,
ΔR values are very large, near 400 yr. On the other hand, at Nansei Island in the southeast, a small
value the value is only ~30 yr.
Similar to the previous report, the shell materials used in this study have been held at the University
Museum, the University of Tokyo. The specimens were collected before World War II and have
labels recording their gathering places, year of collection, and the collector’s name. In this study,
dates are measured for the samples collected in the main island of Japan, especially the Kanto
 Abstracts 119

region, and Izu Island and Ryukyu Island. From the Tohoku region through the Kanto region, the
regions are affected by the Kuroshio Current and/or the Oyashio Current (and/or Tsugaru Warm
Current) and the results have important implications.
Finally, we present the local reservoir effect, ΔR values, of the Japanese archipelago from north to
south. This data will play an important role in the study of archaeological and environmental
changes.

P-78

A WEATHERING ASSESSMENT OF CALCAREOUS SANDSTONES USING RADIOCARBON

Masayo Minami1, Hidekazu Yoshida2, Tsuyoshi Tanaka3
1. Center for Chronological Research, Nagoya University, Nagoya 464-8602, Japan.
2. The Nagoya University Museum, Material Research Section, Nagoya 464-8601, Japan.
3. Graduate School of Environmental Studies, Nagoya University, Nagoya 464-8601, Japan.

Weathering rinds in two Triassic calcareous sandstones collected from Tatehama and Namiita at the
Sanriku seashore, northeast Japan, were investigated to clarify the process of elemental migration,
especially modern 14C accumulation caused by atmospheric CO2, during chemical weathering.
Major element and carbon concentrations together with 14C abundances were measured across the
weathering rind in the sandstones. A specific depletion of calcium and carbon concentrations, and
enrichment of iron concentration, were observed in the weathering rind for both calcareous sand-
stones. The results indicate that the calcareous phase in the weathering rind has been leached, and
that ferrous ion has been accumulating in the rims by oxidization after migration under reducing
subsurface during chemical weathering. It is noted that the Tatehama and Namiita sandstones have
different behaviors of migration for some elements such as calcium and magnesium in the extreme
rims: little enrichment and gradual depletion for the Tatehama and Namiita samples, respectively.
The difference might be due to seawater-rock interaction because the Tatehama sample occurs near
the seashore lapped by waves and the Namiita sample does not. On the other hand, both sandstones
show the same 14C accumulation behavior in the weathering rind. 14C was not detected in the fresh
fractions of the calcareous sandstones, due to 14C-free (dead) carbon contained in a fresh Triassic
limestone, while modern 14C was detected in the weathering rind. The modern 14C could have orig-
inated from meteoric water including modern atmospheric CO2. From this study, 14C is found to be
a sensitive and general index of weathering in calcareous stones. The process of 14C uptake during
weathering of calcareous sandstones will be discussed.

P-79

AN ANNUALLY-RESOLVED MARINE RADIOCARBON BOMB-PULSE COMPILATION FROM

THE TEMPERATE NORTH ATLANTIC USING LONG-LIVED MOLLUSKS (ARCTICA ISLANDICA)
James D Scourse1, Jan Heinemeier2, Alan D Wanamaker Jr1, Chris Weidman3,
Christopher A Richardson1
1. School of Ocean Sciences, College of Natural Sciences, Bangor University, Menai Bridge, Anglesey LL59 5AB, Wales,
UK.
120 Abstracts

2. AMS 14C Dating Centre, Department of Physics and Astronomy, University of Aarhus, DK-8000 Aarhus C, Denmark.
3. Waquoit Bay National Estuarine Research Reserve, PO Box 3092, Waquoit, MA 02536-3092, USA.

Radiocarbon measurements from increments of annually-banded corals covering the past 60 years
from subtropical and tropical contexts provide valuable records of the marine expression of the
atmospheric excess radiocarbon “bomb pulse” due to post-war nuclear weapons tests. These records
can be used as calibration series for high-resolution post-bomb marine radiocarbon dating and con-
stitute tracers for identifying water-mass age and mixing processes. Until now, such applications
have been restricted in temperate shelf seas because of the lack of widespread measurements from
annually-resolved archives. Here, we present a compilation of bomb-pulse data from annual growth
increments of the bivalve mollusk Arctica islandica from relatively shallow sites (<200 m) across
the temperate North Atlantic (Georges Bank, north Icelandic shelf, north Norway, North Sea). The
temporal response is highly correlated at all sites, but the amplitude of the bomb-pulse varies, with
the highest values attained in the North Sea and the most damped response on the north Icelandic
shelf. These differences can be attributed to the integrated hydrographic context of these sites
(entrainment of deep, old water; rates of air-sea exchange; fluvial runoff; removal of high radiocar-
bon level surface waters through north Atlantic deep-water formation). The north Icelandic data
contain a reversal in the rising limb of the bomb-pulse, which is not present elsewhere, even in the
more sensitive sites. This reversal appeared to be coincident with instrumental data characterizing
the Great Salinity Anomaly of the 1960s when cold, relatively fresh, and old (with respect to radio-
carbon; ΔR = +200 years) waters of the East Icelandic Current flooded the north Icelandic shelf as
a result of southward migration of the Polar Front. Upon closer inspection, however, this reversal
turns out be a result of the short hiatus in bomb testing in the late 1950s. The evolution of bomb-
pulse data will be discussed as well as other potential applications of these records.

P-80

THE PASSAGE OF THE BOMB RADIOCARBON PULSE INTO THE PACIFIC OCEAN
W J Jenkins, K L Elder, A P McNichol
National Ocean Sciences Accelerator Mass Spectrometer Facility Woods Hole Oceanographic Institution, Woods Hole, MA
02543, USA

The GEOSECS survey of the early 1970s provided us with the first broad-brush outline of the
impingement of bomb radiocarbon into the ocean. Decades later, the high-resolution radiocarbon
picture built from the World Ocean Circulation Experiment (WOCE) global survey allowed us a
detailed visualization of the ocean ventilation pathways for shallow, intermediate, and deep water
masses. Subsequent reoccupations of the WOCE survey lines by the CLIVAR Repeat Hydrography
Program now provide us with the first high-resolution imagery of the time-evolving radiocarbon
distribution. By re-analyzing Δ14C in archived WOCE CO2 samples, we can demonstrate that there
are no significant systematic biases in our measurements.
We present here a comparison of two meridional radiocarbon sections along 150°W in the South
Pacific taken in 1991 (WOCE) and 2005 (CLIVAR). Differences as large as 70‰ in Δ14C are
observed, more than an order of magnitude larger than analytical and systematic errors. These dif-
ferences form patterns that are consistent with expectations about ocean circulation and ventilation.
During the ~14 years, we observe an overall decrease in surface water Δ14C, an embedded positive
 Abstracts 121

anomaly in the Antarctic Intermediate Water, and a surprisingly large increase in the bottom waters.
The first two phenomena are readily explained, but the last raises interesting questions. Is the
increase due to bomb radiocarbon invasion, or is it a record of longer-term changes in the global
conveyor? On the basis of these observations, inferences can be made about the magnitude and
nature of air-sea exchange, water mass ventilation, and circulation, and possibly about climatic
changes in these processes.

P-81

ΔR IN THE SOUTHERN OCEAN

Bruce McFadgen1, Rodger Sparks2, Christine Prior3
1. Research Associate, Department of Maori Studies, Victoria University of Wellington, New Zealand.
2. Emeritus Scientist, GNS Science, Lower Hutt, New Zealand.
3. Rafter Radiocarbon Laboratory, GNS Science, Lower Hutt, New Zealand.

Two mussel shells (Modiolus areolatus) collected live on known dates from the Auckland Islands
(50°40′S, 166°10′E) returned unusual values for ΔR. The results suggest significant changes in
upwelling of old water over periods of time as short as several decades. The causes of the variation,
currently unknown, are possibly related to ENSO events, although changes in the position of the
sub-Antarctic convergence may also be significant.

P-82

DAILY VARIABILITY OF DISSOLVED INORGANIC RADIOCARBON IN SURFACE WATERS AT

A TIME-SERIES SITE IN THE NORTHEAST PACIFIC OCEAN
Ellen R M Druffel1, Sheila Griffin1, Steven Beaupré1, Jeomshik Hwang2
1. Department of Earth System Science, University of California, Irvine, CA 92697-3100, USA.
2. School of Environmental Science and Engineering, POSTECH, Pohang 790-784, Korea.

We report radiocarbon measurements of dissolved inorganic carbon (DIC) in surface water samples
collected daily during 12 cruises to Station M (34°50′W 123°00′W) in the northeast Pacific. This
time-series site is located 220 km west of Point Conception, California, and lies in the path of the
southward-flowing California Current. There was a general decrease of average Δ14C values from
1994–1995 (52–61‰) to 2004 (32‰). This trend shows an annual decrease of 3‰ per year, about
the same as that measured in surface corals from the northwest Pacific during the 1990s (Morimoto
et al. 2004) and about half of that observed in atmospheric CO2 during the same period (Levin and
Kromer 2004). The ranges of daily Δ14C values were greatest during the two winter cruises (18‰)
and one fall cruise (18‰) and lower during the other nine cruises (2–15‰). These results compare
with the total uncertainty of the Δ14C measurements of ±7.8‰ (2 σ). The ranges of daily Δ14C values
from this site are somewhat lower than those reported from three other sites (Druffel and Griffin
2008) that contained three to four times the number of daily Δ14C measurements per cruise. These
results suggest that solitary measurements of surface Δ14C values are not necessarily representative
122 Abstracts

of the average surface waters because of variability that exists in the DIC Δ14C signature of the sur-
face ocean.

REFERENCES
Druffel E, Griffin S. 2008. Daily variability of dissolved inorganic radiocarbon at three sites in the surface ocean. Marine
Chemistry 110:185–9.
Levin I, Kromer B. 2004. The tropospheric 14CO2 level in mid-latitudes of the Northern Hemisphere (1959–2003). Radiocar-
bon 46(3):1261–72.
Morimoto M, Kitagawa H, Shibata Y, Kayane H. 2004. Seasonal radiocarbon variation of surface seawater recorded in a coral
from Kikai Island, subtropical northwestern Pacific. Radiocarbon 46(2):643–8.

METHODS AND DEVELOPMENTS

P-83

COMPOUND-SPECIFIC RADIOCARBON ANALYSES OF PHOSPHOLIPID FATTY ACIDS AND

ALKANES IN OCEANIC SEDIMENTS USING PREPARATIVE CAPILLARY GAS CHROMATO-
GRAPHIC AND AMS TECHNIQUES
Ellen R M Druffel, Dachun Zhang, Xiaomei Xu, Lori A Ziolkowski, John R Southon,
Guaciara dos Santos, Susan E Trumbore
Department of Earth System Science, University of California, Irvine, CA 92697-3100, USA.

We report radiocarbon analyses of lipid classes and individual compounds extracted from oceanic
sediments collected from the coast of southern California, USA. Phospholipid fatty acids (PLFAs)
and neutral lipids were extracted and separated using standard techniques (Zelles 1999). Alkanes
were extracted using urea adduction according to standard techniques (Pearson and Eglinton 2000).
PLFAs and alkanes were collected by preparative capillary gas chromatography in sample sizes that
ranged from 13–40 μg C. We report Δ14C measurements of compounds extracted from three depth
horizons in the surface sediment. The PLFAs had near-modern Δ14C values that decreased with
depth in the sediment. Neutral lipids had much lower Δ14C values. Blank carbon from PCGC col-
umn bleed, previous sample memory, and solvents were assessed in this study. Implications for the
cycling of organic matter in continental margin sediments are presented.

REFERENCES
Pearson A, Eglinton T. 2000. The origin of n-alkanes in Santa Monica Basin surface sediment: a model based on compound-
specific Δ14C and δ13C data. Organic Geochemistry 31:1103–16.
Zelles L. 1999. Fatty acid patterns of phospholipids and lipopolysaccharides in the characterisation of microbial communities
in soil: a review. Biol Fertil Soils 29:111–29.
 Abstracts 123

P-84

AN ALTERNATIVE METHOD OF DILUTING DISSOLVED ORGANIC CARBON (DOC) HIGH CON-

CENTRATE SAMPLES FOR 14C ANALYSIS
Sheila Griffin, Steven R. Beaupré, Ellen R M Druffel
Department of Earth System Science, University of California-Irvine, CA, 92697, USA.

A modified, low blank, ultraviolet oxidation and vacuum line system to convert DOC in seawater
samples to carbon dioxide for concentration, Δ14C, and δ13C analyses was reported recently (Beau-
pré et al. 2007). They report a low background (~0.2 μM) using Milli-Q water that had been pre-irra-
diated prior to the addition of the sample. We present a revised method that uses Milli-Q water
directly from the Millipore system to dilute fresh and seawater samples. The DOC concentration of
this Milli-Q water is 0.8 ± 0.2 μM (n = 13). Therefore, samples that are highly concentrated, but not
available in 1-L aliquots, can be diluted with Milli-Q water that has not been pre-irratiated without
losing significant precision of the Δ14C measurements. This revised method reduces the time
required to extract CO2 for high-precision analyses from two days to one.

REFERENCE
Beaupré SR, Druffel ERM, Griffin S. 2007. A low-blank photochemical extraction system for concentration and isotopic anal-
yses of marine dissolved organic carbon. Limnol. Oceanogr.: Methods 5:174–84).

P-85

RADIOCARBON DETERMINATION OF THE AGE OF FOSSIL SINTER DEPOSITS

Christine A Prior
National Isotope Centre, GNS Science, Lower Hutt, New Zealand.

Siliceous sinters are formed around hot springs as emerging geothermal water cools at the surface.
The hot water contains high concentrations of dissolved silica (SiO2), which precipitates to form the
sinter. Organic material such as plant fragments and pollen often becomes incorporated into the sin-
ter, which, once formed, protects it from decay. Radiocarbon analysis of the enclosed organic mate-
rial enables researchers to date the formation of sinters and study the evolution of geothermal areas.
This poster describes a method for extracting and concentrating microscopic plant material from sil-
iceous sinters and presents examples from the dating of sinters in the geothermal regions of the cen-
tral North Island of New Zealand.

P-86

BLANK ASSESSMENT FOR ULTRA-SMALL SAMPLES: EXTRACTION CHEMISTRY VERSUS

AMS
Guaciara M Santos and John R Southon
Earth System Science Dept, University of California, Irvine, CA 92612, USA.

The KCCAMS/UCI laboratory has developed protocols for analyzing 14C in samples containing as
small as 0.002 mg of carbon (Santos et al. 2007a). Mass balance background corrections of modern
124 Abstracts

and dead carbon contamination can be quantified by measuring blanks and modern samples submit-
ted to the same sample processing as the unknown samples. The validation of this small sample
approach is obtained by measuring secondary standards of similar size and composition to the
unknown samples. Today, the laboratory routinely produces graphite samples down to 0.015 mg car-
bon for outside submitters and internal research samples as small as 0.001 mg C, with precision of
~1% determined on standard samples ~0.010 mg C (Santos et al. 2007b). However, when dealing
with unknown small samples produced by complex sample processing (such as compound-specific
extractions), a more elaborate blank assessment is required. Chemistry blanks (e.g. from GC column
bleed and solvent residue) may be far larger than those of the AMS laboratory, and possible contam-
ination by 14C-free carbon, as well as modern carbon, must be carefully evaluated. In some cases,
the magnitude of the blanks maybe so high that sample results are meaningless if low processing
yields result in very small samples. Case studies will be presented to illustrate this issue.

REFERENCES
Santos et al. 2007a. Ultra small-mass 14C-AMS sample preparation and analysis at the KCCAMS Facility. Nuclear Instru-
ments and Methods in Physics Research B 259:293–302.
Santos et al. 2007b. AMS 14C sample preparation at the KCCAMS/UCI Facility: status report and performance of small sam-
ples. Radiocarbon 49(2):255–69.

P-87

MEASUREMENT OF 14 C IN QUARTZ SAMPLES FROM NORTHERN ABUKAMA REGION,

JAPAN, USING NATURAL GRAPHITE AS PRECURSOR OF CARRIER
Takumi Kubota1, Yasunori Mahara1, Tomoko Ohta1, Toshio Nakamura2
1. Kyoto University Research Reactor Institute, Japan.
2. Nagoya University Dating and Material Research Center, Japan.

Quartz samples (which were collected at up to 120 g/cm2 in the northern Abukuma region, Japan),
natural graphite and copper oxide were stepwisely heated to 1900 °C in molybdenum crucible cov-
ered with tantalum vessel using a tantalum heater outside the vessel. In this heat treatment, 14C was
extracted from melted quartz and natural graphite was oxidized with copper oxide to carbon carrier
gas, which was produced at each treatment. In order to decrease the background in our extraction
system, natural graphite was used as precursor of carrier gas because natural graphite has low 14C
content and it is unlikely that isotopic exchange occurs with atmospheric carbon during storage.
Carbon monoxide in the resulting gas was converted to CO2 through a heated mixture of copper
oxide and platinum. CO2 was purified using a mixture of liquid nitrogen and ethanol (ca. –70 °C),
then its volume was measured to evaluate the carbon amount at room temperature, and CO2 was iso-
lated and sealed into glass tube prior to graphitizing for AMS measurement.
AMS measurement of natural graphite showed that the 14C date was 50,600 ± 750 BP, equating to
(1.2 ± 0.1) × 105 (14C atom/mg graphite). Based on this value, the extraction system blank was eval-
uated from results obtained with natural graphite and copper oxide, without the quartz sample, to be
(2.0 ± 0.5) × 105 (14C atom). The gross amount ranged from 1.2 × 106 to 1.9 × 106 (14C atom), while
the value of background ranged from 3.6x × 105 to 6.5 × 105 (14C atom). The concentration of 14C in
quartz ranged from 6.6 × 104 to 1.5 × 105 (14C atom/g quartz), however, indicating less dependence
on depth from ground surface.
 Abstracts 125

P-88

THE USE OF HUMAN HAIR AND NAILS AS BIOINDICATORS OF OCCUPATIONAL 14C CON-
TAMINATION
Kristina Stenström and Ingmar Unkel
Lund University, Department of Physics, Division of Nuclear Physics, P.O. Box 118, SE-221 00 Lund, Sweden.

As 14C is a low-energy beta-emitter (Eβ(max) = 156 keV, Eβ(mean) = 49.5 keV), it does not present
an external radiation hazard. However, carbon—including the radioactive 14C isotope—is easily
incorporated in living organisms, resulting in an internal radiation exposure. Thus, for retrospective
dose assessment, the internal 14C contamination in the body is of main interest. Workers handling
14C-labeled compounds or working in a 14C-enriched environment are exposed to the risk of acci-

dental intake of additional 14C into their bodies through ingestion, inhalation, or dermal absorption.
This paper presents the experiences from a study in which specific activities of 14C in hair and nails
were used as a bioindicators of the integrated 14C exposure (over several months) of workers han-
dling 14C-labeled substances, or working in potentially 14C-contaminated premises. However, in
hair from personnel at pharmaceutical development laboratories, the majority of the 14C appeared to
originate from airborne 14C compounds, which had adsorbed on the hair. The difficulties to remove
this external 14C contamination prior to accelerator mass spectrometry (AMS) analysis are
addressed, and the graphitization method used for hair and nail samples is presented.

P-89

NEW PROJECTS IN THE 14C MEASUREMENT LABORATORY LMC14 IN SACLAY, FRANCE

A Quiles1, C Moreau1, G Andreu2, D Bavay1, I Caffy1, C Comby1, E Delqué-KoliÊ1,
J-P Dumoulin1, S Ferkane1, S Hain1, G Pierrat2, V Setti1, C Souprayen1
1. LMC14 UMS 2572, Bât 450 porte 4E, CEA Saclay, 91191 Gif sur Yvette, France.
2. Département des Antiquités Égyptiennes, Palais Royal, Musée du Louvre 75001 Paris, France.

