(a) Institute for Biodiversity and Ecosystem
Dynamics, University of Amsterdam, the Netherlands, mauquoy@science.uva.nl
and vangeel@science.uva.nl
(b) Centre for Isotope Research, Groningen University, Nijenborgh
4, 9747 AG Groningen, the Netherlands, plicht@phys.rug.nl
Introduction
Radiocarbon wiggle-match dating of organic
deposits (WMD) is based on the phenomenon that plants (e.g., trees
and also plants living on raised bogs), while assimilating CO2,
all record the same fluctuations in atmospheric 14CO2
concentration. Past changes in atmospheric 14CO2
content during the Holocene have been recorded as fluctuations
of 14C ages of tree-ring sequences of exactly known
age (14C calibration curve INTCAL98; Stuiver et al.,
1998).
A high-resolution sequence of 14C dates from plant
remains extracted from a peat core should show the same wiggles
as the 14C calibration curve. When the wiggles of both
sequences are matched to each other, a high-precision calendar
age chronology can be obtained for the peat core. Precision obtainable
with WMD depends on, among others, the shape of the relevant part
of the calibration curve (Blaauw et al., 2003, 2004; Mauquoy et
al., 2004; van der Plicht et al., 2004)
General guidelines
-A peat core should usually contain enough material for reliable 14C dating. Although raised bog deposits consist nearly entirely of organic material, one should use above-ground and identifiable material only (no bulk dates). In peat with a high degree of decomposition, there is only little of such above-ground material left. As sample sizes are too small for conventional dating, AMS dating has to be used in all cases.
-If we prepare 14C samples ourselves before submitting, the amount of analytical work needed at the AMS laboratory will be reduced considerably. As analytical work often is the limiting factor in AMS laboratories, preparing the material ourselves will significantly speed up waiting times and will also reduce prices for 14C dating by 25%.
-While slicing up a core, collect 10 samples for 'range-finder' 14C dates throughout the core at regular intervals and/or relevant levels, and submit these samples (via Amsterdam; see below) to the AMS laboratory in an early stage of the project. These 10 samples for 14C dating will also have to be used for macrofossil analysis (analyse the macrofossils before picking the material to dated).
-When the range finder 14C AMS
samples have been submitted, perform macrofossil analysis for
the rest of the core. During the analysis, note for every level
of the core whether it contains sufficient aboveground remains
for AMS 14C dating, and if so, which
remains would be best to use.
-Upon receiving the range finder AMS 14C
dates, a preliminary chronology can be made. With only 10 dates,
no real wiggle-match can yet be performed. Only high-resolution
14C sequences can be securely matched to the calibration
curve. Levels where more dates are needed (and where sufficient,
suitable above-ground remains are present) can be identified.
Here one should aim that the sequence of 14C dates
follows the wiggles in the
calibration curve (if any), as well as aim to date relevant depths
(e.g., depths where important changes in lithology (humification,
charcoal levels, macrofossil composition, pollen concentration)
occur. Some 'educated guessing' is required. Often there is a
compromise between precision of chronology desired, and time and
budget available. During the Helsinki meeting in April 2004 it
was decided to focus with detailed, high resolution work on the
following periods:
2550
to 2050 cal. BC
Figures
1 and 2 show the 14C calibration curve and corresponding
fluctuations of delta 14C for this interval.
1000 to 500 cal. BC
Figures
3 and 4 show the 14C calibration curve and corresponding
fluctuations of delta 14C for this interval.
800 AD to 1800 cal. AD
Figures
5 - 8 show the 14C calibration curve and corresponding
fluctuations of delta 14C for this interval (please
note 100 years overlap of horizontal axis).-Nearly always a third set of 14C dates
will be necessary for 'zooming in' into the most satisfactory
14C wiggle-match chronology. It nearly always will
take some months after submission of samples before you will obtain
the 14C dates.
-For WMD, information about the accumulation history (possible accumulation rate changes or hiatuses) of a peat core is needed. In order to reconstruct the accumulation history of a peat core, information about, e.g., pollen concentration, bulk density, local vegetation composition (macrofossils), or occurrence of charcoal layers is important.