Installed in 2003 in Saclay (France), the Artemis facility routinely measures 3800 samples per year
for French organizations covering several fields of research (e.g. environmental science, archaeol-
ogy, geology). Since October 2008, a PhD project on beam optic simulation on the NEC Pelletron
AMS of the LMC14 was carried out to improve the technical factors involved. Simulations of the
optical spectrum are based on computer tools used in nuclear and particle physics. After the facility
presentation, we will focus on the first attempts of simulations.
In addition to this methodological project on the facility itself, a program of radiocarbon dating on
organic Egyptian remains has begun. The project is in collaboration with the Oxford Radiocarbon
Unit of Oxford University (ORAU, UK) who has done research on this topic over the last few years.
The aim of this program is to create a database for this period to investigate the consistency of the
historical Egyptian chronology with radiocarbon dates. The Departement des Antiquités Egypti-
ennes from Le Louvre in Paris actively participates in this project, providing samples from its col-
lection. The first measurements, done on samples from Deir el-Medineh, under the reign of Thut-
moses III, will be presented and confronted with Egyptological data, allowing a new proposal for the
chronological extent of the reign of Thutmoses III.
126 Abstracts

P-90

I M P RO V E M E N T S I N T H E B E N Z E N E S Y N T H E S I S A T T H E E N E A RA D I O C AR B O N
LABORATORY
Giuseppe Magnani, Paolo Bartolomei, Teresa La Torretta, Ernesto Claudio Marino
ENEA, Radiocarbon Laboratory, Via dei Colli, 16, 40136 Bologna, Italy.

The ENEA Radiocarbon Laboratory has been involved in radiocarbon dating since 1985. The orig-
inal benzene synthesis line was completely renewed in 2006 with the introduction of important tech-
nological improvements aimed to reduce the laboratory background and memory effect. An innova-
tive combustion system based on ceramic IR heaters has been designed and installed in order to
ensure the complete conversion of the organic carbon in CO2. Lithium carbide synthesis is per-
formed in an Inconel reactor in order to minimize the memory effect, while a turbomolecular pump-
ing system is used to reduce the laboratory background. Results of the analyses of IAEA radiocar-
bon standards will be presented to show the performance of this new benzene synthesis line.

P-91

A PREPARATIVE 2D-CHROMATOGRAPHY METHOD FOR COMPOUND-SPECIFIC RADIOCAR-

BON ANALYSIS OF AEROSOL COMPONENTS
S Fahrni1,2,3, H W Gäggeler1,2,3, M Ruff1,2, L Wacker4, S Szidat1,3
1. Department of Chemistry and Biochemistry, University of Bern, Switzerland.
2. Laboratory for Radiochemistry and Environmental Chemistry, Paul Scherrer Institute, Villigen, Switzerland.
3. Oeschger Centre for Climate Change Research, University of Bern, Switzerland.
4. Ion Beam Physics, ETH Hönggerberg, Zurich, Switzerland.

As carbonaceous aerosols influence the global radiation balance and can affect health, there is a
great scientific demand for an assessment of the sources and formation processes of their compo-
nents. One of the main compounds found in such aerosols is oxalic acid. This small dicarboxylic
acid is thought to be produced through the decomposition of secondary organic aerosols (SOA),
which make up for about half of the mass in carbonaceous aerosols. Nevertheless, formation pro-
cesses and sources of oxalic acid are not completely understood and are thus subject to speculation.
As compound-specific radiocarbon analysis (CSRA) offers a unique tool for the assessment of the
fossil and non-fossil sources of environmental pollutants, we have developed a method for the sep-
aration of oxalic acid from aerosols. This method consists of a water extraction from air filters fol-
lowed by a 2D-chromatography separation based on ion chromatography and HPLC and a gentle
drying step. Upon separation, radiocarbon measurement of small samples (10 μg range) are con-
ducted at the gas ion source of the small accelerator mass spectrometry facility MICADAS. We
present a validation of the method with reference materials.
 Abstracts 127

P-92

METHOD DEVELOPMENT FOR VERY SMALL MASS SAMPLES AT THE OXFORD RADIOCAR-
BON ACCELERATOR UNIT
R A Staff, F Brock, D Chivall, P Ditchfield, M Humm, P Leach, C Bronk Ramsey
Research Laboratory for Archaeology and the History of Art, University of Oxford, UK.

As at other radiocarbon laboratories, Oxford Radiocarbon Accelerator Unit (ORAU) has continued
to hone its methodology to produce ever more robust radiocarbon determinations. Particular
demands from the wider radiocarbon community are towards: i) extending the background of the
technique; ii) improving dating reliability, both in terms of accuracy and precision; and iii) the abil-
ity to generate reliable dates from ever smaller sample sizes. Here, the empirical results of experi-
ments to address this third point are presented, and recommendations made as to the protocols,
which will be adopted in future for such small-mass samples at ORAU.

P-93

AGE AND GROWTH RATE DYNAMICS OF AN OLD AFRICAN BAOBAB DETERMINED BY

RADIOCARBON DATING

Adrian Patrut1, Karl F von Reden2, Diana H Mayne3, Daniel A Lowy4, Sarah Venter3,
Ann P McNichol2, Mark L Roberts2, Dragos Margineanu1
1. Department of Chemistry, Babes-Bolyai University, 400028 Cluj-Napoca, Romania.
2. NOSAMS Facility, Department of Geology & Geophysics, Woods Hole Oceanographic Institute, Woods Hole, MA 02543,
USA.
3. Baobab Trust, Box 1566, Parklands 2121, Johannesburg, South Africa.
4. Nova Research Inc., Alexandria, VA 22308, USA.

The Makulu Makete Wildlife Reserve, located in the NNW area of South Africa, contains several
large African baobabs (Adansonia digitata L.). In May 2008, one half of the biggest tree in the
reserve (22°34.584′S, 25°52.261′E), with a cbh (i.e. circumference at breast height) of 22.25 m, was
found lying on the ground. Several wood samples were collected from the large quasi-central cavity
located in the still-standing northern section, which was observed only after the tree split, as well as
from the fallen southern section. The northern section also split and collapsed in November 2008.
The wood samples were processed and investigated by AMS radiocarbon dating. The radiocarbon
date of the oldest sample was found to be 1016 ± 22 BP (Fm = 0.8812 ± 0.0027), corresponding to
a calibrated age of 1000 ± 15 yrs. Dating results of other samples collected from different positions
of the cavity show that the apparent growth rate of the trunk, measured by the annual increase in
radius Δr, decreased gradually during its life cycle. On the other hand, the true growth rate of the
trunk, which is measured by the increase in volume and is proportional with Δ(r2), remained almost
constant or even increased slightly over time. The age of a collapsed high branch indicates that it
belonged to the original crown structure. According to radiocarbon analysis, the Makulu Makete Big
tree becomes the second oldest African baobab that has been accurately dated to at least 1000 yrs.
128 Abstracts

P-94

FIRE HISTORY OF A GIANT AFRICAN BAOBAB EVINCED BY RADIOCARBON DATING: COM-

PARATIVE CALIBRATION WITH NORTHERN VS. SOUTHERN HEMISPHERE DATA SETS
Adrian Patrut1, Karl F von Reden2, Daniel A Lowy3, Diana H Mayne4, Robert van Pelt5,
Ann P McNichol2, Mark L Roberts2, Dragos Margineanu1
1. Department of Chemistry, Babes-Bolyai University, 400028 Cluj-Napoca, Romania.
2. NOSAMS Facility, Department of Geology & Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA
02543, USA.
3. Nova Research Inc., Alexandria, VA 22308, USA.
4. Baobab Trust, Box 1566, Parklands 2121, Johannesburg, South Africa.
5. College of Forest Resources, University of Washington, Seattle, WA 98195, USA.

The research reports the first dating of a live and standing African baobab (Adansonia digitata L.),
by investigating samples collected from inner cavities. The giant Platland tree, also called Sunland
baobab, is a remote specimen located in the Limpopo Province, South Africa (23°37.361′S,
30°11.864′E). Its large trunk comprises two cavities inside the two partially fused stems, which
communicate via a small opening/passage. Several wood samples were collected from different
positions of the two inner cavities.
Here, we present the investigation of 14 segments extracted from determined positions of the origi-
nal samples, which correspond to a depth of up to 20 cm in the wood. The respective segments were
processed and analyzed by 14C AMS. Fraction modern (Fm) values were converted to calendar ages
by using comparatively Northern vs. Southern Hemisphere data sets, in order to verify the influence
of the Intertropical Convergence Zone on regional/local radiocarbon concentration. For 8 segments
showing values Fm >1, the post-bomb NH zone 3 and post-bomb SH data sets were employed. The
6 segments with values Fm <1 were calibrated with the IntCal04 and SHCal04 data sets.
Calibrated calendar ages of segments show lower values than the expected ones and are not corre-
lated with their positions in the cavities/stems of the tree. Dating results indicate that the investi-
gated segments consist of new growth layers, with a thickness of several centimeters, which were
triggered by successive fire damages inside the cavities. Sample ages document five large fires that
occurred in the cavities of Platland tree. The respective fires were dated around AD 1650, 1750–
1780, 1900, 1955, and 1990. Age differences due to the calibration data sets used are very small.
The obtained results demonstrate that the radiocarbon dating method is suitable for evincing succes-
sive fire episodes suffered by the African baobab.

P-95

EXPANDING ACCESS TO RADIOCARBON ANALYSES IN THE OCEAN SCIENCES: IMPROVE-

MENTS AT NOSAMS
A P McNichol, A R Gagnon, L Xu, D S Gerlach, and NOSAMS SPL
NOSAMS, Woods Hole Oceanographic Institution, Woods Hole, MA 02543 USA.

The Sample Preparation Laboratory (SPL) at the National Ocean Sciences Accelerator Mass Spec-
trometry Facility (NOSAMS) is responsible for providing radiocarbon analyses to the US ocean sci-
ences community. Samples for organic carbon measurement collected in a marine environment
 Abstracts 129

present challenges for chemical pretreatment due to the high salt content and relatively high inor-
ganic carbon concentration of seawater. Here, we detail the methods and facilities we have recently
added for isolating and measuring radiocarbon in dissolved organic carbon (DOC) and individual
organic compounds or compound classes. Improvements we have made in other routine pretreat-
ment procedures will be presented.
DOC, operationally defined as the organic carbon in water that passes through a 0.2- to 1-μm pore
size filter, is a key component in the marine carbon budget and global carbon cycle. The natural car-
bon isotope ratios of DOC and other carbon pools render information on the nature and rate of these
processes and are, thus, of great importance for understanding the global carbon cycle. We have set
up an ultraviolet ray oxidation method (Beaupré et al. 2007) to decompose the DOC in seawater
samples.
Organic matter in the oceanic water column and sediments can come from a variety of sources with
a range of ages. Measuring the radiocarbon content of individual organic compounds isolated from
these matrices can improve our understanding of these complex mixtures. We have expanded our
compound isolation capabilities with the addition of a microwave extraction system and are pres-
ently improving our HPLC capabilities. These improvements allow us to better serve visitors who
would like to isolate compounds from marine samples for radiocarbon analysis.

P-96

USING DIRECT ABSORPTION METHOD AND LIQUID SCINTILLATION COUNTING FOR 14C

MEASUREMENTS IN ORGANIC SEDIMENTS

Irina Vagner1, Ioan Stefanescu1, Stela Cuna2, Carmen Varlam1, Ionut Faurescu1
1. Institute of Cryogenics and Isotopic Technologies, Ramnicu Valcea, Romania.
2. Institute of Isotopic and Molecular Technologies, Cluj Napoca, Romania.

The CO2 direct absorption method of sample preparation for radiocarbon analysis by liquid scintil-
lation counting has been successfully employed in many laboratories primarily for “dating” ground-
water by measuring 14C from dissolved inorganic carbon. The timeless consumption for the direct
absorption method is a strong argument in extending its use for various types of samples, yet the
method is following the similar principles as the benzene synthesis method. In order to ensure the
consistency of the procedure for various samples, different types of soil and slurry from the bottom
sediments of rivers and lakes have been investigated. The total carbon in these samples can be inor-
ganic and organic carbon type. Inorganic carbon can be straightforward analyzed using the direct
absorption method by sample acidification and CO2 capture. In this work, a hybrid method that con-
nects the wet oxidation of organic carbon to the direct absorption method has been extensively
investigated. The procedure applied to the samples was based on wet oxidation with potassium
dichromate solution 1N, in the presence of concentrated sulphuric acid and silver sulphate, prior
acidified to release inorganic carbon as CO2. In order to overcome the concern of incomplete diges-
tion of organic matter, extensive heating during wet oxidation of the sample has been conducted. We
investigated the occurrence possibilities of isotopic fractionation during CO2 production from raw
material by measuring δ13C from the sample and obtained CO2. The reproducibility and accuracy of
this adapted method of sample preparation for 14C measurements by liquid scintillation have been
quantified.
130 Abstracts

P-97

HYDRODYNAMIC CONTROLS ON ARCHAEAL TETRAETHER LIPID COMPOSITIONS IN

WASHINGTON MARGIN SEDIMENTS: INSIGHTS FROM COMPOUND-SPECIFIC RADIOCAR-
BON MEASUREMENTS
M Uchida1,2, T I Eglinton1, D Montluçon1, A Pearson3, J M Hayes4
1. Woods Hole Oceanographic Institution, Dept. of Marine Chemistry & Geochemistry, Woods Hole, MA 02543, USA.
2. Now at AMS Facility (NIES-TERRA), National Institute for Environmental Studies, Tsukuba, 305-8506, Japan.
3. Dept. of Earth and Planetary Sciences, Harvard University, 20 Oxford Street, Cambridge, MA 02138, USA.
4. Woods Hole Oceanographic Institution, Department of Geology & Geophysics, Woods Hole, MA 02543, USA.

Continental margin sediments represent a large sink of organic carbon derived from marine and ter-
restrial sources. Archaeal glycerol dibiphytanyl glycerol tetraether lipids (GDGTs) are derived from
both marine and terrestrial sources and have been used both for reconstruction of paleo sea surface
temperatures and as an index of terrestrial carbon input to the marine sediments. However, the
sources and modes of supply as well as the preservation of GDGTs in marginal sediments are poorly
understood. The distribution and deposition of GDGTs is further complicated by hydrodynamic pro-
cesses. We have analyzed a suite of surface sediment samples collected along a transect from the
mouth of the Columbia River, across the Washington Margin, to the Cascadia Basin in the northeast
Pacific Ocean. Sediments were separated according to their grain size and hydrodynamic properties,
and the organic matter characterized in terms of its bulk elemental, isotopic, and molecular proper-
ties. Here, we present radiocarbon measurements on individual GDGTs, alkenones, and fatty acids
from size-fractionated sediments from shelf and slope sediments, and discuss the results in the con-
text of previous studies of the molecular abundances and isotopic compositions of sedimentary
organic matter for in this region. Systematic variations in elemental, isotopic, and molecular-level
composition are observed across the different particle classes. Moreover, these variations are mani-
fested in the isotopic composition of different molecular markers of both marine and terrestrial
sources organic matter. Both marine-derived lipids, including alkenones and marine archaeal tetra-
ethers, and soil microbe-derived tetraether lipids show strong distributional and isotopic variations
among the size-fractionated sediments. These variations in terrestrial and marine biomarker proper-
ties inform on the sources, particle dynamics, and transport history of organic matter buried on river-
influenced continental margins. The implications of these findings for the application of molecular
markers as proxies of organic matter input, and on the interpretation of past marine and continental
environmental conditions from sedimentary records, will also be discussed.

P-98

EVIDENCE OF HETEROTROPHIC MICROBIAL DECOMPOSITION OF SOIL ANCIENT CARBON,

SVALVARD, NORWAY ARCTIC: INSIGHTS FROM COMPOUND-SPECIFIC RADIOCARBON
ANALYSIS
Masao Uchida1, Masaki Uchida2, Miyuki Kondo1, Yasuyuki Shibata1
1. AMS Facility (NIES-TERRA), National Institute for Environmental Studies (NIES), Tsukuba, Japan.
2. National Institute of Polar Research, Tokyo, Japan.

Spitsbergen Island is located in the Norwegian High Arctic. Recent works report fragile character-
istics in the soil organic carbon reservoir under near future global warming. The high latitude region
 Abstracts 131

including the Arctic has the largest reservoir of soil organic carbon on Earth. Since the high latitude
region would be significantly influenced with environmental change due to global warming, the fra-
gility of the soil carbon reservoir should be studied. Thus, we investigated here the carbon source of
soil respiration to evaluate fragile characteristics of the soil organic carbon using radiocarbon. We
used radiocarbon for carbon source apportionment from modern plant and ancient carbon, which
accumulated in the geological timescale. Soil samples down to 50 cm depth were collected for com-
pound-specific radiocarbon analysis (CSRA) of bulk organic matter, microbial lipids (phospholipids
fatty acids; PLFAs), and inaerobic microbial lipids (glycerol dibiphytanyl glycerol tetraether;
GDGTs) in situ living in the soil. In addition, CO2 from the soil and soil-respired CO2 were collected
using a vacuum-tight glass bottle for radiocarbon analysis. We report the instability of Arctic soil
carbon reservoir in the light of carbon sources used by the heterotrophic microbial community.

PRETREATMENT

P-99

ANSTO AMS FACILITY SAMPLE PROCESSING AND TARGET PREPARATION: AN UPDATE

Geraldine E Jacobsen, Linda Barry, Fiona Bertuch, Quan Hua, Charles Mifsud, Prabha Pratap,
Neil Reilly, Simon Varley, Alan Williams
Institute for Environmental Research, Australian Nuclear Science and Technology, Organisation, PMB1, Menai, NSW 2234,
Australia.

The ANSTO AMS Facility has been operating for the past 17 years, and comprises two accelerators
complemented with a suite of chemistry laboratories dedicated to the processing of samples for car-
bon, beryllium, aluminium, iodine, and actinide analyses. The facility performs and supports a wide
range of research in the areas of paleoclimate change, water resource sustainability, archaeology,
geomorphology, and nuclear safeguards. As a result, the chemistry laboratories are called upon to
process a large variety of sample types and increasing numbers of samples.
The radiocarbon laboratories process charcoal, wood, sediments, pollen, carbonates, waters, tex-
tiles, and bone though the pretreatment stages, combustion or hydrolysis, and graphitization. Over
the years, we have continually worked to improve pretreatment methods, reduce sample size, and
reduce background. Construction of a dedicated low-background combustion and graphitization
system is underway.
The cosmogenic laboratories process quartz-bearing rocks and sediments through cleaning, dissolu-
tion, separation, and purification of Be and Al and preparation of targets as oxides. In this poster, we
will summarize the current methods and developments in the radiocarbon and cosmogenic chemis-
try laboratories.
132 Abstracts

P-100

MEASUREMENT OF 14C IN CARBONACEOUS AEROSOL FOR SOURCE APPORTIONMENT AT

LUND UNIVERSITY
J Genberg, K Stenström, E Swietlicki
Nuclear Physics, Department of Physics, Lund University, Box 118, Se-22100, Lund, Sweden.