-The accumulation rate of a peat core should have been fast enough (and degree of compaction small enough) for a 14C sequence to follow wiggles in the 14C calibration curve. Along the same line of reasoning, the density of 14C dates should be high enough to be able to reconstruct the wiggles.
-For WMD, 14C dates are most valuable when they fit in periods of pronounced wiggles in the calibration curve. It does not make sense to have a large number of 14C dates in a plateau. Also, periods with less pronounced fluctuations in the 14C calibration curve result in less precise WMD chronologies (Blaauw et al., 2003). In such cases, possibly a limited number of calibrated 14C dates are sufficient, and the high-resolution WMD strategy would probably only be waste of time and finances.
-Great care should be taken to clean the samples as thoroughly as possible. Several studies report problematic 14C ages, many of which can be attributed to contamination (Wohlfarth et al., 1998; Speranza et al., 2000; Kilian et al., 2001; Nilsson et al., 2001). There are several stages in the preparation of 14C samples where contamination can enter, and even small amounts of contamination could result in "erroneous" 14C ages (e.g., Kilian et al., 1995, 2000). If examined carefully, macrofossil remains can show unexpectedly high degrees of (fossil) fungal contamination. Studies at our laboratory have shown that when samples consisted of above-ground plant remains only and were carefully cleaned of contamination, WMD placed the 14C ages of almost all samples investigated within the 1 s error envelope of the 14C calibration curve (Blaauw et al., 2003). Moreover, duplicate 14C samples of different species of macrofossils from the same levels showed no systematic, species-specific 14C age differences. This means that no indications were found for systematic, species-specific errors in 14C age of carefully cleaned aboveground parts of peat-forming plants. Vegetative remains of Cyperaceae should be avoided because they may show 14C ages that are too young (Kilian et al., 2000).
-Sometimes a sequence of 14C dates appears to have a constant 14C error, e.g., caused by a constant contamination of bulk dates (Kilian et al., 1995, 2000), or by a hard-water effect in lakes (Kilian et al., 2001). Calibration of individual 14C dates does not make sense in this case. With WMD however, such an effect can be recognized and corrected for, at least during periods with major wiggles and plateaus in the 14C calibration curve.
-Seeds (e.g., of Rhynchospora alba,
Scheuchzeria palustris, or Andromeda polifolia)
are very useful for 14C dating. They are easy to collect,
can readily be checked for purity and cleaned if necessary, and
already a few seeds provide a dry weight sufficient for AMS
dating (see below). Seeds can be severely contaminated however,
e.g., the interior of R. alba seeds regularly contains
large amounts of black matter resembling fungal remains.
-Sphagnum stems, branches and capitula/capsules are easy to collect although they also can be contaminated considerably. Sphagnum stems and leaves are often penetrated by ericaceous rootlets (Kilian et al., 1995, 2000). Stems of Sphagnum may appear clean on the outside, but when opened by forceps they can reveal large amounts of fungal hyphae. Close examination of Sphagnum leaves often reveals considerable visible contamination with fungal hyphae. Contamination would potentially have a smaller effect on 14C ages in Sphagnum stems than in leaves, because Sphagnum stems are much denser than Sphagnum leaves. If there is sufficient Sphagnum, we advise the selection of stems (with the leaves removed) and the capitula/capsules.
-Flowers of Ericaceae are often abundant in peat cores, but they easily collect large amounts of ericaceous rootlets and fungal hyphae. Removal of such contamination takes much time and effort, and often large parts of the flowers have to be discarded due to too much contamination. Stems and leaves of Ericaceae generally are easier to collect, although also here contamination can be considerable and difficult to remove. On the other hand, a sample of contaminated Calluna vulgaris leaved stems gave a 14C age similar to that of a clean sample of C. vulgaris from the same level (Blaauw et al. 2003).
-Leaves and bud-scales of Betula spec. (Tomlinson, 1985) are very easy to collect, although contamination can be considerable and difficult to remove.
-Hydathodes of Scheuchzeria palustris are useful for 14C dating. They can at times be abundant and are easy to collect and to clean (fungal contamination can be considerable).
Practical guidelines
-Several cases are known where strange 14C age results could be traced back to the fact that in the laboratory where the samples had been prepared, studies had been performed using 14C tracers! Even very small amounts of 14C tracers could mess up 14C age determinations. Check for 14C contamination if 14C tracers have ever been used in or near your laboratory. Avoid "shaking hands" with scientists who work with 14C tracers.
-Cut a core into contiguous slices of 1 cm, taking care to follow the orientation of any visible layering. Take samples of 2.5 cm diameter from the slices (they are meant for macrofossil analysis as well as for 14C AMS dating, see macrofossil protocol by Mauquoy), and boil these in 5% KOH (samples should be kept at boiling point for ca. 10 minutes, while stirring every few minutes so that the slices fall apart into individual plant remains). Filter the macroremains over a 100 µm stainless steel sieve and wash with demineralised water until the effluent is no longer visibly stained. Split the samples in manageable sub-samples and analyse these for macrofossils in demineralised water in a petri-dish. Store the samples at 4°C in demineralised water in marked sealed plastic bags, with some droplets of 5% HCl added.
-Samples for 14C dating should
be collected and pre-treated as follows: from a petri dish containing
demineralised water and a manageable sub-sample of macrofossils
(which had already been boiled in 5% KOH for macrofossil analysis),
collect suitable macroremains by forceps. Put the selected macroremains
in separate clean containers (e.g., glass-ware or china cups)
and clean them thoroughly of any visible contamination such as
fungal hyphae, rootlets, or material introduced during laboratory
work. Use a binocular microscope with 6-50x magnification, forceps
and purified (milli-Q) water. If possible, macrofossils are to
be split open to check for internal fungal contamination. In case
visible contamination of a macrofossil cannot not be removed entirely,
do not add the
macrofossil to the 14C sample ('dirty' material can
be taken only if there is very little clean material to be dated,
A note should then be made about this). After a first phase of
selection and cleaning, samples should be further cleaned, and
this procedure is to be repeated until the samples show no visible
contamination. Be very sure that the samples are as clean as possible.
Put the samples in cleaned glass vials with milli-Q water and
add some droplets of 4% HCl (1 part 37% HCl diluted in 9 parts
milli-Q water). Water could easily remove marker ink, so it would
be better to mark the vials with, e.g., paper labels, and then
cover the label with plastic tape. Store at 4ºC.
The following procedure is called "AAA"
by the AMS laboratory people. It is a washing procedure with alkalis
and acids to remove lots of unwanted material, including for example
recent CO2 dissolved in water. The procedure takes
about 4 days:
Day 1: put samples in HCl.
Day 2: Neutralize the samples, clean them and put them in the
oven.
Day 4: Weigh the samples, and send them to the AMS laboratory).
For convenient working, prepare ca. ten samples at a go (up to
20 samples are possible, but this will result in some very long
and busy days).
Take great care to avoid contamination at any time. Work with clean water (milli-Q), clean glassware (cleaned with plenty of clean water) and clean forceps. Avoid hand contact with your material. Don't use paper tissues for, e.g., cleaning or drying.
Put the samples in 100 ml glass beakers
containing milli-Q water with 4% HCl (1 part 37% HCL diluted in
9 parts milli-Q water). The glass beakers should have been cleaned
with milli-Q water, and carefully marked on the outside of the
glass with a marker. Chances are that droplets of acid will dissolve
the marker, so write down the sample code on several places on
the glass. Sometimes, small chips of marker ink will get loose
from the glass and start floating in a water film on the glass
beaker: take care that no such ink
remains end up in your samples. Cover the beakers with clean lids
(watch-glasses), in order to prevent contamination such as dust
entering the sample. Store the samples in a cupboard.
After one night, neutralise the samples by rinsing with milli-Q water through a sieve constructed of cleaned scrynel plankton cambric with mesh size 100 µm (HD, Plato, Diemen, the Netherlands), after which any remaining acid is allowed to dissolve out of the samples by submerging the samples in plenty milli-Q water until the pH of the water is neutral. As an alternative, just decant the fluid without using a sieve, taking care that none of the macrofossil remains are discarded during decanting, and then refill with plenty milli-Q water.