Aerosol particles contribute to a cooling effect of the Earth’s climate, but the magnitude is shrouded
by large uncertainties. In order to acquire a better understanding regarding the aerosols, the sources
need to be identified. A significant part of the aerosol mass is found in carbonaceous matter, origi-
nating from combustion of fossil fuels, biomass burning, as well as from biogenic sources (the latter
mainly associated to secondary particles formed from gaseous emissions like terpenes). Thus,
source apportionment of aerosols from these sources is of particular importance.
Carbonaceous aerosol is generally divided into two fractions: organic and elemental carbon (OC and
EC, respectively). The dividing line is not fully clear, but EC absorbs light and is thermally more sta-
ble compared to OC. EC originates from incomplete combustion and is therefore not present in bio-
genic aerosol. EC is also insoluble in water, whereas OC may be separated into a water-soluble frac-
tion and a water-insoluble fraction.
Using known and estimated values of the 14C content (Fm) of the different sources and measure-
ments of the EC/OC ratio as well as the 14C content of the different fractions, the relative contribu-
tion of the sources to the carbonaceous aerosol mass can be determined. This study presents a
scheme for separating the carbonaceous components with limited use of specialized equipment,
making it more accessible to other laboratories. The physical separation of OC and EC uses the
water insolubility and thermal stability of EC. After water and heat treatment, the remaining carbon
is regarded as pure EC and analyzed by accelerator mass spectrometry (AMS).

P-101

DEVELOPMENT OF GRAPHITIZATION OF μg-SIZED SAMPLES AT LUND UNIVERSITY

J Genberg, K Stenström, M Elfman, M Olsson
Nuclear Physics, Department of Physics, Lund University, Box 118, Se-22100, Lund, Sweden.

Accelerator mass spectrometry (AMS) analysis of atmospheric aerosol samples is a tool used to
reveal their sources, which is of importance to understand the present climate. This leads to require-
ments of handling small sample sizes since the available carbon content may be as low as 5 μg.
Therefore, a new graphitization line was constructed at Lund University influenced by the work of
Santos et al. (2007). The new line is equipped with six on-line combustion tubes to enable higher
sample throughput and six graphitization reactors, the latter with an internal volume of approxi-
mately 1.5 cm3. The small volume results in a complete reaction normally within 1 hour. Compared
to the line’s predecessor, the reactors are equipped with smaller pressure transducers capable of
measuring pressure change less than 1 mbar. An absolute pressure sensor may be connected to the
line in order to calibrate the built-in transducers. Water is removed from the reaction using solid
magnesium perchlorate. Using CuO as oxidizing agent and Cu (to remove any free O2 formed in the
process) led to a rather high and variable background level with approximately 2 μg of carbon as
 Abstracts 133

carbon dioxide. This background was present in newly cleaned chemicals exposed to atmospheric
conditions for only a few minutes. A combustion process in pure oxygen will therefore be presented.

REFERENCE
Santos et al. 2007. Nuclear Instruments and Methods in Physics Research B 259.

P-102

APPLICATION OF DIFFERENT ANALYTICAL TECHNIQUES ON POTTERY FOOD CRUSTS AS

A SAMPLE QUALITY CONTROL FOR 14C DATING
Mathieu Boudin1, Wim Fremout1, Mark Van Strydonck1, Philippe Crombé2
1. Royal Institute for Cultural Heritage, Jubelpark 1, B-1000 Brussels, Belgium.
2. Ghent University, Department of Archaeology and Ancient History of Europe, Blandijnberg 2, B-9000 Ghent, Belgium.

In order to verify the relative dating of the Swifterbant pottery collected at the Final Mesolithic site
of Doel “Deurganckdok,” with dating based on pot morphology and decoration, a series of direct
dates were obtained on samples of food crusts preserved on the inner surface of numerous potsherds.
In addition, a number of indirect dates were done on samples of organic material and cremated
bones originating from presumed surface-hearths. The results indicate an important incompatibility
between the food crust dates and the other dates, the former being systematically older. This incom-
pability may be explained by a reservoir effect of the food crusts, caused by the processing of (fresh-
water) fish.
Different analytical techniques (gas chromatography mass spectrometry of lipids, gas chromatogra-
phy combustion isotope ratio mass spectrometer of lipids, element analyzer isotope ratio mass spec-
trometry of the food crusts, nano-HPLC-ESI-QTof-MS/MS of peptides) were applied and none can
prove with 100% reliability the presence of aquatic products in food crusts of pottery. Direct dating
of food crusts remains problematic.

P-103

A NEW AUTOMATED EXTRACTION SYSTEM FOR 14C MEASUREMENT IN ATMOSPHERIC CO2

Jocelyn Turnbull1, Scott Lehman2, Stephen Morgan2, Chad Wolak2
1. NOAA Earth System Research Laboratory, 325 Broadway, Boulder, CO 80305, USA.
2. INSTAAR, University of Colorado, Boulder 80309, USA.

The radiocarbon content of atmospheric CO2 (Δ14CO2) has long been of interest to atmospheric and
earth system researchers. In the last few years, improvements in 14C measurement precision and
sample size requirements have made it possible to measure 14C in conjunction with existing trace
gas sampling networks, with numerous applications, most notably as a method to quantify fossil fuel
CO2 in the atmosphere. At NOAA/ESRL and INSTAAR, we currently prepare 300–600 atmo-
spheric 14C samples from around the globe each year, and as results garner interest from the atmo-
spheric community, that number is rapidly increasing. In order to accommodate the larger number
134 Abstracts

of samples, we developed an automated extraction system to quantitatively isolate CO2 from whole
air for AMS 14C analysis.
Our system is able to extract CO2 from 1–5 liters of whole air collected in a variety of storage media,
including “standard” glass and steel flasks at or below ambient pressure, multisample, pressurized
NOAA/ESRL portable flask packages (PFPs), and high-pressure cylinders typically used as stan-
dards. Up to 20 samples can be extracted in one fully automated run, which takes 10–12 hours to
complete, but only about 2 hours of operator time. This represents a substantial improvement over
our previous manual extraction method from which 4–6 samples could be extracted per day, and
required continuous operator attention. CO2 is extracted cryogenically by flowing the whole air over
a liquid nitrogen trap, after first removing water in a trap at –85 °C. Large volume vacuum lines are
used to extract ~30 μmol of CO2 in less than 10 minutes, thus keeping contamination from leaks to
a minimum, as well as allowing rapid processing of large numbers of samples. δ13C measurements
on the resultant CO2 demonstrate that extraction is quantitative, and extractions of 14C-free air show
that no significant modern contamination occurs.

P-104

THE UTILITY OF CHROMOSORB® FOR MONITORING CONTAMINATION DURING 14C PRE-

TREATMENT
M W Dee, F Brock, T F G Higham, A D Bowles, C Bronk Ramsey
University of Oxford, RLAHA, Dyson Perrins Building, Oxford, Oxfordshire OX1 3QY, UK.

Analyzing the efficiency of 14C pretreatment is frequently problematic. If the approach taken uti-
lizes known-age standards, a large number of 14C measurements are needed for valid inferences to
be drawn from the normally distributed results. Alternatively, techniques based on mass changes
resulting from the removal of exogenous substances are frequently undermined by the simultaneous
loss of sample material. The procedure presented here employs the inert non-carbonaceous material,
Chromosorb, to monitor the flow of contamination during pretreatment. Chromosorb was substi-
tuted for 14C samples and then subjected to routine pretreatment procedures. After each step, one
Chromosorb “sample” was withdrawn and analyzed for carbon content. Using this approach, the
quantity of accrued contamination could be detected. In addition, tests were conducted on stocks of
Chromosorb that had been contaminated with glues, preservatives, and environmental contami-
nants. In such cases, it is possible to monitor the removal of contamination during treatment.

P-105

ALTERNATIVE METHODS FOR CELLULOSE PREPARATION FOR AMS MEASUREMENT

M NÏmec1,2, L Wacker2, I Hajdas2
1. Dept. of Chemistry and Biochemistry, University of Bern, Switzerland.
2. Ion Beam Physics, ETH Zurich, Switzerland.

The main methods applied to clean plant material for radiocarbon dating are not compound-specific
and generally remove only the easily exchangeable components by an acid-base-acid sequence and
additional optional steps like Soxhlet extraction of resins and oxidative bleaching with NaClO2. The
 Abstracts 135

products are normally clean enough for standard radiocarbon measurement, but in some cases it is
desirable to have pure cellulose as the main building component of plant cell walls. It stays
unchanged and immobile over long time ranges, representing the original plant material. In this
work, two specific but still simple methods were tested to separate the cellulose from wood.
The first procedure is based on the industrial viscose process modified for small samples. The milled
wood sample is first treated several times with NaOH. CS2 is added and alkali cellulose forms a cel-
lulose xanthate, which is then dissolved in the NaOH forming an orange solution. The cellulose is
precipitated with hot H2SO4, washed, and dried.
The second procedure is based on the cellulose dissolution in ionic liquid 1-butyl-3-methylimidazo-
lium chloride [C4mim]Cl, recently reported in literature, but still not used in sample preparation pro-
cedures for radiocarbon dating. In the experiments, the milled wood was added to the ionic liquid
[C4mim]Cl at 110 °C, which dissolves the cellulose. The purified cellulose is precipitated again by
simply adding a water-acetone mixture.
The cellulose samples from both processes were graphitized in the AGE (Automated Graphitization
Equipment) and measured at MICADAS (MIni CArbon DAting System) at ETH Zurich to deter-
mine the possible contamination of the cellulose during the procedures. The procedures and the
results will be presented in detail.

P-106

IMPROVEMENTS ON CO2 EXTRACTION AND PURIFICATION USING AUTOMATED SYSTEM

AND FIRST ATTEMPTS FOR GRAPHITIZING FEW μg OF CARBON AT LMC14 IN SACLAY,
FRANCE
J-P Dumoulin, C Moreau, D Bavay, I Caffy, C Comby, E Delque-KoliÊ, S Ferkane, S Hain,
A Quiles, V Setti, C Souprayen
LMC14 UMS 2572, Bat 450 porte 4E, CEA Saclay, 91191 Gif sur Yvette, France.

Two main technical improvements are discussed in this poster. First, we present the recent develop-
ments on CO2 extraction and purification using an automated extraction line initially developed and
described by Hatté et al. (2003). Those lines have been reconstructed for the LMC14-ARTEMIS
radiocarbon facility. The automated extraction lines have been developed to extract carbon from
charcoal, burned woods, peats, and plant remains (macro residuals). However, purification of CO2
was insufficient to extract carbon from sediment organic matter and consequently samples did not
properly reduce to graphite. This is because the on-line burning of such samples results in an ineffi-
cient contact with the silver layer and large internal temperature gradients. We have investigated the
technical realization to work with sealed tubes, which provide a complete burning of the samples
and a better sulphur removal for optimized graphitization. This cannot be done in a fully automated
procedure. Therefore, we have constructed a line to seal tubes at low vacuum (10–6 mbar). The tubes
are then transferred to an automated extraction line with a newly developed inlet system. This
allows preparing numerous samples manually that can be extracted afterwards using the automated
line (semi-automatic procedure), which implies an efficient handling and less time-consuming gas
production procedures.
On the other hand, to answer the increasing demand for measuring very small samples (<100 μg of
carbon), we have tested our graphitization line with small aliquots of backgrounds and standards
136 Abstracts

produced in large volumes. The first results obtained for samples of carbon from 100 to 10 μg will
be presented and the influence of graphitization and measurements parameters will be discussed.

REFERENCE
Hatté C, Poupeau J-J, Tannau J-F, Paterne M. 2003. Development of an automated system for preparation of organic samples.
Radiocarbon 45(3):421–30.

P-107

CHARACTERIZATION AND ONGOING DEVELOPMENT OF THE ANSTO AMS RADIOCARBON

SMALL MASS H2/Fe GRAPHITIZATION LINES
Alan Williams and Simon Varley
Australian Nuclear Science and Technology Organisation, PMB1, Menai, NSW 2234, Australia.

The ANSTO AMS radiocarbon graphitization laboratory was originally established with a set of 8
graphitization lines, using Zn/Fe and having volumes around 10 to 13 mL. As the demand for
smaller sample masses increased, 12 smaller volume (~3.5 mL) H2/Fe graphitization lines were
developed. These lines proved to be versatile, as the operating volume could be varied by changing
the volume of a removable cold finger, allowing small mass samples of <100 μg C up to ~3 mg C to
be graphitized reliably. Following the success of these lines, the original Zn/Fe lines were replaced
in 2004 with a second set of 12 H2/Fe lines.
In 2006, following the serial failure of the original gauges used for the 3.5-mL lines we had the
opportunity to decrease the reaction volume further and thus reduce the graphitization mass limit.
This poster will describe the smaller volume (~2.5 mL) versions, of the original H2/Fe graphitization
lines; these new lines were developed using stainless steel diaphragm pressure sensors. This devel-
opment was successful in extending our lower working mass limit down to ~10 μg C at ~50% graph-
itization efficiency, while still retaining the flexibility to graphitize up to ~3 mg C reliably. We will
present the efficiency data and the characterization tests that these lines underwent, following their
rebuilding as ~2.5-mL lines.
In addition, we will present the development of new graphitization lines, which are in the process of
construction. These comprise 3 very small volume (~0.9 mL) H2/Fe graphitization lines (using
ceramic diaphragm pressure sensors), with the primary objective of minimizing, and stabilizing, the
absolute mass of added extraneous carbon and stabilizing its pMC value. To assist in this objective,
we are also developing revised methods for our sample combustion and transfer procedures, and the
development of a modified transfer line. Preliminary results from this work will be presented.

P-108

A COMPARISON OF CELLULOSE EXTRACTION AND ABA PRETREATMENT METHODS FOR

AMS 14C DATING
J Southon and A Magana
Keck Carbon Cycle AMS Facility, Earth System Science Dept, University of California, Irvine, CA 92697-3100, USA.
 Abstracts 137

We have compared AMS radiocarbon results on wood samples at or near the limit of 14C dating, pre-
treated with a standard acid-base-acid (ABA) protocol, with those obtained from alpha-cellulose
prepared from the same samples by Soxhlet extraction, acid-chlorite bleaching, and treatment with
strong NaOH. We are also investigating variations of the alpha-cellulose extraction procedure, in an
effort to streamline the method and reduce processing times. These include the use of ABA as a sub-
stitute for Soxhlet treatment and the use of strong (fully dissociated) vs. weak acids in the bleaching
step. Initial results suggest that for at least some samples near the dating limit, ABA-treated wood
produces ages as old as those on cellulose; but as other researchers have found, a subset of “diffi-
cult” samples exists for which more rigorous extraction methods are clearly required.

P-109

RADIOCARBON IN CLAY MINERALS: A STEPPED-COMBUSTION APPROACH

G S Burr1,2, Jack McGeehin3, Yue-Gau Chen2, James Mayer1, Alex Leonard1, Li Cheng1, A J T Jull1
1. NSF-Arizona AMS Facility, University of Arizona, Department of Physics, Tucson, AZ, USA.
2. National Taiwan University, Geosciences Department, Taipei, Taiwan.
3. USGS National Center, Radiocarbon Laboratory, Reston, VA USA

This study describes the results of stepped-combustion experiments on clay minerals, and clay-rich
Holocene paleosols collected from Taiwan. Earlier studies have focused on the radiocarbon content
of CO2 released during combustion at two temperatures, 400 °C (low temperature) and 1100 °C
(high temperature). Those results showed that the radiocarbon ages from low temperature experi-
ments on the <64-μm size fraction produced ages that are consistent with charcoal ages in the same
horizon, albeit with larger uncertainties. It was also shown previously that radiocarbon dates of
humic acids from the <64-μm fraction were consistent with charcoal ages as well.
Here, we examined a new site, and focused on the reproducibility of the method by sampling a
paleosol along a horizontal transect ~30 meters long and at two depths. The samples produced con-
sistent results for both the low temperature and humic acid fractions, and are in agreement with our
previous findings.
We then performed stepped-combustion experiments on pure clay samples (illite, montmorillonite,
smectite, ripidiolite, and kaolinite) in an effort to relate these findings to the clay-rich paleosols.
Stepped-combustion experiments were conducted at 50 °C intervals on the clays, from 300 °C to
1100 °C, and the 14C and 13C isotopic composition were determined. The results show that the radio-
carbon content can be quite sensitive to temperature fraction and may help us refine the temperature
ranges used in the method.

P-110

DEVELOPMENT OF AN AUTOMATIC SAMPLING UNIT FOR MEASURING RADIOCARBON

CONTENT OF GROUNDWATER
R Janovics1, M Molnár2, I Futó2, L Rinyu2, É Svingor2, M Veres3, I Somogyi4, I Barnabás5
1. University of Debrecen-ATOMKI, Debrecen, Hungary.
138 Abstracts

2. Hertelendi Ede Laboratory of Environmental Studies, MTA ATOMKI, Debrecen, Hungary.

3. Isotoptech Zrt, Debrecen, Hungary.
4. Iontech Kft, Litér, Hungary.
5. Public Agency for Radioactive Waste Management (PURAM) of Hungary, Budaörs, Hungary.

Continuous monitoring of uncontrolled tritium emission from Paks nuclear power plant (NPP) has
been carried out since the start of the NPP. In order to control the release of fission and corrosion
products as well as 14C, automatic samplers were designed and installed into 20 observation wells
in the vicinity of the reactor units. The automatic samplers contain two columns filled with anion
and cation exchanger resins. The samplers are working 2 m below the groundwater level. Water
pumps fed by an accumulator ensure that water flows through the ion-exchanger columns. After a
two-month working period, the trapped ions are eluted from the resins. The activity of the gamma
emitters, 14C, 90S, and transuranium elements are measured in the dried elute, while the tritium activ-
ity is measured in the collected water phase.
The sampling method is based on the principle of ion exchange. The anion-exchanger resin changes
the dissolved HCO3– ions to OH– ions. The radiocarbon content of the carbon released in the form
of CO2 from the dissolved HCO3– of the resin is determined by the conventional beta-counting tech-
nique. The amount of the generated CO2 and the activity concentration of 14C were measured after
exposure of carbonate samples gained by the automatic sampler and by conventional barium-chlo-
ride precipitation. Differences were found between the results of the two methods in both the
amount of CO2 and 14C content. The amount of CO2 gained by ion exchange from the resin was
higher by 10–20% than that resulting from the conventional method. The 14C content of the carbon-
ate dissolved from the resin was about 10% lower than the 14C content of the barium carbonate pre-
cipitate. The difference is well-defined; therefore, it can be factored into a correction.

P-111

CLOSED-TUBES PREPARATION OF GRAPHITE FOR HIGH-PRECISION AMS RADIOCARBON

ANALYSIS
Irka Hajdas1, Danuta MichczyÒska1,2, Georges Bonani1, Mantana Maurer1, Lukas Wacker1
1. ETH Zürich, Laboratory of Ion Beam Physics, Radiocarbon Dating, Schafmattstrasse 20, 8093 Zürich, Switzerland.
2.Silesian University of Technology, GADAM Centre of Excellence Krzywoustego 2, 44-100 Gliwice, Poland.

Recent technical developments in accelerator mass spectrometry (AMS) opened opportunities for
new applications. Moreover, high-resolution records of the past changes require high-resolution
chronologies, i.e. sampling for 14C dating. As a result, the field of applications is rapidly expanding
and the number of radiocarbon analyses is growing rapidly. Nowadays, dedicated 14C AMS
machines have a great capacity for analysis, but in order to keep up with the demand for analysis and
provide the results as fast as possible, a very efficient method of sample preparation is required.
Sample preparation for 14C AMS analysis consists of two steps: separation of relevant carbon from
the sample material and preparation of graphite for AMS analysis.
In the early 90s, a method was proposed (Vogel 1992) and applied by a few laboratories, mainly for
environmental or biomedical samples, which can be easily and flexibly applied to produce graphite.
In this procedure, a sample of CO2 is frozen into the tube containing preconditioned Zn/TiH2 and Fe
catalyst. Torch-sealed tubes are then placed in the stepwise-heated oven at 500–550 ºC and left to
 Abstracts 139

react for several hours. The greatest problem is the lack of control of the completeness of the reac-
tion and considerable fractionation. However, recently reported results (Xu et al. 2007) suggest that
high-precision dating using graphite produced in closed tubes might be possible. We will present
results of radiocarbon dating on a set of standards and secondary IAEA standards to demonstrate to
what level this method can be used for high-precision radiocarbon dating.

REFERENCES
Xu X, Trumbore SE et al. 2007. Modifying a sealed tube zinc reduction method for preparation of AMS graphite targets: re-
ducing background and attaining high precision. Nuclear Instruments and Methods in Physics Research B 259(1):320–9.
Vogel JS. 1992. Rapid production of graphite without contamination for biomedical AMS. Radiocarbon 34(3):344–50.

P-112

DOES PERFORMANCE OF ZINC- AND HYDROGEN-REDUCED GRAPHITE DIFFER?

Ron W Reimer1, Kerry R Allen1, David W Beilman1, Susan E Crow1, Stewart J Fallon2,
Paula J Reimer1
1. 14CHRONO Centre for Climate, the Environment and Chronology, School of Geography, Archaeology and Palaeoecol-
ogy, Queen’s University Belfast BT7 1NN, UK
2. Radiocarbon Dating Facility, Research School of Earth Sciences, Mills Road, The Australian National University, Can-
berra, ACT 0200 Australia

The 14CHRONO Centre at Queen’s University Belfast (QUB) and the Research School of Earth
Sciences at the Australian National University Canberra (ANU) recently conducted an intercompar-
ison of their graphite production facilities. The Belfast laboratory uses both zinc and hydrogen
reduction rigs for producing graphite and the ANU laboratory uses hydrogen reduction. Each labo-
ratory took splits of their graphite and sent one half to the other laboratory to be measured, keeping
back the other half for measuring at home. Both laboratories have reasonably new facilities (<2
years old), with both the hydrogen reduction rigs being based on the CAMS/CCAMS methodolo-
gies. The properties of the graphite were determined from measurements on an NEC Single Stage
Accelerator Mass Spectrometer at ANU and an NEC compact model 0.5MV at QUB. We find inter-
esting comparisons between the zinc- and hydrogen-reduced graphite.

P-113

RADIOCARBON DATING OF WATERLOGGED WOODS TREATED WITH A CONSERVATION

MATERIAL PEG: TEST OF PEG REMOVAL BY USING DENDRODATED ARCHAEOLOGICAL
WOOD REMAINS
Hiroshi Nishimoto1, Hideki Takada2, Shinya Nakamura3, Katsuhiko Kimura4, Hirotaka Oda5,
Toshio Nakamura5
1. Graduate School of Environmental Studies, Nagoya University, Aichi, Japan.
2. The Noto-Town Board of Education, Mawaki Noto-cho, Ishikawa, Japan.
3. Department of Cultural Properties and Heritage, Kanazawa Gakuin University, Ishikawa, Japan.
4. Faculty of Symbiotic Systems Science, Fukushima University, Fukushima, Japan.
5. Center for Chronological Research, Nagoya University, Aichi, Japan.
140 Abstracts

Polyethylene glycol (PEG) is an effective conservation substance and commonly used for stabiliza-
tion of the size and shape of waterlogged woods that are excavated from many archaeological sites.
When radiocarbon dating of the conserved woods is required, we must eliminate PEG completely
from the woods, because it is made from coal or oil. In theory, PEG can be dissolved easily in water,
but there is no acceptable result indicating that a normal decantation process with even hot water can
completely extract PEG in previous studies. It is also important to check whether removal of PEG
from any woods including different species and degradation degree is possible. In order to estimate
the degree of PEG elimination with water from sample woods, we tried elimination of PEG with the
normal decantation process at around 70–80 °C from known-age test samples prepared from water-
logged woods, and performed AMS radiocarbon dating of the test sample remains recovered after
treatment. Another method to eliminate PEG, a Soxhlet extractor with water solution, was
employed. The extractor can remove the conservation substance more efficiently than the normal
decantation method. Results of radiocarbon dating of test samples treated by the two methods were
compared. In addition, the PEG elimination process was applied to chestnut timber remains exca-
vated at the Mawaki archaeological site, Japan. The timbers impregnated with PEG for the conser-
vation about 10 years ago were washed by the above two methods, and dated by AMS. To verify the
estimated dates, the timbers were also cross-dated dendrochronologically with other timbers, which
were previously dated by 14C wiggle-matching precisely.

P-114

STUDIES ON THE PREPARATION OF SMALL 14 C SAMPLES WITH AN RGA AND 13 C-

ENRICHED MATERIAL
Peter Steier, Florian Handle, Jakob Liebl, Walter Kutschera, Roswitha Avalos Ortiz,
Eva Maria Wild
Vienna Environmental Research Accelerator (VERA), Faculty of Physics, Isotope Research, University of Vienna,
Währinger Straße 17, A-1090 Vienna, Austria.

The minimum size of radiocarbon samples for which reliable results can be obtained is in many
cases imposed by laboratory contamination. Efforts to reduce the sample size limit down to a few µg
carbon require comprehensive systematic investigations (Drosg et al. 2007). The average laboratory
blank is usually assessed by a complete series of AMS measurements on samples of different known
carbon mass and 14C/12C content. The yield of sample pretreatment and combustion is calculated
from the CO2 pressure measured in the graphitization reactor. The yield of the graphitization reac-
tion is estimated from the pressure change during the reaction.
We have introduced additional methods that work faster and provide more reliable results. An RGA
(residual gas analyzer) is used to study the combustion and graphitization reactions and to achieve
quasi-quantitative results despite the difficulties mentioned e.g. in McNichol et al. (1992). We could
optimize the reaction kinematics at small CO2 pressures and could identify detrimental side reac-
tions. Knowing the composition of the residual gas allows a reliable judgment on the completeness
of the graphitization.
Further, we use isotopically pure 13C as test material, which offers significant advantages: 1) the
measurement of 12C/13C is possible online with the RGA, which significantly reduces turnaround
 Abstracts 141

times; 2) since the 12C/13C of the laboratory contamination will show the natural isotopic ratio both
the yield and the contamination can be determined from a single sample.

REFERENCES
[1] Drosg R, Kutschera W, Scholz K, Steier P, Wagenbach D, Wild EM. 2007. Treatment of small samples of particulate or-
ganic carbon (POC) for radiocarbon dating of ice. Nuclear Instruments and Methods in Physics Research B 259(1):340–4.
[2] McNichol AP, Gagnon AR, Jones GA, Osborne EA. 1992. Illumination of a black box: analysis of gas composition during
graphite target preparation. Radiocarbon 34(3):321–9.

P-115

A NEW LINE FOR PREPARATION OF GRAPHITE TARGET AT XI’AN AMS CENTER

Xuefeng Lu1,2, Weijian Zhou1,2, Wennian Zhao2, Peng Cheng1,2, Hua Du1,2
1. Institute of Earth Environment, CAS, Xi’an 710075, China.
2. Xi’an Acceleration Mass Spectrometry Center, Xi’an 710054, China.

A new line for the preparation of graphite targets for accelerator mass spectrometry (AMS) radio-
carbon dating has been built in the Xi’an AMS Center. The new graphitization system use H2/Fe
reduction of CO2 and calcium chloride anhydrous to remove the water during reduction. The quality
of our line has been tested in series of AMS 14C measurements of background and Chinese sucrose
carbon standard samples. Most background contamination was introduced during combustion and
from dryer reagent, which has been greatly reduced. A new method of cleaning dryer reagent used
in sample processing, by baking them in a vacuum system, is described. The technical details of this
system, the contamination reduction, and the reproducibility tests are discussed.

RADIONUCLIDE TRACERS

P-116

14C IN BIOLOGICAL SAMPLES AND IN THE ATMOSPHERE IN THE VICINITY OF THE KRŠKO

NUCLEAR POWER PLANT (SLOVENIA)

B ObeliÊ1, I Krajcar BroniÊ1, J BarešiÊ1, N HorvatinËiÊ1, B Breznik2

1. Rudjer BoškoviÊ Institute, BijeniËka 54, HR-10000 Zagreb, Croatia.
2. Krško Nuclear Power Plant, Vrbina 12, Krško, Slovenia.

Monitoring of 14C in biological material and in the atmospheric CO2 in the area of Krško nuclear
power plant (NEK) in Slovenia has been performed since 2006. Sampling campaigns of biological
samples (apples, corn, cereals, grass) from the sites in the immediate vicinity of the plant, as well as
at a control site 10 km away, were performed twice a year. Carbon 14C activity was measured by
direct absorption technique in a Quantulus 1220 LSC. Atmospheric CO2 was collected in bimonthly
periods at two locations inside the NEK area by absorption of CO2 in saturated NaOH. The CO2
142 Abstracts

recovered from the carbonate was used for benzene synthesis, and used for subsequent measurement
in LSC.
Most of the locations in the immediate vicinity of NEK showed slightly higher 14C activities in
plants than the control site. Spatial distribution of 14C activities around NEK showed the depen-
dence on the distance and on the direction. In all campaigns, the highest activities were obtained at
the locations that coincided with the most pronounced wind directions. At the control site, the mea-
sured activities were more-or-less constant. Different biological materials taken at the same location
gave the same 14C activity.
14C activity in atmospheric CO2 at the location inside the NEK area, which lies in the dominant wind
direction, was always higher than at the other location, where no evident deviation from natural 14C
activities was observed during regular operation of NEK. However, during the refueling outage peri-
ods and 2–3 months after it, higher 14C activities were detected at both sites.
Based on the measurements, an assessment of the effective dose to the public could be obtained. The
dose received by ingestion of food (fruits and vegetables) grown in the vicinity of the Krško nuclear
power plant resulted in the contribution to the total dose of less than 1 μSv for most exposed indi-
viduals.

P-117

RADIOCARBON IN THE BRATISLAVA AIR: FORTY YEARS OF INVESTIGATIONS

P P Povinec, A Šivo, K Holý, J Šimon, M Richtáriková
Comenius University, Faculty of Mathematics, Physics and Informatics, Department of Nuclear Physics and Biophysics, SK-
842 48 Bratislava, Slovakia.

Radiocarbon variations with typical maxima in summer and minima in winter have been observed
in the atmospheric CO2 of Bratislava (Slovakia) since 1967. While the earlier Δ14C variations were
mostly due to stratospheric-tropospheric mixing, the recent ones were due to 14CO2 releases from
the Bohunice nuclear power plant (situated about 60 km NE from Bratislava), as well as due to CO2
from combustion of fossil fuels. The observed Δ14C variations with the attenuating amplitudes and
decreasing mean values have maxima in summer and minima in winter mainly caused by large
emissions of fossil CO2 during winter months. During the 1980s and early 1990s, the Δ14C values in
heavily polluted atmosphere of Bratislava were up to 100‰ and at Bohunice up to 50‰ lower than
the Δ14C values represented by the European clean air measured at the Jungfraujoch monitoring sta-
tion (Levin and Kromer 2004). Deep Δ14C minima were observed mostly in winters, however, in a
few cases also during summer. Very negative δ13C values measured in these samples (e.g. –21.9‰
in July 1993) suggest that the Bratislava air was polluted by fossil CO2. After 1993, the Δ14C values
were similar at both Bratislava and Bohunice sites, and from 2003 they were close to the European
clean air levels.
 Abstracts 143

P-118

GROWTH RATE OF ANTARCTIC MOSSES DERIVED FROM BOMB RADIOCARBON

David Fink1, Quan Hua1, Laurence J Clarke2, Sharon A Robinson3
1. Australian Nuclear Science and Technology Organisation, Menai, NSW 2234, Australia.
2. School of Biological Sciences, Flinders University, Adelaide, SA 5001, Australia.
3. Institute for Conservation Biology, University of Wollongong, NSW 2522, Australia.

At the previous Radiocarbon Conference in Oxford, 2006, we reported our preliminary results on
the use of bomb radiocarbon for the determination of growth rates of two moss species collected
from east Antarctica in January 2005. Here, we present an update of this study for four different
moss species (Bryoerythrophyllum recurvirostre, Bryum pseudotriquetrum, Schistidium antarctici,
and Ceratodon purpureus) from the Vestfold Hills and Windmill Islands in east Antarctica.
Samples were collected as a “core” from moss turfs. Approximately 10–20 of the longest shoots
were separated from each core and cut into 3-mm sections for AMS 14C analysis. The samples were
cleaned with dilute HCl acid, then combusted to CO2 and converted to graphite. AMS radiocarbon
measurements have been carried out using the STAR facility at ANSTO. The 14C profile in most
cores showed both the rising and falling limbs of bomb radiocarbon, implying that these mosses
began growing over 50 years ago. Our results indicated that these Antarctic mosses have grown
slowly with average rates ranging from 0.6 to 1.3 mm yr–1. This study is the first to determine the
growth rates of Antarctic mosses over a time period spanning decades rather than individual or mul-
tiple field seasons. The results also give us an opportunity to investigate whether temporal variations
in moss growth rates correlate with climatic change in Antarctica.
This work was supported by Ainse grants 04/068, 05/142P & 06/155 and the ANSTO CcASH
project.

P-119

BOMB-PULSE DATING OF HUMAN MATERIAL: MODELING THE INFLUENCE OF DIET

Elisavet Georgiadou and Kristina Stenström
Lund University, Department of Physics, Division of Nuclear Physics, Box 118, SE-221 00, Lund, Sweden

Atmospheric testing of nuclear weapons in the late 1950s and early 1960s almost doubled the
amount of 14C in the atmosphere. The resulting 14C “bomb pulse” has proven to provide useful age
information in e.g. forensic and environmental sciences, biology, and geosciences. The technique is
also currently being used for retrospective cell dating in man, in order to provide fundamental
insight about the rate of formation of new cells in the human body.
Bomb-pulse dating relies on precise measurements of the declining 14C concentration in atmo-
spheric carbon dioxide collected at clear air sites. However, the matter is complicated by in some
cases significant local variations in the 14C specific activity of air and foodstuff, caused by natural
processes as well as by various human activities. As 14C enters the human body mainly through the
diet, variations of the 14C concentration in foodstuff need to be considered. The marine component
of the diet is believed to be of particular importance, due to the non-equilibrium in 14C specific activ-
ity between the atmosphere and aquatic reservoirs during the bomb pulse.
144 Abstracts

We review the 14C concentration in different marine environments, as well as marine foodstuffs dur-
ing the bomb pulse era. 14C concentrations in some other foodstuffs are also summarized. Conclu-
sions about possible variations in the 14C concentration in the human body during the bomb pulse
are drawn.

P-120

14CFLUXES ACROSS THE WATER-ATMOSPHERE INTERFACE IN A EUTROPHIC RIVER DUR-

ING RADIOACTIVE LIQUID RELEASES FROM NUCLEAR POWER PLANTS
Michel Fontugne1, Gwenael Abril2, Arnaud Bacon1, Dominique Baumier1, Marc Vincent
Commarieu2, Olivier Connan3, Henri Etcheber2, Pierre Germain3, Christine Hatté1, Didier
Hébert3, Philippe Jean-Baptiste1, Caroline Lozay4, Denis Maro3, Aline Olivier4, Martine Paterne1,
Vincent Podevin4, Dominique Poirier2, Marianne Rozet3, Lionel Tenailleau4, Françoise Siclet5
1. Laboratoire des Sciences du Climat et de l’Environnement (UMR1572 CNRS/CEA/UVSQ), Domaine du CNRS, 91198-
Gif sur Yvette Cedex, France.
2. Département Géologie Océanographie, EPOC (Environnements et Paléoenvironnements OCéaniques), UMR 5805 CNRS/
Université Bordeaux 1, Avenue des Facultés, 33405 Talence Cedex, France.
3. Laboratoire de Radioécologie de Cherbourg-Octeville (IRSN/DEI/SECRE/LRC), Rue Max-Pol Fouchet, 50130 Cher-
bourg-Octeville, France.
4. Groupe d’Etudes Atomiques, Ecole des Applications Militaires de l’Energie Atomique, Marine Nationale, BP19-50115
Cherbourg Armées, France.
5. EDF R&D Dept LNHE, 6 Quai Watier, 78400 Chatou, France.

To improve our understanding of 14C loss from eutrophic rivers following radioactive discharges,
field measurements of water and air 14C concentrations were performed during three seasons (spring,
summer, and fall) along the Loire River, downstream three nuclear power plants. Water and air pCO2,
as well as water-air CO2 fluxes, were also measured in order to estimate gas transfer velocity. In rela-
tion to the extent of phytoplankton growth, the Loire River behaved as a sink of CO2 in spring (May
2005 F(CO2) = –30 mmol m–2 d–1) and summer (Sept 2004: F(CO2) = –60 mmol m–2 d–1) and as a
source of atmospheric CO2 during fall (Nov 2005: F(CO2) = +30 mmol m–2 d–1). Downstream of the
power plants, our experiments clearly show that, whatever the season, CO2 at the liquid-gas boundary
is enriched in 14C, in accordance with a diffusion of 14CO2 along its own concentration gradient.
During the summer, when the Loire River is globally a sink for CO2, 14C is re-equilibrated at the
water-air interface resulting in relative 14C enrichment of the atmospheric CO2, the concentration of
which is lowered by photosynthetic activity. Near the air-water interface, such a process leads
causes the weakest releases and a decrease of 14C air radioactivity with 14C-enriched CO2. For
higher 14C activity in water (strong releases), calculations indicate a positive 14C flux to the atmo-
sphere. During winter, the relative 14C enrichment is due to a 14CO2 flux to the atmosphere propor-
tionally higher than the CO2 flux.
 Abstracts 145

P-121

14C AND TRITIUM LEVELS ALONG ROMANIAN LOWER DANUBE RIVER

Carmen Varlam1, Ioan Stefanescu1, Stela Cuna2, Irina Vagner1, Ionut Faurescu1
1. Institute of Cryogenics and Isotopic Technologies, Ramnicu Valcea, Romania.
2. Institute of Isotopic and Molecular Technologies, Cluj Napoca, Romania.

The Danube River is the second largest river in Europe, lying west of the Black Sea in central and
southeastern Europe. The Lower Danube Basin covers the Romanian-Bulgarian sub-basin down-
stream of Cazane Gorge and the sub-basins of the Siret and Prut rivers. About 65% of the Lower
Danube Basin forms the natural border between Romania and Bulgaria. Two important nuclear
objectives for both countries are found in this region: Kozloduy nuclear power plant (NPP) and Cer-
navoda NPP. Taking into account the future development of the Romanian nuclear objective and
knowing the nuclear activity developed along Danube River (over 25 nuclear reactors), knowledge
of the present condition of tritium and 14C concentration levels becomes a necessity. Therefore, an
extensive monitoring program for these isotopes has been developed along the Romanian sector of
the Danube River Basin, starting with the entrance of the Danube River in Romania and ending with
the three branches of the Danube Delta. The tributaries from this sector (Cerna, Jiu, Olt, and Arges)
were also included in this monitoring program. Four sampling campaigns took place during 2006–
2008. The mean tritium concentration for Danube water in monitored areas was 15.1 ± 2.4 TU with
higher values for two locations, downstream NPP Kozloduy and downstream NPP Cernovoda. The
same behavior was recorded for the 14C level in DIC of Danube water, with a mean value of 99.7 ±
2.7 pMC and higher values for the two locations downstream the nuclear reactors. Despite the
nuclear activity in the observed areas, tritium and 14C activities present slightly higher values for
specific locations without any influence in the catching river.