It is difficult, not to say impossible, to precisely measure the pH of milli-Q water (the water does not contain much buffer, and it is this buffer that is used to measure pH by electrodes and indicator paper). We use indicator paper; drop some drops of the sample water on a strip of such paper instead of putting the strip in the water. If the indicated pH is about the same as that of the milli-Q water used, that's okay. This normally means one or two rounds of rinsing. Bigger samples will need more rinsing than fragile samples.
This might look like a strange remark, but perhaps it is better not to wear, e.g., a woollen sweater during this part of the procedure, as such clothes easily shed tiny fibres that could contaminate your samples. Wear a lab coat instead.
Put the material in a petri-dish containing
some milli-Q water so that the samples float but do not move around,
and check for contamination under a binocular microscope (nearly
always, some contamination enters during the acidification and
neutralisation
procedure). Put the selected samples in pre-weighed tin capsules
(pressed, standard weight, size 8 x 5 mm, Elemental Microanalysis
Limited. For weighing use a scale with a precision of at least
0.1 mg). A handy way of doing this is placing two petri dishes
under the microscope. The one under the lens contains the samples
that are to be cleaned and selected, and the one next to that
contains the tin capsule. The tin capsule can be fixed onto the
bottom of the dish by "gluing" it with a drop of clean
water. Put a drop of clean water in the tin capsule so that the
selected macroremains slide more easily from the forceps into
the tin cup. Select clean macroremains and place them in the tin
cup one by one, starting with the cleanest, most "reliable"
samples (the rest can always be added if there appears to be insufficient
material to fill the tin capsule).
Don't fill the entire tin capsule, as often this will result in dry weights more than 10 mg. Such too big samples will need extra work at the AMS lab (reducing into manageable samples). Samples smaller than 1-2 mg dry weight might not give enough material for the AMS. So, try to obtain dry weights of ca. 2-10 mg per sample (5 mg would be perfect). We can't give guidelines as to when a capsule contains just enough material; this appears to depend on density of the material selected. Perhaps it is best to use as much as possible the first time you prepare the samples, and then reduce sample size to 10 mg when determining the dry weight. Next time you will more or less know how much material makes up 10 mg dry weight in your case.
When the capsules are filled with macroremains, put them in empty 100 ml glass beakers (you can use the ones that originally contained the samples, because they will be clean). The beakers are to be covered partly with lids (watch-glasses), in order to prevent contamination such as dust entering the sample, while still enabling moisture to evaporate from the samples. Dry the samples for two nights at 80°C in an oven (preferably shut off any outside air circulation).
Weigh the samples (subtracting the weight of the tin capsule; use a balance with a precision of at least 0.1 mg) and remove material if the samples weigh more than 10 mg (perhaps storing the remainder in a separate capsules for future duplicate purposes). Close the tin capsules securely by folding the tops several times and then forming the capsules into small, dense cubes or balls using clean, dry forceps (no material should leak out through holes, and all air should be pressed out. If by accident the tin capsule starts to leak, it can be wrapped in another tin cup).
Store the capsules in a plastic box with
marked cups ("minesweeper" box). Write your name, institute,
and study site on the box with a marker. Provide information about
the samples on AMS forms that can be downloaded from the AMS pages
at http://www.cio.phys.rug.nl. Also, don't forget to indicate
which sample is put in which cup of the "minesweeper"
box. Indicate this clearly on the AMS forms, and on an additional
list where the information of your samples is summarized. Send
the samples and the forms, securely protected in a jiffy bag,
to:
Dr. Bas van Geel/Dmitri Mauquoy
Institute for Biodiversity and Ecosystem Dynamics
University of Amsterdam
Kruislaan 318
1098 SM Amsterdam
The Netherlands
-The first round of 14C dating
will be subject to quantity, quality and purity control. Collect
material for the 10 range-finder 14C dates. Select
the most suitable material and clean it as thoroughly as possible.