P-122

RADIOCARBON MONITORING IN THE CZECH REPUBLIC AND HUNGARY

I SvÏtlík1, M Molnár2, É Svingor2, T Bujtás3, M Váòa4, P Rulík5
1. Nuclear Physics Institute AS CR, Prague, Czech Republic.
2. Nuclear Research Institute HAS (ATOMKI), Debrecen, Hungary.
3. Paks nuclear power plant, Paks, Hungary.
4. Czech Hydrometeorological Institute, Prague, Czech Republic.
5. National Radiation Protection Institute, Prague, Czech Republic.

In the Czech Republic and Hungary, monitoring of atmospheric 14CO2 is performed at several loca-
tions. The purposes of this monitoring network are (1) quantification of actual 14C activity levels in
the areas influenced by regional Suess effect and (2) determination of local anthropogenic influenc-
ing caused both by 14C releases from nuclear power plants (NPPs) and by fossil fuel combustion. It
can be assumed that each area in central Europe is affected by a local and regional Suess effect.
Resulting activities of atmospheric 14CO2 were compared with long-term extrapolation of available
data from other European monitoring stations to estimate the local and regional components of the
Suess effect.
146 Abstracts

In the surroundings of NPPs 14C activity surplus can be partly compensated by influence from local
sources of fossil carbon. Such a “compensation” effect is eliminated in the case of 14CO2 activity
concentration. This parameter is given only by the 14C amount in a volume unit of air with normal-
ized conditions (or in weight unit of dry air) and is not influenced by the presence of fossil CO2. To
calculate activity concentration of atmospheric 14CO2, it is necessary to measure both 14CO2 activity
and concentration of CO2. Likewise, additional parameters including temperature, pressure, and
humidity of sampled air are necessary. In the Prague-Bulovka area (Czech Republic), monitoring of
activity concentration experimentally was launched during 2001 using volumetric determination of
CO2 concentration.
This presentation will discuss the results of our monitoring, including a brief comparison with pub-
lished data from other monitoring stations.

P-123

EFFECTS OF RAINFALL ON CARBON ISOTOPES OF POC IN THE TESHIO RIVER, NORTHERN

JAPAN
Takafumi Aramaki1, Seiya Nagao2, Yo-hei Nakamura3, Masao Uchida1, Yasuyuki Shibata1
1. National Institute for Environmental Studies, Japan.
2. Kanazawa University, Japan.
3. Hokkaido University, Japan.

Knowledge of organic carbon dynamics in river basins is required to study the role of rivers in glo-
bal biogeochemical cycles and organic carbon flux. As a large amount of suspended solids wash out
from river basins to coastal regions during rainfall, it is considered that a large amount of organic
matter and nutrients is also carried to coastal regions. Therefore, observational study on the trans-
port behavior of organic matter in river basins has been focused on rainfall events. The objective of
this study is to investigate the behavior of particulate organic carbon (POC) in a river before and
after rainfall using carbon isotopic signatures such as Δ14C and δ13C. As the first step, we selected
the Teshio River located in Hokkaido, northern Japan, where there is lower turbidity during the year
except for the spring snowmelt period. Water sampling was carried out at a fixed point from 4 to 6
September 2006. During the observation period, rain began in the night of 4 September and cleared
up in the night of the 5th.
Suspended solid samples were concentrated from 45–90 L of river waters by a single-bowl contin-
uous-flow centrifuge technique, and carbon isotopes and C/N molar ratios of organic matter in were
measured. Though the Δ14C of POC before rainfall was –56‰, in the evening of September 5th,
when the turbidity of river water was at its maximum, Δ14C increased to –10‰. Δ14C has a high pos-
itive correlation with C/N molar ratios except for the data from the evening of September 5th. These
results may indicate that organic matter supplied from different sources from before rainfall flowed
into the river because of rising water levels.
 Abstracts 147

P-124

RADIOCARBON CONCENTRATION, CARBON AND OXYGEN STABLE ISOTOPE RATIOS IN

ANNUAL-RING CELLULOSE OF A PINE TREE FROM NAGOYA, CENTRAL JAPAN
Kazuki Hayashi1 and Masayo Minami2
1. Earth and Environmental Sciences, Graduate School of Environmental Studies, Nagoya University, Nagoya 464-8601,
Japan.
2. Center for Chronological Research, Nagoya University, Nagoya 464-8602, Japan.

Nagoya is the fourth largest city in Japan with a population of over 2 million. Nagoya is located in
Aichi Prefecture, central Japan. We analyzed 14C concentration and stable carbon and oxygen isoto-
pic compositions of annual-ring cellulose of a pine tree (Pinus thunbergii) with annual rings from
1955 to 2002, grown on the campus of Nagoya University, in order to investigate the relationships
between the annual tree-ring data and climatic and/or artificial factors in the urban area. The cli-
matic factors used are monthly total values of sunshine duration and precipitation, monthly average
values of day temperature, relative humidity, and vapor pressure in the Nagoya area.
The annual rings of the sample are very wide in the periods from 1960–1970. This remarkable
growth might be related to local environmental changes, not with the regional environment changes,
because the annual-ring width shows no correlation with the climatic factors. The δ13C fluctuations
of the annual rings in the period from 1955 to 1979 are different from those in the period from 1980
to 2002. The results show the δ13C values in the two periods are different. In the period from 1955
to 1979, the δ13C fluctuations are not related with the climatic factors, and no climatic environmen-
tal factors, that is, artificial factors such as air pollution, could effect the δ13C fluctuations. In the
period from 1980 to 2002, on the other hand, the monthly average values of relative humidity are in
good correlation with the δ13C fluctuations of earlywood and latewood fractions, respectively.
The δ18O fluctuations of the annual rings of the sample show negative correlation mainly with rela-
tive humidity. The results indicate that δ18O fluctuations of tree rings originate only from the
weather factors without being affected by the artificial factors.

P-125

TRENDS AND GRADIENTS IN Δ14C OF ATMOSPHERIC CO2 OBSERVED BY THE SCRIPPS

GLOBAL FLASK NETWORK
Heather D Graven1,2, Thomas P Guilderson3,4, Ralph F Keeling2
1. Institute of Biogeochemistry and Pollutant Dynamics, ETH Zurich, Switzerland.
2. Scripps Institution of Oceanography, University of California-San Diego, San Diego, CA, USA.
3. Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, Livermore, CA, USA.
4. Department of Ocean Sciences, University of California-Santa Cruz, Santa Cruz, CA, USA.

High-precision measurement of Δ14C was conducted on atmospheric CO2 samples from the Scripps
Institution of Oceanography’s CO2 Program, producing monthly records at Point Barrow, Alaska;
La Jolla, California; Kumukahi and Mauna Loa, Hawaii, USA; Cape Matatula, Samoa; and Palmer
Station and the South Pole, Antarctica. Analysis was conducted at the Center for Accelerator Mass
Spectrometry at Lawrence Livermore National Laboratory with measurement uncertainty of 1.7‰
for most samples. The observations provide a significant contribution to the characterization of
148 Abstracts

recent patterns in Δ14C of CO2. Δ14C decreased at roughly –5‰ yr–1, yet significant interannual vari-
ability was evident in the trend of Δ14C over 1992–2007 at La Jolla. The most prominent excursion
from a linear or exponential trend occurred in 2000 when Δ14C decreased rapidly (–10‰ yr–1), pos-
sibly due to enhanced deep-water ventilation in the North Pacific Ocean. Mean Δ14C at each station
over 2005–2007 shows that Northern Hemisphere Δ14C was approximately 6‰ lower than Southern
Hemisphere Δ14C. The interhemispheric gradient appears to have shifted since 1988–1989 when the
Northern Hemisphere Δ14C was slightly higher than the South. Using results from the Transcom 3
atmospheric transport modeling experiment, we calculate that fossil fuel emissions growth explains
only 20% of the shift in the interhemispheric Δ14C gradient, indicating that regional 14C exchanges
with the ocean and the land biosphere also changed between 1987–89 and 2005–07. Adjustments in
oceanic and biospheric 14C fluxes over this time period are related to turnover and ventilation rates
that are currently not well understood. Continued observation and investigation of atmospheric Δ14C
should therefore provide new constraints on carbon cycling between the ocean, land biosphere, and
the atmosphere.

P-126

Δ14C OF ATMOSPHERIC CO2 AT POINT BARROW, ALASKA

Xiaomei Xu and Susan E Trumbore
Earth System Science Department, University of California, Irvine, CA 92697, USA.

We report here a continuous, high-precision and high-temporal-resolution Δ14CO2 record obtained

at the Point Barrow Observatory, Alaska (71°N, 157°W) since July 2003. Sample collection was
through the NOAA/CMDL flask sampling network program, which enables us to compare radiocar-
bon data with other trace gases and isotopes, including CO, CO2 mixing ratios and δ13C and δ18O of
CO2. Before 2008, two duplicate samples were collected biweekly. Because of good reproducibility
between flasks, after 2007 one sample was collected weekly. Air samples were pumped into pre-
evacuated 6-L canisters (up to 2 atmospheres pressure) and mailed to UC Irvine. In the laboratory,
CO2 was extracted cryogenically on a vacuum line and converted to graphite using the sealed-tube
Zn reduction method, and measured for 14C at the Keck AMS Facility at the University of Califor-
nia, Irvine. During the collection period, Δ14C of CO2 decreased by ~5‰/year, to ~42‰ in Decem-
ber 2008. We find distinct seasonal cycles for 14C, with a broad minimum around March and a max-
imum in September with an amplitude of ~10‰. This seasonal pattern is highly variable from year
to year. Increasing 14C values may reflect injection of stratospheric air in April and May, and higher
soil respiration with enriched 14CO2 between May to August; rapid declines may be due to reduction
in soil respiration and changes in the poleward advection of fossil fuel burned in the winter months.
Our record adds to the few available records worldwide to provide observational constraints for the
roles of 14C isotope disequilibrium among different reservoirs, and hopefully can enhance our
understanding of the patterns of atmospheric 14CO2 distribution and its seasonal variation.
 Abstracts 149

P-127

RADIOCARBON CONCENTRATION IN TREE RINGS FROM A HIGH EMISSION AREA OF VOL-

CANIC CARBON DIOXIDE
Kazuho Horiuchi1, Takuji Chiba1, Yasuharu Hoshino2, Motonari Ohyama2, Takayuki Tanaka3,
Hikaru Amano4
1. Graduate School of Science and Technology, Hirosaki University, Japan.
2. Botanical Gardens, Tohoku University, Japan.
3. Aomori Research and Development Center, Japan Atomic Energy Agency, Japan.
4. Japan Chemical Analysis Center, Japan.

Local atmospheric concentration of 14C is controlled by hemispheric background (“clean air” con-
centration of 14C), regional exchange rate between stratosphere and troposphere, regional/local
ocean influence, local emission of biospheric CO2, local emission of “fossil” CO2 originating from
dead carbon, and some other sporadic factors. In these factors, the “fossil” CO2 emission is impor-
tant in relation to the global warming issue; thus, the regional/local anthropogenic CO2 emission
from fossil fuel is becoming a growing subject in radiocarbon analyses of plants and their byprod-
ucts (e.g. Hsueh et al. 2007; Palstra et al. 2008). Tree rings have an outstanding advantage for such
studies because temporal information can be retrieved from an offset of their 14C profiles (e.g.
Kromer et al. 2001; Hua et al. 2007). A possible weak point of those profiles is the insufficient infor-
mation concerning the processes of uptake of CO2 from local air into the tree rings.
The purposes of this study are to investigate how volcanic CO2 emission affects the 14C concentra-
tions in tree rings and to obtain certain information on whether we can use this information as a
tracer of local emission of fossil CO2 gas. In addition, unique data on the validity of the 14C dating
for areas of intensive volcanic activity is also obtained. We took tree-ring samples from beech trees
living in Hakkoda caldera, north Japan, where an extremely high emission of volcanic CO2 gas had
been reported (Hernández et al. 2003).
14C analysis was performed by using the 3MV Tandetoron accelerator of Japan Atomic Energy
Agency. The obtained 14C concentrations were, surprisingly, quite near the values of the atmo-
spheric CO2, but slightly (around 1%) lower than those of the control samples taken from a beech
living outside of the caldera. We will discuss these results and their implications.

P-128

RADIOCARBON BASED-SOURCE APPORTIONMENT OF PM10 AEROSOLS AT CAPE HEDO,

OKINAWA, JAPAN
Daishi Handa1, Masao Uchida2, Takemitsu Arakaki1, Hitomi Nakajima1, Hidetoshi Kumata3,
Yasuyuki Shibata2
1. Graduate School of Engineering and Science, University of the Ryukyus, 1 Senbaru Nishihara-cho, Okinawa 903-0213,
Japan.
2. AMS Facility (NIES-TERRA), Environmental Chemistry Division, National Institute for Environmental Studies, 16-2
Onogawa, Tsukuba, Ibaraki, 305-8506, Japan.
3. School of Life Sciences, Tokyo University of Pharmacy and Life Sciences, Horinouchi, Hachioji, Tokyo 192-0392, Japan.

A study was initiated to understand the sources and behaviors of carbon-containing compounds in
the atmospheric aerosols in the East Asian region. As an initial attempt, we collected airborne par-
150 Abstracts

ticulate matter (APM) with a diameter 10 micrometers (PM10) in Okinawa, Japan, and analyzed
radiocarbon content in BC and total organic carbon (TOC) to apportion source of biomass and fossil
fuel combustion in PM10 aerosols. Okinawa is situated ~1500 km south of Tokyo, Japan, 2000 km
southeast of Beijing, China, and 1000 km south of South Korea. Its location is well suited for study-
ing long-range transport of air pollutants in East Asia. Aerosol particles were collected at the Cape
Hedo Atmosphere and Aerosol Monitoring Station (CHAAMS) in Okinawa, Japan. Each sample
was collected for two weeks with a high-volume air sampler at a rate of 1000 L min–1 to collect aero-
sol particles on a quartz filter, which was precombusted before use to eliminate residual organic
compounds. The results showed that during Asian dust events in March and April, both APM and
BC in PM10 aerosols collected in Okinawa were much higher than those of non-Asian dust events.
The percent modern carbon (pMC) of EC and TOC was much smaller for the PM10 aerosols col-
lected during Asian dust events than those of non-dust events. When maritime air mass prevailed,
biomass-originated organic compounds were the major TOC in the aerosols. We will report radio-
carbon data of BC and TOC in PM10 aerosols collected at CHAAMS during March to October 2008.

P-129

ORIGIN AND SPATIAL DISTRIBUTION OF THE ORGANIC CARBON FROM THE RHONE RIVER
IN THE MEDITERRANEAN: A MULTITRACER APPROACH (Δ14C, δ13C)
C Cathalot1, N Tisnerat-Laborde1, R Buscail2, P Kerhervé2, C Treignier1, B Lansard1,
C Rabouille1
1. Laboratoire des Sciences du Climat et de l’Environnement, UMR CEA-CNRS, Bât 12, avenue de la Terrasse, 91198, Gif-
sur-Yvette, France.
2. Centre de Formation et Recherche sur l’Environnement Marin, CNRS - UMR 5110, Université de Perpignan, 52 avenue
Paul Alduy, 66860 Perpignan Cedex, France.

At the continent-ocean interface, the coastal zone receives organic matter and nutrients from the riv-
ers that sustain local primary production. The fate of the organic particles at the Rhône River mouth
is largely constrained by biogeochemical and physical processes. Seasonal variations of environ-
mental condition (T, light, hydrology) and organic matter inputs from the river govern degradation,
transport, and deposition phenomena.
Cruises were performed off the Rhône River to study the export and biogeochemical transforma-
tions of particulate inputs and investigate their spatial distributions over the continental shelf.
Radiocarbon (Δ14C) and stable carbon isotope (δ13C) distributions were measured in suspended par-
ticulate organic carbon and in sediment in the Rhône subaqueous delta and Mediterranean continen-
tal shelf. Carbon isotopes in sediments near the river mouth display the same signature as the mod-
ern and terrestrial riverine material (δ13C = –27‰, Δ14C = 0–100‰). A dilution pattern with old
marine organic matter (δ13C = –22‰, Δ14C = –390‰) is visible offshore, preferentially in a south-
west direction. This spatial sedimentary distribution is most likely linked to the well-documented
main hydrodynamic circulation and sediment deposition in this area. The δ13C signature of the sus-
pended particles is homogeneous from the river delta to the shelf (δ13C = –23.7‰), indicating a
decoupling between the water column and the sediment particles. This suggests the existence of a
pool of particles with long residence time and multiple erosion-deposition cycles over the shelf.
 Abstracts 151

P-130

RADIOCARBON DATING AND DIET-DERIVED OFFSETS FROM GEOTHERMAL SOURCED

FRESHWATER SYSTEMS: A CASE STUDY FROM THE ARCHAEOLOGICAL SITE OF KLIN
YAR, RUSSIAN NORTH CAUCASUS
Tom Higham1, Rebecca Warren1,2, Rachel Wood1
1. Oxford Radiocarbon Accelerator Unit, RLAHA, University of Oxford, UK.
2. Now at University of Glasgow, UK.

The influence of geothermally-derived carbon on the radiocarbon dating of human bone is poorly
understood and has rarely been examined with independent age control. This study identifies a res-
ervoir effect at the archaeological site of Klin Yar in the Russian North Caucasus. This large site
spans the Koban (Iron Age, ~1100–400 BC) to Sarmatian (400 BC–AD 400) and Alanic periods
(~4–7th centuries AD), and contains an estimated 2000–3000 human burials. Human bones from
Sarmatian and Alanic burials that were AMS dated and yielded results that were older than expected
when compared with the dates of terrestrial ungulate bones found in direct archaeological associa-
tion. In some instances, it was possible to measure directly the degree of offset by comparing the
human radiocarbon ages to Byzantine coins of known date. The AMS dating of modern plant, fish,
and water samples demonstrates that the source of the old carbon is likely to be the geothermally-
derived local water, which has an apparent age of several thousand years. Unfortunately, it is diffi-
cult to show this definitively, firstly, because the archaeological record is silent in showing the pres-
ence of fish remains in occupation contexts. In addition, enriched δ15N values of fauna and the pos-
sible presence of millet in the foodchain make the identification of freshwater-derived protein in the
diet difficult.
Due to the highly variable nature of the radiocarbon offset (~0 to 334 years), no suitable average cor-
rection factor is applicable to correct for the human dates at the site. A radiocarbon chronology
based on dates obtained from terrestrial ungulate bone is instead a more reliable indicator of age.

P-131

RADIOCARBON ON-LINE ANALYSIS OF ATMOSPHERIC SAMPLES

N Perron1, S Szidat2, S Fahrni2, L Wacker3, A S H Prévôt1, U Baltensperger1
1. Laboratory of Atmospheric Chemistry, PSI, Villigen, Switzerland.
2. Department of Chemistry and Biochemistry, University of Bern, Switzerland.
3. Institute for Particle Physics, ETH Hönggerberg, Zürich, Switzerland.

Radiocarbon has been recognized in the field of atmospheric chemistry as a powerful tracer that
enables discrimination between fossil and non-fossil sources of carbonaceous aerosol (1). This part
of particulate matter divides into organic carbon (OC) and elemental carbon (EC), which exhibit
very different chemical and physical properties—EC is much more refractory and light-absorbing
than OC—and stem from different sources. Therefore, efforts have been made to isolate them so as
to analyze their radiocarbon content separately.
To reach this goal, the sample preparation described in Szidat et al. (2) is currently being adapted to
a thermo-optical EC/OC analyzer. In this device, a filter punch containing the aerosol is gradually
heated, thus combusting carbonaceous fractions of increasing refractiveness. Meanwhile, the degra-
152 Abstracts

dation of light-absorbing material is monitored by detection of a laser signal through the filter,
which allows to visualize the release of EC.
For each carbonaceous fraction, the resulting CO2 can be trapped cryogenically and sealed in glass
ampoules for later off-line acceleration mass spectroscopy analysis of its radiocarbon content. Since
the whole preparation procedure is time-consuming and 14C information cannot be obtained imme-
diately, injecting the CO2 directly into the AMS for direct on-line radiocarbon measurements
appears of major interest.
Therefore, we have coupled a Sunset EC/OC analyzer (RT 3080, Sunset Laboratory Inc.) with the
MICADAS facility presented in Synal et al. (3, 4) by using the already existing carbon dioxide feed-
ing system. We will discuss the results of the first analysis of standards and blanks.

REFERENCES
(1) Szidat S. 2009. Science 323:470–1.
(2) Szidat S et al. 2006. J. Geophys. Res. 111: D07206; doi:10.1029/2005JD006590.
(3) Synal H-A et al. 2004. Nucl. Instr. and Meth. B 223-224:339–45.
(4) Synal H-A et al. 2007. Nucl. Instr. and Meth. B 259:7–13.

SOILS AND SEDIMENTS

P-132

RADIOCARBON DATING OF THE LAST VOLCANIC ERUPTION OF THE CIOMADUL VOL-

CANO, SOUTHEAST CARPATHIANS
Sz Harangi1, A P Vinkler1, M Molnár2, A J T Jull3, A E Leonard3
1. Department of Petrology and Geochemistry, Eötvös University, 1117 Budapest Pázmány sétány 1/C, Hungary.
2. ATOMKI, Debrecen, Hungary.
3. NSF Arizona AMS Laboratory, University of Arizona, 1118 East Fourth St, Tucson, AZ 85721, USA.

The last volcanic eruption in the Carpathian-Pannonian region occurred within the Ciomadul Vol-
cano, southeast Carpathians. This is a lava dome complex active for about 900 kyr. Following the
effusive activity, at least two major explosive volcanic eruptions occurred, forming two craters
(Mohos and St. Anna). These eruptions resulted in pumiceous pyroclastic fall and flow deposits. In
order to understand the behavior of this volcano and evaluate the possible renewal of the volcanic
eruption, it is crucial to constrain the length of the active and repose periods and know the date of
the last eruption. The available K/Ar and radiocarbon data are controversial and give only a rough
estimate for the active phases. The age data for the last eruption is in the range from 10,000 to
40,000 yr. In this study, we focus on the time of the last volcanic eruptions. We analyzed charcoal
fragments found in pumiceous pyroclastic flow deposits from two localities.
14C dating from the first locality was executed by the GPC technique in the ATOMKI, and the fol-
lowing five dates in the NSF Lab in Tucson by the AMS technique (one repeated dating of the
ATOMKI sample, with excellent match; one resampled charcoal from the first locality, and one sam-
ple from a second locality, and dating of NaOH-soluble fractions coming from sample preparation
for all the three charcoal samples). The radiocarbon data clearly show that the last volcanic eruption
 Abstracts 153

took place ~28,000 BP. In addition, it was not a single eruption as previously assumed, but followed
another violent explosive eruption, which occurred 39,000 BP. These data are consistent with the
different composition of the volcanic products from the two localities.
This study is part of the research project supported by the OTKA grant # K68587.

P-133

NEW RADIOCARBON DATES FROM THE LATE NEOLITHIC TELL SETTLEMENT OF HÓDMEZ-
VÁSÁRHELY-GORZSA, SE HUNGARY
S Gulyás1, P Sümegi1, F Horváth2, M Molnár3
1. University of Szeged, Department of Geology and Paleontology H-6722 Szeged, Egyetem u. 2-6. Hungary.
2. Ferenc Móra Museum, H-6720 Szeged, Roosevelt tér 1-3, Hungary.
3. Laboratory of Environmental Studies, Institute of Nuclear Research of the Hungarian Academy of Sciences, 4025 Debre-
cen, Bem tér 18/C, Hungary.

Understanding the internal chronology of the Late Neolithic Tisza culture and the Neolithic of the
Tisza region is the subject of debate in both Hungarian and international prehistoric research. The
layer sequence of the Late Neolithic Gorzsa tell from SE Hungary offers ideal match points for
determining the successive phases of the Tisza culture. According to the results published so far, in
the Gorzsa sequence the Tisza culture was divided into 4 main phases, the fifth phase representing
the transitional period to the Early Copper Age. Excavations were carried out in 33 profiles covering
about 2% of the original area of the entire settlement. The archaeostratigraphy established was based
on the identification of microhorizons corresponding to settlement levels. Radiocarbon dates pub-
lished so far were created using a pool of various objects of differing microhorizons deriving from
different profiles. However, as archaeological results revealed, the settlement was characterized by
frequent minor spatial shifts during its evolution into a tell complex. Here, we would like to present
a succession of 7 radiocarbon dates deriving from a single profile located on the northeastern flank
of the excavation area. The 7 dates span the entire profile from the uppermost microhorizons down
to the lowermost ones. The new dates were compared with the system of relative stratigraphy men-
tioned above. According to our findings, the evolution of the tell ended earlier in this part of the pro-
file than in other parts, probably as a result of the mentioned spatial movement of people within the
site itself.

P-134

RADIOCARBON-DATED PALEOENVIRONMENTAL CHANGES ON A LAKE AND PEAT SEDI-

MENT SEQUENCE FROM THE CENTRAL PART OF THE GREAT HUNGARIAN PLAINS (CEN-
TRAL EUROPE) DURING THE LAST 25,000 YEARS
P Sümegi1,2, M Molnár3, G Jakab4, G Persaits1, P Majkut1, D G Páll1, S Gulyás1,
A J Timothy Jull5, T Törıcsik1
1. University of Szeged, Department of Geology and Palaeontology, 6701 Szeged, P.O. Box 658, Hungary.
2. Archaeological Institute of the Hungarian Academy of Sciences, 1014 Budapest Úri u. 49, Hungary.
3. Laboratory of Environmental Studies, Institute of Nuclear Research of the Hungarian Academy of Sciences, 4025 Debre-
cen, Bem tér 18/C, Hungary.
154 Abstracts

4. Tessedik Sámuel College 5540 Szarvas Szabadság str. 2, Hungary.

5. NSF Arizona AMS Laboratory, University of Arizona, 1118 East Fourth St, Tucson, AZ 85721, USA.

One of Hungary’s geological and environmental treasures is nestled in the heart of the Great Hun-
garian Plains. The catchment basin of Lake Kolon was subjected to detailed environmental histori-
cal studies starting in 2005. Undisturbed cores taken along transects of the basin were subjected to
detailed sedimentological, paleoecological, and geochemical studies. In order to establish a reliable
timeframe of the lacustrine and marshland sedimentary sequence, 22 samples were subjected to
AMS analyses implemented in the radiocarbon laboratories of Poznan and Tucson. With the new
results in hand, we had the opportunity to elucidate the geological evolution of the area for the past
25,000 years. This sequence is highly outstanding in so far it is probably the most well-dated profile
of the Quaternary from the mentioned area. The new absolute dates enabled the comparison of local
geological evolution of the studied area with those of global climatic changes. As seen from our
findings, the geological evolution of the catchment basin was congruent with major climatic events
during the Pleistocene and the entire Holocene. However, a very peculiar trajectory was identified
for the terminal part of the Pleistocene and the opening of the Holocene regarding the evolution of
the landscape, the vegetation and the fauna of this part of the Great Hungarian Plains.

P-135

14CDATING OF DANUBE DELTA SEDIMENTS: IMPLICATIONS FOR THE DELTAIC SYSTEM

EVOLUTION DURING THE PAST 12 KYR
Gh Oaie1, N Panin1, C Varlam2, A Seghedi3, O G Duliu4
1. Institute of Marine Geology and Geoecology, Bucharest, Romania.
2. Institute of Cryogenics and Isotopic Technologies, Ramnicu Valcea, Romania.
3. Geological Institute of Romania, Bucharest, Romania
4. University of Bucharest, Department of Atomic and Nuclear Physics, Bucharest, Romania

Situated in the northwestern part of the Black Sea, the Danube Delta is the result of the Danube
River-Black Sea interaction during the Upper Pleistocene-Holocene. The deltaic system is divided
into three main units: the Delta Plain area (5800 km2), with fluvial delta plain and marine delta plain
sub-units; the Delta Front (1300 km2); and Danube Prodelta (more than 5500 km2). While the west-
ern part of the delta is formed under the river influence, the eastern part is predominantly influenced
by the sea. During the last 12,000 years, the solid discharge of the river Danube formed large lobes,
named paleodeltas. According to the first radiocarbon dating, the Letea-Caraorman initial spit accu-
mulated between 11,700–7500 BP in the main phase of the deltaic evolution; the main paleodeltas,
as depositional lobes, formed in the following intervals: 9000–7200 BP (St. George I Delta), 7200–
2000 BP (Sulina Delta), 2000 BP to present (St. George II and Chilia deltas), and 3500–1500 BP
(the secondary Cosna-Sinoe Delta). The main objective of the study is to achieve a better knowledge
of the Danube Delta evolution, using new radiocarbon dates on shell layers interbedded in the del-
taic sediments.
Research carried out in the Danube Delta included sampling of sediment from trenches and cores,
digital radiography of cores, selection of sediment samples for grain size, macro- and microfauna,
TLD and radiocarbon datings. Seven samples have been analyzed using the direct absorption
method and liquid scintillation counting (IAEA-C6 reference standard). Final calibrated 14C ages
(OxCal 3.10) were estimated between 3100 and 12,500 BC.
 Abstracts 155

Correlation between different layers with the same age has led to reliable results. The scientific
results were subjected to critical analysis, reservoir correction, and TLD dating in order to define an
age model for the investigated areas.

P-136

RADIOCARBON DATING OF CHARCOALS IN LATERALLY AMALGAMATED FLUVIAL CHAN-

NEL BODIES: THE LATE HOLOCENE ABAK CREEK SUCCESSION (AXUM, NORTHERN ETHI-
OPIA)

M Ghinassi1, M E Fedi2, F Taccetti2, M Sagri3

1. Dipartimento di Scienze della Terra, Università di Padova, via Giotto 1, 35137 Padova, Italy.
2. INFN Sezione di Firenze e Dipartimento di Fisica dell’Università, via Bruno Rossi 1, 50019 Sesto Fiorentino (Fi), Italy.
3. Dipartimento di Scienze della Terra, Università di Firenze, via La Pira 4, Firenze, Italy.

Radiocarbon dating of charcoal fragments is widely used in stratigraphic, geomorphologic, and

geoarchaeological studies on landscape evolution, even in those areas where the basic “cake-like
architecture” of sedimentary successions (e.g. tabular, laterally extensive, and vertically stacked
depositional units) is not commonly respected. An example of these situations is represented by
alluvial successions where overbank fines are not preserved and the product of alluvial plain
dynamic is a sheet-like body made of laterally amalgamated channel-fill deposits (e.g. meander
belt), which can record repeated phases of erosion and deposition in a low accommodation space
setting. Under these conditions, coeval channel-fill units can be separated by older units showing
identical sedimentary features, whereas units separated by erosional surfaces were deposits by the
same channel. The occurrence of relatively scarce exposures, which commonly characterize recent
deposits, can significantly complicate this picture.
The late Holocene fluvial deposits of the Abak Creek (Axum, northern Ethiopia) well represent such
a geological frame. These deposits consist of laterally amalgamated channel-fill bodies formed in a
moderate to low sinuosity river setting. In order to identify possible erosional-aggradational phases
hidden within the same depositional unit, several charcoal samples have been collected from chan-
nel-fill deposits in 10 different localities and dated by accelerator mass spectrometry at the LABEC
laboratory in Florence. Results have shown that the examined depositional unit is indeed uniform
without any gap or “external” phase.

P-137

HIGH-PRECISION DETERMINATIONS OF ERUPTION AGE OF THE TOWADA OFUDO (T-Of)

TEPHRA IN NORTHEAST JAPAN
Yosuke Miyairi1, Hiroyuki Matsuzaki1, Kazuho Horiuchi2
1. Department of Nuclear Engineering and Management, School of Engineering, The University of Tokyo, 7-3-1 Hongo,
Bunkyo-ku Tokyo 113-0032, Japan.
2. Graduate School of Science and Technology, Hirosaki University, Hirosaki, Japan.

A widespread tephra sedimentary layer is often used as a “time marker bed” in geology and geogra-
phy. It means that the age of a tephra layer is used as a criterion to estimate the other layer’s age.
156 Abstracts

Therefore, the establishment of a high-precision volcanic eruption dating method is required. How-
ever, the reported 14C ages show large variations even for the same widespread tephra in many cases.
Towada Ofudo (T-Of) tephra was erupted by the Towada Volcano at the late phase of the marine iso-
tope stage 3 (MIS 3). The late phase of the MIS 3 is an important period in Japanese archaeology
because that is the oldest period for ancient Paleolithic ruins found in Japan. So this tephra is a very
important time marker in northeast Japan.
So far, many 14C ages have been reported for this tephra, but reported 14C ages show large varia-
tions. We corrected the wood from buried the T-Of tephra layer. The alpha-cellulose was extracted
from this wood sample. The influence of the contamination can be excluded by refining to the alpha
cellulose. Thus, we are able to determine the age of the T-Of tephra with high accuracy.

P-138

COMPARISON OF LAKE SEDIMENTS AGE-DEPTH MODEL BASED ON HIGH-RESOLUTION

14C DATINGS WITH VARVE CHRONOLOGY

Irka Hajdas1, Adam MichczyÒski2

1. ETH Zurich, Ion Beam Physics, Radiocarbon Laboratory, Schafmattstrasse 20, 8093 Zurich, Switzerland.
2. GADAM Centre of Excellence, Institute of Physics, Silesian University of Technology, Krzywoustego 2, Gliwice, Poland.

Two chronological Holocene records were obtained for laminated sediments of Lake Soppensee
(Switzerland)—high-resolution radiocarbon chronology and varve chronology. The new statistical
tools, which are available in OxCal 4.0 (Ramsey 2006), allowed creation of an improved age-depth
model for sediments recovered from Soppensee. The comparison of this model with the varve chro-
nology was done by calculation of agreement index between the modeled calibrated 14C age proba-
bility distribution and the varve age distribution for each pair of radiocarbon and varve dates. The
method of agreement index calculation was the same as the one used in OxCal to compare prior and
posterior distributions. Finally, the overall agreement index of analyzed sections of chronologies
was calculated. The results show very good agreement of the age-depth model with the varve chro-
nology, especially for the section for depths 420–500 cm (overall agreement index >70%). The sig-
nificant differences between the model and varve chronology occur only for the profile section
where lamination of the sediment was poor. This presents a discussion of the achieved results. The
comparison between the varve chronology and age-depth model illustrates the great potential of
high-resolution radiocarbon chronologies for precise dating of Holocene and Late Glacial records.

P-139

SOURCES OF PLANT RESIDUES IN A SEDIMENT CORE (PY608W-PC) FROM LAKE

PUMOYUM CO DURING THE LAST 19 KYR
Takahiro Watanabe1, Tetsuya Matsunaka2, Toshio Nakamura3, Mitsugu Nishimura2, Yasuhiro
Izutsu2, Motoyasu Minami4, Fumiko Watanabe Nara1, Takeshi Kakegawa1, Liping Zhou5
1. Graduate School of Science, Tohoku University, Japan.
 Abstracts 157

2. School of Marine Science and Technology, Tokai University, Japan.

3. Center for Chronological Research, Nagoya University, Japan.
4. Department of Environmental Biology, Chubu University, Japan.
5. Institute of Tibetan Plateau, Chinese Academy of Science, China.

The Tibetan plateau is highly sensitive to monsoon variability because it is near the limit of Asian
monsoonal influence. Therefore, paleoclimatic and environmental records from the Tibetan plateau
provide important clues for understanding the Asian climate system. Pumoyum Co is a freshwater
lake formed by fault action on the southeastern Tibetan plateau (28°34′N, 90°24′E, altitude: ~5030
m asl, lake surface area: 281 km2, maximum water depth: 65 m). The new continuous sediment core
(PY608W-PC; core length, 3.9 m) for reconstruction of climatic changes in the area was taken from
the eastern part of Pumoyum Co in August 2006. In this study, fine (>125 μm) plant remains from
the sediment layers of the core from Pumoyum Co were concentrated by wet-sieving and picked out
for dating in order to preclude an old carbon effect. The main objective of our study was to evaluate
the resulting suite of 42 new AMS radiocarbon ages to determine an accurate chronology for the
cores. Because stable carbon isotope ratios have proved to be useful for deducing the sources of
plant fragments in lake sediments, we carried out both high-time-resolution radiocarbon dating and
stable carbon isotope ratio measurements of plant residues from the PY608W-PC covering the
period from the last glacial to the Holocene. In this study, the aquatic plant residues disappeared
between 300 cm and 280 cm in core depth of PY608W-PC, and silt-silty clay (with a small amount
of terrestrial plant residues) took their place. The calibrated ages for base of the silt layer (~280 cm
in core depth) were 15.5–12.9 cal kyr BP, which correspond with the timing of Bølling-Allerød
warm phase.

P-140

14C
DATING OF HOLOCENE SOILS FROM AN ISLAND ON LAKE PUMOYUM CO (SOUTHEAST-
ERN TIBETAN PLATEAU)
Takahiro Watanabe1, Tetsuya Matsunaka2, Toshio Nakamura3, Mitsugu Nishimura2, Yasuhiro
Izutsu2, Motoyasu Minami4, Fumiko Watanabe Nara1, Takeshi Kakegawa1, Liping Zhou5
1 Graduate School of Science, Tohoku University, Japan.
2 School of Marine Science and Technology, Tokai University, Japan.
3 Center for Chronological Research, Nagoya University, Japan.
4 Department of Environmental Biology, Chubu University, Japan.
5 Institute of Tibetan Plateau, Chinese Academy of Science, China.

Previous studies using soils, loess deposits, and lake sediments have documented environmental
changes in the central, northern, and western Tibetan plateau. However, the detailed environmental
changes during the late Quaternary are still largely unknown in the southeastern Tibetan plateau
because of its very high altitude (~5000 m asl) and severe environmental conditions. Lake
Pumoyum Co is located on the southeastern Tibetan plateau (28°34′N, 90°24′E, altitude: ~5030 m
asl). Continuous soil samples (85-cm-long section) were taken from an island on Lake Pumoyum Co
in August 2006. In order to estimate past water circulation (monsoon activities) and lake-level
changes of Pumoyum Co during the Holocene period, we analyzed AMS 14C ages, stable carbon
isotope compositions, and total organic carbon/total nitrogen atomic ratios of the soil samples from
the island on the lake. The 14C measurements were performed by a Tandetron Accelerator Mass
Spectrometry system (Model-4130 AMS, HVEE), at the Center for Chronological Research,
158 Abstracts

Nagoya University. The 14C concentration of the surface layer (101 pMC, at 5 cm depth) was nearly
modern. Based on the present 14C chronology, continuous soil sedimentation began after ca. 5800
cal BP (at a depth of 63 cm, top of fluvial gravel layer) for the island on Pumoyum Co. These results
could be caused by a decrease in the lake level during the middle Holocene. The age of the obvious
lithologic boundary (ca. 5800 cal BP) was in agreement with the end of Holocene climate optimum.