Store the samples in small glass vials containing clean water
with some drops of 4% HCl, pack the vials securely, provide all
information including the AMS forms, and send all this to the
Amsterdam laboratory. Then the AMS 14C dating control
team (Dmitri Mauquoy, Bas van Geel, Hans van der Plicht) will
check for quantity, quality and purity of the samples, and subsequently
perform the AAA analysis, drying, weighing as described above.
For subsequent rounds of 14C dating, it is planned
that the site investigators themselves will perform the procedures
as described above. Decisions about which extra levels are to
be dated, will be taken on the basis of discussions between (1)
the ACCROTELM coordinator (Frank Chambers), (2) the principal
site investigator and (3) the partners responsible for AMS wiggle-match
dating (Bas van Geel, Dmitri Mauquoy and Hans van der Plicht).
A detailed proposal (including diagrams) outlining the levels
to be 14C wiggle-matched for the three time periods
we are interested in should be sent to Dr. Bas van Geel/Dmitri
Mauquoy in the first instance. Based on this information we can
advise the best selection in order to generate a satisfactory
wiggle-match fit. See figures 1-8 for the calibration curve and
corresponding fluctuations of delta 14C during the
relevant time intervals.
References
Aerts-Bijma, A.T., van der Plicht, J.
and Meijer, H.A.J. 2001. Automatic
AMS sample combustion and CO2 collection. Radiocarbon 43, 293-298.
Blaauw, M., Heuvelink, G.B.M., Mauquoy, D., van der Plicht,
J. and van Geel, B., 2003. A numerical approach to 14C
wiggle-match dating of organic deposits: best fits and confidence
intervals. Quaternary Science Reviews 22: 1485-1500.
Blaauw, M., van Geel, B., Mauquoy, D. and van der Plicht, J.,
2004. Carbon-14 wiggle-match dating of peat deposits: advantages
and limitations. Journal of Quaternary Science 19: 177-181.
Kilian, M.R., van der Plicht, J. and van Geel, B. 1995.
Dating raised bogs: new aspects of AMS 14C wiggle matching,
a reservoir effect and climatic change. Quaternary Science Reviews
14, 959-966.
Kilian, M.R., van Geel, B. and van der Plicht, J. 2000.
14C AMS wiggle matching of raised bog deposits and models of peat
accumulation. Quaternary Science Reviews 19, 1011-1033.
Kilian, M.R., van der Plicht, J., van Geel, B. and Goslar,
T. 2001. Problematic 14C -AMS dates of pollen concentrates
from Lake Gosciaz (Poland). Quaternary International 88, 21-26.
Mauquoy, D., van Geel, B., Blaauw, M., Speranza, A. and van
der Plicht, J., 2004. Changes in solar activity and Holocene
climate shifts derived from 14C wiggle-match dated
peat deposits. The Holocene 14: 45-52.
Nilsson, M., Klarqvist, M., Bohlin, E. and Possnert, G. 2001.
Variation in 14C age of macrofossils and different
fractions of minute peat samples dated by AMS. The Holocene 11,
579-586.
Speranza, A.O.M., van der Plicht, J. and van Geel, B. 2000.
Improving the time control of the Subboreal/Subatlantic transition
in a Czech peat sequence by 14C wiggle-matching. Quaternary
Science Reviews 19, 1589-1604.
Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S.,
Hughen, K.A., Kromer, B., McCormac, F.G., van der Plicht, J. and
Spurk, M. 1998. INTCAL98 radiocarbon age calibration, 24,000-0
cal BP. Radiocarbon 40, 1041-1083.
Tomlinson, P. 1985. An aid to the identification of fossil
buds, bud-scales and catkin-bracts of British trees and shrubs.
Circaea 3, 45-130.
van der Plicht, J., van Geel, B., Bohncke, S.J.P., Bos, J.A.A.,
Blaauw, M., Speranza, A.O.M., Muscheler, R. and Björck, S.,
2004. Early Holocene solar forcing of climate change in Europe.
Journal of Quaternary Science 19: 263-269.
Wohlfarth, B., Skog, G., Possnert, G. and Holmquist, B.
1998. Pitfalls in the AMS radiocarbon-dating of terrestrial macrofossils.
Journal of Quaternary Science 13, 137-145.