P-141

HOLOCENIC PHYTOLITH AND ISOTOPIC RECORD FROM AN UMBRIC EPIPEDON OF AN

OXISOL (SALINAS, MINAS GERAIS STATE, SOUTHEASTERN BRAZIL)
Marcia Regina Calegari1, Pablo Vidal-Torrado2, Marco Madella3, Luiz Carlos R Pessenda4,
Flávio Adriano Marques5
1. Departamento de Geografia, Universidade Estadual do Oeste do Paraná – UNIOESTE, Mal. Cândido Rondon/PR, Brazil.
2. Departamento de Ciência do Solo, Campus “Luiz de Queiroz”/Universidade de São Paulo, ESALQ/USP, Piracicaba/SP,
Brazil.
3. ICREA, Dept. of Archaeology and Anthropology, Institució Milà i Fontanals, Spanish National Research Council (CSIC),
Barcelona, Spain.
4. Laboratório de 14C, Centro de Energia Nuclear na Agricultura (CENA), Piracicaba/SP, Brazil.
5. Doutorando em Solos e Nutrição de Plantas, Departamento de Ciência do Solo, ESALQ/USP, Piracicaba/SP, Brazil.

The reconstruction of paleoenvironmental conditions—which contributed to the pedogenetic pro-

cesses, organic matter input, and stabilization in a very thick (>1 m) umbric epipedon of a Oxisol
profile—was performed through comparisons between phytolith assemblages and carbon data (total
C, organic C, and δ13C) and the radiocarbon dating chronology. The profile is located in Salinas
(16°16′16.3″S; 42°58′50.5″W), Minas Gerais State, southeastern Brazil. The modern vegetation
cover is cerrado (savanna). The same studies considered this Oxisol subgroup as relict soil devel-
oped under different climatic conditions. The aim of this study was to detect eventual environmental
changes that occurred during the Holocene and that contributed to the soil development. The phy-
tolith assemblages and the isotopic organic matter values revealed that the humic horizon was devel-
oped under a mixture of (dominant) C3 and C4 plants. The most enriched δ13C values (–21.04‰)
recorded at the bottom of the umbric epipedon (Early-Middle Holocene), suggest the presence of
more open vegetation (cerrado), probably associated to a drier climate. From Late Holocene to the
present the δ13C became more depleted (–25.02‰), probably associated with the modern vegetation
cover (woody savanna) and more humid climate. The phytolith index used in the study showed vari-
ability, indicative of Holocene variations in the vegetation structure of the area.
Research funded by FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo).

P-142

VEGETATION AND CLIMATE RECONSTRUCTION IN NORTHEASTERN BRAZIL DURING THE

LAST 15,000 YEARS INFERRED FROM CARBON ISOTOPES AND CHARCOAL SOIL DISTRI-
BUTION
L C R Pessenda1, S E M Gouveia1, J A Bendassolli2, R Aravena3, S E M G Saia1
1. 14C Laboratory-Centre for Nuclear Energy in Agriculture, Piracicaba, S. Paulo, Brazil.
 Abstracts 159

2. Stable Isotopes Laboratory-Centre for Nuclear Energy in Agriculture-Piracicaba-Brazil.

3. Environmental Isotopes Laboratory, University of Waterloo, Canada.

This poster presents results with the objective to expand the vegetation and climate studies for north-
eastern Brazil in the last 15,000 years. Samplings have been made in three states of the region
(Ceará, Piauí, and Paraíba), covering a linear distance of ~900 km. Soil profiles were sampled for
δ13C analysis and buried charcoal fragments used for 14C dating. In the area of Flona (Ceará), the
isotope enrichment from –24‰ to –19‰ during the early-mid Holocene is an indication of an open-
ing of arboreal vegetation (woody savanna), associated with the expansion of C4 plants; a trend
toward more depleted values (~ –32‰) in the late Holocene indicates the dominance of C3 plants. A
similar trend is observed at Parna (Piauí) and Rebio (Paraíba) where values of ca. –24‰ indicate the
presence of more open vegetation, woody savanna, during the late Pleistocene. In the early-mid
Holocene, more enriched values (up to –18.0‰) suggest a more significant presence of C4 plants.
The presence of buried charcoal fragments in several soil profiles suggested the occurrence of paleo-
fires during the Holocene, reinforcing the dry early to mid-Holocene period. Based on the isotopic
data, it is possible to postulate that approximately since the late Pleistocene (~15,000 yr BP) to the
early Holocene (~10,000-9000 yr BP), arboreal vegetation was present in the study areas and related
to a more humid period. Since ~9000 yr BP until 4000–3000 yr BP, the savanna expanded, which is
associated with the presence of a drier climate. From ~4000–3000 yr BP to the present, a trend
toward more depleted values was interpreted as an arboreal vegetation expansion due to the return
of more humid phase. These results agree with studies developed in northeastern Brazil and also in
the south and central Amazon region, implying similar climatic conditions affecting these areas
since the late Pleistocene.

P-143

COMPARISON OF SOIL CARBON ACCUMULATION IN CONIFER AND BROAD-LEAF FOR-

ESTS USING RADIOCARBON MEASUREMENTS
Ayu Toyota1, Ichiro Tayasu2, Reiji Fujimaki3, Nobuhiro Kaneko3, Masao Uchida4,
Yasuyuki Shibata4, Tsutom Hiura5
1. Japan Wildlife Research Center, Japan.
2. Kyoto University, Japan.
3. Yokohama National University, Japan.
4. National Institute for Environmental Studies, Japan.
5. Hokkaido University, Japan.

Considerable interest is recently being paid how vegetation type and living organisms affect soil car-
bon accumulation. Soil macro-invertebrates often consume litter and soil; however, the effect of soil
animals on soil carbon accumulation has not been studied. To investigate the effects of vegetation
type and animals within the soil on soil carbon accumulation, we examined three types of cool-tem-
perate forest, where the vegetation switched from conifer to broad-leaf trees (CB), broad-leaf trees
to conifers (BC), and conifer trees (CC), respectively. To separate the vegetation effect using radio-
carbon measurements, the vegetation switch timing is 1954–1963, before the 1964 peak in bomb
radiocarbon in atmospheric CO2. Sampling soil layer is accumulated after an eruption of Mt. Taru-
mae in 1739.
160 Abstracts

We found that existing litter accumulation was significantly lower, and earthworm biomass was sig-
nificantly higher in CB than in CC and BC. Earthworms are dominant soil macro-invertebrates in all
sites. We hypothesized that these differences in litter accumulation patterns and earthworm biomass
caused changes in soil carbon accumulation processes have enhanced recent carbon accumulation in
soil at CB due to the existing high earthworm biomass.
Distribution of 14C with depth in CC and BC peaked in the litter and surface soil layer (in CB peaked
in the soil layer at 4–6 cm depth) supported our hypothesis. The 14C distribution in CC is similar to
the changes in atmospheric radiocarbon with depth. Compared to CC, 14C distribution in CB and BC
were constant throughout the depth. Low 14C values before the peak in bomb 14C were not detected
in BC even at the deepest soil layer. Soil-feeding earthworms, which dominated in BC, were moving
between the surface and deeper soil layer. These results suggest that earthworms probably import
new carbon into the deep soil horizons in BC and CB; thus, soil carbon accumulation depends on
past and current vegetation closely related with earthworm activity.

P-144

SEQUENTIALLY DENSITY FRACTIONATION AND RADIOCARBON-BASED RESIDENCE TIME

FOR VOLCANIC ASH SOIL IN A JAPANESE COOL-TEMPERATE FOREST
Miyuki Kondo, Masao Uchida, Yasuyuki Shibata
AMS Facility (NIES-TERRA), Environmental Chemistry Division, National Institute for Environmental Studies, 16-2
Onogawa, Tsukuba 305-8506, Japan.

Although Andisols cover only approximately 0.8% of the world’s land surface (ISSS Working
Group RB 1998), they represent a very important land resource in regions where active and recently
extinct volcanoes are located, such as in the Pacific Rim region. Andisols generally contain much
higher concentrations of SOC than other soils. However, its capacity and the instability of carbon are
not known well. We conducted sequential density separations and measured radiocarbon contents in
order to evaluate the residence time for volcanic ash soil in a Japanese cool-temperate forest.
We find that at the deep soil layer (40–45 cm depth), both heavy fraction (>2.1 g cm–3) and lower
density fractions (<2.1 g cm–3) have significantly longer residence time, ranging from 2040 to 3680
yrs. On the other hand, at the soil surface (5–10 cm depth), the residence times varied from 12 yr at
1.0–1.6 g cm–3 to 350 yr at 2.1–2.4 g cm–3. We are also conducting mineralogy analyses for each
fraction and will examine its relation to residence time. We will show the results at this meeting.

P-145

RADIOCARBON ANALYSIS OF DEEP PEAT CO2 USING A PASSIVE SAMPLING TECHNIQUE

M H Garnett1 and S M L Hardie2
1. NERC Radiocarbon Facility (Environment), Rankine Avenue, East Kilbride, G75 0QF, UK.
2. Chemistry Department, Joseph Black Building, University of Glasgow, Glasgow G12 8QQ, UK.

We developed and tested a method to collect CO2 from the surface to deep layers of a peatland for
radiocarbon analysis. The method comprises two components: i) a probe equipped with a hydropho-
 Abstracts 161

bic filter that allows entry of peat gases by diffusion, while simultaneously excluding water, and, ii)
a cartridge containing zeolite molecular sieve that traps CO2 passively. We field-tested the method
by sampling at depths of between 0.25 and 4 m at duplicate sites within a temperate raised peat bog
(Langlands Moss, near Glasgow, UK). CO2 was trapped at a depth-dependent rate of between ~0.2–
0.8 mL day–1, enabling sufficient CO2 for 14C analysis to be collected when left in place for several
weeks. The age of peatland CO2 increased with depth from modern to ~170 BP for samples col-
lected from 0.25 m, to ~ 4000 BP at 4 m, and was younger, but followed a similar trend, to the age
of bulk peat previously reported for the site. δ13C values of recovered CO2 increased with depth.
CO2 from the deepest samples was considerably 13C-enriched (up to ~ +9‰) and agreed well with
results reported for other peatlands, where this phenomenon has been attributed to fermentation pro-
cesses. CO2 collected from plant-free static chambers at the surface of the mire was slightly 14C-
enriched compared to the contemporary atmosphere, suggesting that the surface soil CO2 flux was
predominantly derived from carbon fixed during the post-bomb era, and therefore, that the aged
deep peat CO2 did not contribute greatly to the surface efflux. The passive sampling technique is
readily portable, easy to install and operate, causes minimal site disturbance, and can be reliably
used to collect peatland CO2 from a wide range of depths.

P-146

QUANTIFYING SITE-SPECIFIC HOLOCENE SOIL EROSION USING DEPTH-PROFILES OF

COSMOGENIC IN SITU 14C AND 10Be
R-H Fülöp1,2, P Naysmith1, G T Cook1, D Fabel2, P Bishop2, C Schnabel1, S Xu1
1. SUERC, Scottish Enterprise Technology Park, Rankine Avenue, East Kilbride G75 0QF, Scotland, UK.
2. Department of Geographical & Earth Sciences, University of Glasgow, Glasgow G12 8QQ, Scotland, UK.

The sustainable use of soils represents one of the key challenges that society faces. We are testing
whether cosmogenic in situ 14C and 10Be depth-profiles can be used to determine amounts and tim-
ing of erosion on a soil formed on a Younger Dryas moraine in Scotland. Eighteen 10Be determina-
tions were carried out on till material of varying size fractions collected from a 240-cm depth-profile
from the moraine crest. The results confirm that the expected cosmogenic nuclide depth-profile
from a sediment body of Holocene age can be reconstructed and, thus, the technique can be applied.
The 14C measurements are instrumental for the success of this project as 10Be (given its long half-
life relative to Holocene timescales) can only show the erosional loss, whereas 14C, with a substan-
tially shorter half-life, can also discriminate between the various timings of surface erosion. We used
the 14C extraction vacuum line of Naysmith et al. (2004) that was built following Lifton et al.
(2001). The purified quartz is heated to 1100 °C in a resistance furnace in the presence of lithium
metaborate and ultra-high purity oxygen. The released carbon is oxidized and the resulting CO2 is
cryogenically cleaned and graphitized for AMS measurement. The preliminary system blanks aver-
age 3.1 ± 0.2 × 105 14C atoms. This value is similar to those obtained by Lifton et al. (2001) and
Pigati et al. (2005). The blanks exhibited substantial fluctuations at the beginning of the study, sug-
gesting that the continuous running of the vacuum system was slowly cleaning contaminant carbon
from the line. These fluctuations in the blanks were also reduced with longer cleaning of the quartz
sleeves and close monitoring of the temperatures of the cryogenic traps used in the gas-cleaning
steps. In this paper, we present preliminary results of our blank and standard (PP4) measurements.
162 Abstracts

REFERENCES
Lifton et al. 2001. A new extraction technique and production rate estimate for in situ cosmogenic 14C in quartz. Geochimica
et Cosmochimica Acta 65:1953–69.
Naysmith et al. 2004. Preliminary results for the extraction and measurement of cosmogenic in situ 14C from quartz. Radio-
carbon 46:201–6.
Pigati et al. 2005. Extracting in situ cosmogenic 14C from olivine: significance for the CRONUS-Earth project. EOS, Trans-
actions of the American Geophysical Union 86(52), Fall Meet. Suppl., Abstract U23B-06.

P-147

RECONSTRUCTION OF SHELL-REWORKING PROCESS BY USING SEDIMENT-ACCUMULA-

TION CURVE: A CASE STUDY FROM THE LATEST PLEISTOCENE TO HOLOCENE INCISED-
VALLEY FILLS UNDER THE TOKYO LOWLAND, CENTRAL JAPAN
Susumu Tanabe1, Rei Nakashima1, Masao Uchida2, Yasuyuki Shibata2
1. Geological Survey of Japan, AIST, Central 7, Higashi 1-1-1, Tsukuba 305-8567, Japan.
2. National Institute for Environmental Studies, Onogawa 16-2, Tsukuba 305-8506, Japan.

When we reconstruct paleoenvironments of the geological record by using shell fossils, we identify
the paleoenvironments by analyzing the community composition of shell fossils. However, in the
geological records, allochthonous shells are regarded as reworked shells, and indigenous shells are
only used to reconstruct the paleoenvironments. But when we reconstruct the paleoenvironments
from the late Quaternary sediments, the shell fossils are radiocarbon dated, and the radiocarbon
dates are compared with the sediment-accumulation curve of the late Quaternary sediments. The
sediment-accumulation curve is connected with the youngest age-depth plots in certain horizons. By
comparing an age-depth plot of shell fossil with a sediment-accumulation curve, we can identify the
location and the age of the strata from which the shell fossil is derived.
Since 2002, Geological Survey of Japan investigated the stratigraphy and distribution of the latest
Pleistocene to Holocene incised-valley fills under the Tokyo Lowland to evaluate the ground ampli-
fication from earthquake. Unconsolidated shallow marine sediments distribute in the portion of the
Paleo-Okutokyo Bay, which existed during the middle Holocene sea-level highstand. By radiocar-
bon dating of shell fossils from the marine sediments in the sediment core, we identified the shell
fossils, which were reworked from tidal flat sediments (8000 cal BP) to bay sediments (4000 cal BP)
and from bay sediments (7000 cal BP) to bay sediments (6000 cal BP). These results indicate that
the wave-reworking processes and the bioclast recycle were dominant in the bay mouth portion of
the Paleo-Okutokyo Bay.

P-148

COMPARISON OF DEPTH PROFILES OF 129I AND 14C CONCENTRATION IN THE SURFACE

LAYER OF SOILS COLLECTED FROM NORTHEASTERN REGION OF JAPAN
Hiroyuki Matsuzaki1, Yoko Sunohara Tsuchiya1, Yasuyuki Muramatsu2, Yuji Maejima3,
Yosuke Miyairi1, Kazuhiro Kato4
1. Dept. of Nuclear Engineering and Management, School of Engineering, University of Tokyo, Hongo 7-3-1, Bunkyo-ku,
Tokyo 113-8656, Japan.
2. Dept of Chemistry, Gakushuin University, Mejiro 1-5-1, Toshima-ku, Tokyo 171-8588, Japan.
 Abstracts 163

3. National Institute for Agro-Environmental Sciences, 3-1-3 Kannondai, Tsukuba 305-8604, Japan.
4. National Institute for Environment Studies, 16-2 Onogawa, Tsukuba 305-8506, Japan.

By means of accelerator mass spectrometry, 129I/127I isotopes become an entirely new tool for envi-
ronmental analyses. The close association between iodine and organic carbon makes 129I/127I a
potential proxy of organic carbon dynamics. To examine this, depth profiles of 129I and 14C concen-
tration in the surface layer of several Japanese soils were measured and compared. 14C in soil is
mainly provided by the litter fall, of which the 14C/12C ratio is in equilibrium with the atmosphere.
Modern 129I is produced artificially and provided from the atmosphere; however, the detailed trans-
fer process is not clear. The nuclides could be concentrated near the surface and decrease exponen-
tially with depth, suggesting a simple downward moving process. In deeper layers, the profiles show
different trends with rather moderate decreasing. We can extract several important parameters rep-
resenting the character of each depth profile, i.e. the concentration at the surface, the steepness of the
profile in the surface layer, and the crossover point between two different profile trends.
Upon closer examination of these parameters as well as the carbon and iodine content, we will dis-
cuss the similarity and difference between 129I and 14C profiles and examine the dynamics of ele-
mental carbon and iodine in the soil environment.

P-149

14C DISTRIBUTION AND SOIL DEVELOPMENT IN A TYPICAL HILL SLOPE

Kexin Liu1, Chengde Shen2, Xingfang Ding1, Ping Ding2, Dongpo Fu1, Naveed Ali1
1. State Key laboratory of Nuclear Physics and Technology & Institute of Heavy Ion Physics, Peking University, Beijing
100871, China.
2. Guangzhou Institute of Geochemistry Chinese Academy of Science, Chinese Academy of Sciences, Guangzhou 510640,
China.

Soil development and erosion are very important for the study of globe change because soil is an
important source and sink of many greenhouse gases such as CO2, methane, and nitrous oxide. We
have investigated the 14C concentration, δ13C, soil organic matter (SOM), and magnetic susceptibil-
ity (MS) of three profiles on a soil slope in the Crane Mountain area, Guangdong Province, China.
The 14C concentration is 101–111 pMC in the upper 20 cm of the profiles, corresponding to the
bomb-produced radiocarbon. The 14C dates of the samples are from zero to 500 years for the 20–40
cm layer and 500–1000 years for the 40–60 cm layer in the profiles. From 60–150 cm in the profiles,
it is clay-rich soil with many pieces of bedrock. The 14C dates of the samples from this layer are
1000–8000 years. Therefore, the creation process of soil by weathering of the underlying bedrock
lasts at least thousands years in this area.
164 Abstracts

P-150

STUDY OF CHANGES OF MORAVA RIVER FLOOD SEDIMENTS IN STRÁéNICKÉ POMORAVÍ

AREA (CZECH REPUBLIC) DURING THE LAST MILLENNIUM
J Kadlec1, I SvÏtlík2, T Grygar3, M MihaljeviË4, Helena Svitavská-Svobodová5
1. Institute of Geology ASCR, Paleomagnetic Laboratory, Rozvojová 269, 165 00 Prague, Czech Republic.
2. Nuclear Physics Institute ASCR, Radiocarbon Laboratory, Na Truhlá°rce 39/64, 180 86 Prague, Czech Republic.
3. Institute of Inorganic Chemistry ASCR, Analytical Laboratory, 250 68 ±Reû, Czech Republic.
4. Institute of Geochemistry, Mineralogy and Mineral Resources, Charles University, Albertov 6, 128 43 Prague, Czech
Republic.
5. Institute of Botany ASCR, Department of Geobotany, Zámek 1, 252 43 Pr˘honice, Czech Republic.

The Morava River, meandering for about 12 km through the Stráûnické Pomoraví area, Czech
Republic, is one of the last river segments without bank regulations. The purpose of our study is to
describe the Morava River floodplain deposits, mainly from the last millennium, using instrumental
analytical and dating methods. Stratigraphic correlations were performed using lithological descrip-
tion in the field followed by chemical and physical analyses. Also, geophysical, palynological, and
geomorphological studies were performed in selected parts of the Stráûnické Pomoraví. To deter-
mine the sedimentary record, we acquired mineral magnetic characteristics and evaluated mineral-
ogical composition of the sediments. The high-resolution logs of the mineral magnetic variations
were obtained by mineral magnetic methods (MS, NRM, ARM, SIRM, S-ratio). The magnetic
record was complemented by grain-size analyses, X-ray diffraction, quantification of expandable
clay minerals (CEC), and X-ray fluorescence spectroscopy. 14C dating was performed using organic
material, such as tree trunk fragments and charcoal. Selected wood samples were processed in den-
drochronological laboratory and the resulting age was compared with 14C dating results. The deter-
minations of DDT, PCB, 137Cs, heavy metals (Pb and Zn) concentration, and variations of stable Pb
isotope ratios were performed in the sediment samples to identify industrial pollution.
Some results connected with changes during the last millennium will be discussed as well as the
benefit and possibilities of 14C dating from the point of view of multidisciplinary studies performed
in this project.

P-151

ISOLATION OF DIFFERENT SOIL COMPONENTS FOR RADIOCARBON DATING OF AN ALLU-

VIAL FAN
S Szidat1,2, S Schmoker1, B Gasser1, H W Gäggeler1,2,3, I Hajdas4, L Wacker4, H Veit5
1. Department of Chemistry and Biochemistry, University of Bern, Switzerland.
2. Oeschger Centre for Climate Change Research, University of Bern, Switzerland.
3. Paul Scherrer Institut, Villigen-PSI, Switzerland.
4. Laboratory of Ion Beam Physics, ETH Hönggerberg, Zurich, Switzerland.
5. Institute of Geological Sciences, University of Bern, Switzerland.

For investigations of the dynamics of alpine rivers, soil samples were taken from the alluvial fan of
the Lütschine River in the region of Interlaken, Switzerland [1]. For stratigraphical purposes, the
ages of these soil samples were determined by radiocarbon dating methods. For some of the sam-
ples, the obtained ages were much older than the samples from proximate layers [1]. These compli-
 Abstracts 165

cations were the motivation to investigate the radiocarbon age of different soil components. The
goal was to use different pretreatment methods for isolation of soil components that were chemically
stable and have not been transported vertically since the time of their deposition in the soil.
The isolation of the humic acids was performed by dissolution of the carbonates using acid-alkaline-
acid extraction. Cellulose was isolated from wood pieces in a leaching process with sodium chlorite
and hydrochloric acid with precedent Soxhlet extraction [2]. In order to isolate glucose, wood parti-
cles and soil materials were hydrolyzed in concentrated hydrochloric acid solution. Oligosaccha-
rides could be isolated from soil samples with this technique, even if wood pieces are not macro-
scopically identifiable. The purification of the glucose from the hydrolyzate was performed by
preparative chromatography. In the isolated material 14C was measured at the small AMS facility
MICADAS at ETH Zurich. With these measurements, the previous dating ambiguities of the allu-
vial fan were clarified.

REFERENCES
[1] L. Schulte et al. Geomorphology, in press (2009).
[2] S. W. Leavitt and S. R. Dancer. Anal. Chem. 65:87–89 (1993).

P-152

AMS RADIOCARBON DATING OF THE U-4, U-3, U-2 TEPHRAS IN ULLENGDO VOLCANO,
SOUTH KOREA
M Okuno1, M Shiihara2, M Torii3, T Nakamura4, K H Kim5, H Domitsu6, H Moriwaki7, M Oda8
1. Dept. of Earth System Science, Faculty of Science, Fukuoka University, Fukuoka 814-0180, Japan.
2. West Japan Engineering Consultants, Inc., Fukuoka 810-0004, Japan.
3. Kumamoto Gakuen University, Kumamoto 862-8680, Japan.
4. Center for Chronological Research, Nagoya University, Nagoya 464-8602, Japan.
5. Dept. of Earth Science Education, Ewha Womans University, Seoul 120-750, South Korea.
6. Dept. of Ecosystem Studies, School of Environmental Science, University of Shiga Prefecture, Hikone 522-8533, Japan.
7. Section of Physical Geography, Faculty of Law, Economics and Humanities, Kagoshima University, Kagoshima 890-
0065, Japan.
8. Graduate School of Science, Tohoku University, Sendai 980-8578, Japan.

The Ullengdo volcanic island, a large stratovolcano, is located in the western part of the East Sea
(Japan Sea), 130 km off the eastern coast of the Korean peninsula. Of the 7 tephra deposits (U-7 to
U-1) identified by Machida et al. (1984), the pumieous U-4, U-3, and U-2 tephra deposits are petro-
chemically and petrographically similar to the Ullengdo-Oki (U-Oki) tephra. The U-Oki is a widely
distributed tephra in and around the Japan Sea (Machida et al. 1981), and is possibly correlated to
any of the three tephra deposits. However, the exact deposit correlated to the U-Oki tephra is not yet
identified.
To determine the eruption age of the U-4, U-3, and U-2 tephras in the Ullengdo Volcano, we con-
ducted 14C dating for 5 soil and 2 charcoal samples with a HVEE Tandetron AMS system at Nagoya
University (Nakamura et al. 2000). The organic carbon and nitrogen contents were also measured
for a humin fraction of soil sample, using a CN analyzer (MT-700, Yanaco). Although the soil sam-
ples have C/N ratios from 5 to 10, the obtained 14C dates are consistent with stratigraphy in the
Ullengdo. The calibrated 14C dates for the U-4, U-3, and U-2 tephras are 11 cal kyr BP, 8.3–9 cal kyr
166 Abstracts

BP, 5.6 cal kyr BP, respectively, indicating that the explosive eruptions occurred in the Ullengdo
with a time interval of 2000 to 3000 years during the early to middle Holocene. The age of U-Oki,
10.7 cal kyr BP, obtained from Lake Suigetsu data set in central Japan (Kitagawa et al. 1995), sug-
gests that the U-4 tephra is most likely correlated with the U-Oki tephra.

P-153

CHANGE IN ORGANIC MATTER SOURCES OF LAKE BAIKAL SEDIMENT CORE DURING THE
PAST 23,000 YEARS REVEALED BY STABLE CARBON ISOTOPE RATIOS WITH 14C DATING
Fumiko Watanabe Nara1, Takahiro Watanabe1, Toshio Nakamura2, Takeshi Kakegawa1,
Akio Imai3, Takayoshi Kawai4
1. Graduate School of Science, Tohoku University, Japan.
2. Center for Chronological Research, Nagoya University, Japan.
3. National Institute for Environmental Studies, Japan.
4. Graduate School of Environmental Studies, Nagoya University, Japan.

A number of studies of stable carbon isotope ratios of total organic carbon (δ13CTOC) profiles with
14C dating using Lake Baikal sediments have been reported in the last decade. However, the high-

time resolution profiles (less than 200 years) of δ13CTOC in Lake Baikal sediment are still limited. In
this study, to evaluate the detailed changes in organic matter sources in Lake Baikal, the δ13CTOC
measurements and 14C dating of TOC and pollen fraction samples of Lake Baikal sediment
(VER99G12) were carried out. The sediment core (VER99G12, core length 4 m) was taken from the
Buguldeika saddle of Lake Baikal in 1999. The 14C measurements for TOC and pollen fraction sam-
ples were performed by a Tandetron accelerator mass spectrometer (Model-4130 AMS, HVEE).
The AMS 14C age showed that core VER99G12 spans the past ~23 cal kyr BP, and the sedimentation
rate of this core is calculated as 18 cm/kyr. This means that the time resolution in VER99G12 sedi-
ment core is less than 60 yr BP.
Additionally, AMS 14C ages of pollen samples were parallel to those of TOC in sediment samples.
These results show that the reservoir effect for 14C ages of TOC on this sediment core is negligible.
The δ13CTOC values in core VER99G12 during the last glacial/post-glacial transition period (from 18
to 10 cal kyr BP) showed light carbon isotopic values up to –31.8‰. On the other hand, the δ13CTOC
values in VER99G12 showed a gradual decrease from –29.0 to –26.6‰ after the last glacial period,
that is, after 10 kyr BP. It is supposed that the rapid change in source of organic matter in Lake
Baikal could have occurred in the last glacial/post-glacial transition period.

P-154

14C DATING OF PHYTOLITH-RICH LAYERS: ATTEMPTS, SUCCESS, AND DIFFICULTIES

E Mintz1, E Boaretto1,2, V Chu1, S Weiner3
1. Radiocarbon and Cosmogenic Isotopes Laboratory, Weizmann Institute of Science, 76100 Rehovot, Israel.
2. Institute of Archaeology, Bar-Ilan University, Ramat-Gan, 52900, Israel.
3. Kimmel Center for Archaeological Science, Weizmann Institute of Science, 76100 Rehovot, Israel.
 Abstracts 167

Phytolith-rich layers visible to the naked eye were found in many different archaeological sites in
Israel over the last few years. They may well be widespread in the Levant and elsewhere. Phytolith-
rich layers are important archaeological contexts for understanding past activities (cereal storage,
animal enclosures), use of space, and the botanical environment. They are formed directly or indi-
rectly by anthropogenic activities and represent an in-situ deposition in the archaeological record.
As the archaeological context of these layers is secure, radiocarbon dating of these layers could rep-
resent an important step for building of chronologies.
To date, it has not been possible to obtain reliable radiocarbon dates from the small amounts of
organic matter occluded inside the phytoliths. We have used several procedures in order to extract
carbon dioxide from pure phytolith samples for radiocarbon dating. So far, the results indicate a very
complicated interaction between these amorphous silicates and the environment during their prepa-
ration. This results in contamination by modern carbon dioxide even after treatment with acidic
solutions and exposure to a vacuum. Such an interaction is not yet understood, and, therefore, radio-
carbon dating of phytolith sample is still not routinely possible.

P-155

THE NATURE AND AGES OF HUMIC SUBSTANCES IN JAPANESE VOLCANIC ASH SOILS AT
DIFFERENT STAGES OF VEGETATION SUCCESSION
Kazumi Katsuno1, Yosuke Miyairi2, Kenji Tamura3, Hiroyuki Matsuzaki2, Kenji Fukuda1
1. The University of Tokyo, Department of Natural Environmental Studies, Graduate School of Frontier Sciences, Japan.
2. The University of Tokyo, Department of Nuclear Engineering and Management, School of Engineering, Japan.
3. University of Tsukuba, Graduate School of Life and Environmental Sciences, Japan.

Soil is an important reservoir of carbon in terrestrial ecosystems. In addition, it is an important part

of the biogeochemical carbon cycle. Total global emission of CO2 from soils is recognized as one of
the largest fluxes in the global carbon cycle, and small changes in the magnitude of soil CO2 flux can
have large effects on the concentration of CO2 in the atmosphere (Schlesinger and Andrews 2000).
There is broad agreement on the major pathways of the soil carbon cycle (Trumbore 2008). Plant lit-
ter and root exudates input organic carbon to soil. It is decomposed by microorganisms and becomes
CO2 to return to atmosphere or dissolved organic matter to be washed away from soil. In that way,
stable organic matter stays in soil.
In Japan, volcanic ash soils exist in wide area, and many of them are Andisols. A great amount of
humic substances is contained in Andisol, which is related to grassland vegetation and maintained
by human disturbance. These humic substances are supposed to decrease when vegetation succes-
sion to be a forest takes place. However, there have been few data demonstrating this hypothesis. In
this study, we collected Andisol samples in sites where a maintained grassland and a pine forest,
which was an abandoned grassland, are next to each other. The samples were separated by acid and
alkali, and their 14C content was measured. From the result, fulvic acid was found to be younger than
other fractions. Moreover, the ages of fulvic acid were older in the forest than in the grassland. This
result suggested that humic substances changed to unstable fulvic acid or decomposed when a grass-
land became a forest. This may be the cause of the decrease in soil carbon during vegetation succes-
sion in Andisols.
168 Abstracts

P-156

THE IMPROVED APPLICATION OF CARBON ISOTOPES IN FORENSIC INVESTIGATION

N Wang1,2, C D Shen1,3, P Ding1,2, W X Yi1, W D Sun1, K X Liu3, X F Ding3, D P Fu3
1. Key Laboratory of Isotope Geochronology and Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of
Sciences, Guangzhou 510640, China.
2. Graduate University of Chinese Academy of Sciences, Beijing, 100049, China.
3. State Key Laboratory of Nuclear Physics and Technology, Peking University, Beijing, 100871, China.

Recent studies show that carbon isotopes can play a significant part in forensic investigations. The
enamel 14C contents of 13 teeth from two areas in China, Guangzhou and Bangbu, are examined in
our study. Most sample formation years obtained by comparing the 14C contents to the 14C bomb
pulse curve show a good agreement with the enamel formation time of the known-age teeth. How-
ever, the samples formed before 1960 yield greater errors, which means the old samples (formed
before 1960) may not be dated by this means. In the intent to date the exact age of the victim, it is
important that two or more teeth collected from one individual formed in different years should be
analyzed. In addition, it is necessary to choose the proper atmospheric radiocarbon bomb pulse
curve according to the area where the samples come from, since the radiocarbon content variation is
quite different in different zones all over the world, especially before 1975.
The enamel stable carbon isotope is correspondingly surveyed. As the δ13C is basically determined
by diet, the enamel d13C value varies in samples from different countries, even at different latitudes
of the same country. The samples from Guangzhou, at a lower latitude, give a lower mean δ13C
value of –14.50‰, while the ones from Bangbu yield a mean value of –13.27‰. With more and
more relevant research, we are going to present a better picture of the stable carbon isotope distribu-
tion in China, then the stable carbon isotope content in the enamel can reveal the information of the
area where the victim grew up, and narrow the search range when investigating the identity.

P-157

ENRICHMENT OF BOMB 14C IN SOIL CO2 FROM A SUBTROPICAL FOREST, SOUTH CHINA
P Ding1,2, C D Shen1,3, N Wang1,2, W X Yi1, X F Ding3, D P Fu3, K X Liu3
1. Key Laboratory of Isotope Geochronology and Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of
Sciences, Guangzhou 510640, China.
2. Graduate University of Chinese Academy of Sciences, Beijing 100049, China.
3. State Key Laboratory of Nuclear physics and Technology, Peking University, Beijing 100871, China.

Bomb 14C in the atmosphere decreased from about 900‰ in 1965 to 60~70‰ at present in terms of
Δ14C. The soil CO2 is not expected to be enriched in bomb 14CO2 as a result of the high soil gas flux
and the carbon sources, which include the present root respiration and the old-age soil organic car-
bon (SOC) decomposition. Contrary results were observed from two soil profiles (DHLS & DHS)
in Dinghu Mountain, south China. The Δ14C values of soil CO2 remain fairly stable along the depth,
ranging from 100.0‰ to 107.2‰ in DHLS (sample depth 90 cm) and 102.5‰ to 112.1‰ in DHS
(sample depth 105 cm). The values are obviously higher than the ones of atmospheric CO2 and SOC
at the same layer except for the top 20 cm of the profiles. Significant bomb 14CO2 enrichment in soil
CO2 has occurred in the deep soil layers (≥30 cm), where the Δ14C values of SOC become negative.
 Abstracts 169

Similar to 14CO2, 13CO2 also accumulates in soil CO2 along the depth. The CO2 δ13C varies from –
24.71‰ to –24.03‰ in the DHLS profile and from –25.19‰ to –22.82‰ in the DHS profile, clearly
more positive compared to the ones of SOC and roots in each profile.
One reason of the enrichment is likely concerned with the difference of diffusion rates between soil
14CO 13 12
2, CO2, and CO2 during soil gas fluxing into the atmosphere. Isotope enrichment during the
SOC decomposition is another possible reason. The enrichment suggests that the soil CO2 is a pos-
sible reservoir for atmospheric bomb 14CO2.

P-158

ORIGIN OF THE FALLING DUST IN BEIJING, APRIL 2006: ANOTHER EVIDENCE FOR AEO-
LIAN ORIGIN OF LOESS
W X Yi1, C D Shen1,3, P Ding1,2, N Wang1,2, X F Ding3, D P Fu3, K X Liu3, X M Xu4
1. Key Laboratory of Isotope Geochronology and Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of
Sciences, Guangzhou 510640, China.
2. Graduate University of Chinese Academy of Sciences, Beijing 100049, China.
3. State Key Laboratory of Nuclear physics and Technology, Peking University, Beijing 100871, China.
4. Department of Earth System Science, University of California, Irvine, CA 92697-3100, USA.

The aeolian origin of loess presumes that falling dust likely has a similar composition to that of
loess. A comparison of isotopes in falling dust and different areas of loess would indicate the falling
dust sources and its accumulation, which finally helps us to reconstruct the formation of the loess.
We collected falling dust samples in April 2006 in Beijing. 10Be, 14C, and δ13C of the samples were
analyzed. The content of 10Be in the falling dust ranges from 1.69 × 108 atoms/g to 2.07 × 108 atoms/
g with a mean value of 1.82 × 108 atoms/g, clearly in the interval of 10Be content of loess ranging
from 1.4 × 108 atoms/g to 2.8 × 108 atoms/g. The δ13C values of black carbon in falling dust, varying
between –23.15‰ and –23.80‰, are similar to the δ13C values of organic carbon in loess, ranging
from –21‰ to –24‰. Both 10Be content and δ13C values in falling dust exhibit obvious difference
from the ones in paleosols. Thus, the falling dust in Beijing likely originated from the loess in north-
west China. According to the 14C ages of the black carbon in falling dust, at least two areas of origin
for the dust can be revealed. One area is thought to be the excessive eroded area with an older 14C
age interval ranging between 4200 and 5200 BP, and the other is probably the less eroded area with
14C ages varying from 3000 to 4000 BP. Finally, the origin of the falling dust from loess would be

another evidence for the aeolian origin of loess.

P-159

14CCHRONOSTRATIGRAPHY AND PALEOCLIMATE INDICATIONS FROM BURIED ANCIENT

FOREST IN GUANGDONG, SOUTH CHINA
C D Shen1,3, W X Yi1, P Ding1,2, N Wang1,2, X F Ding3, D P Fu3, K X Liu3
1. Key Laboratory of Isotope Geochronology and Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of
Sciences, Guangzhou 510640, China.
170 Abstracts

2. Graduate University of Chinese Academy of Sciences, Beijing 100049, China.

3. State Key Laboratory of Nuclear physics and Technology, Peking University, Beijing 100871, China.

A great snow storm happened in the early spring of 2008 and brought great damage to the forest eco-
system in south China. This kind of event also likely occurred in ancient times, which finally led to
the disappearance of the ancient forest in Guangdong, south China. To study the mechanism of the
climate change and to better predict the climate variation in the future, we have carefully investi-
gated the 14C chronostratigraphy of the buried ancient forest in Sihui and Gaoyao, Guangdong Prov-
ince.
The ancient forest was dominated by Glyptostrobus pensilis formed in swamps at 4210 ± 45 cal yr
BP and disappeared with the wetland at about 3300 cal yr BP, lasting for nearly 1000 years. During
that period, the climate was warm, humid, and stable. However, it changed to cold and dry at around
3300 cal yr BP, corresponding to the start of the Hemer little ice age. A continuous freeze was
assumed to be main reason for the death of Glyptostrobus pensilis in large amounts, and a subse-
quent drought finally terminated the development of the forest. Almost at the same time, the global
climate also turned to cold and dry, revealed by paleoclimate records all over the world.
In recent years, the global climate trends have been unstable, and extreme weather appears fre-
quently. Would the great forest damage and the snow disaster of the low-mid latitudes of south
China forewarn us of a new little ice age coming?