Hostname: page-component-74656f9884-kv2lp Total loading time: 0 Render date: 2026-11-06T07:40:52.439Z Has data issue: false hasContentIssue false
  • English
  • Français

AN INTERCOMPARISON PROJECT ON 14C FROM SINGLE-YEAR TREE RINGS

Published online by Cambridge University Press:  07 September 2021

Sabrina G K Kudsk Open the ORCID record for Sabrina G K Kudsk [Opens in a new window] ,
Jesper Olsen Open the ORCID record for Jesper Olsen [Opens in a new window] ,
Gregory W L Hodgins ,
Mihály Molnár Open the ORCID record for Mihály Molnár [Opens in a new window] ,
Todd E Lange ,
Jessica A Nordby ,
A J Timothy Jull Open the ORCID record for A J Timothy Jull [Opens in a new window] ,
Tamás Varga Open the ORCID record for Tamás Varga [Opens in a new window] ,
Christoffer Karoff  and
Mads F Knudsen
Sabrina G K Kudsk*
Affiliation:
Institute for Geoscience, Aarhus University, Høegh-Guldbergs Gade 2, DK-8000 Aarhus C, Denmark
Jesper Olsen
Affiliation:
Aarhus AMS Centre (AARAMS), Department of Physics and Astronomy, Aarhus University, Ny Munkegade 120, DK-8000 Aarhus C, Denmark
Gregory W L Hodgins
Affiliation:
University of Arizona AMS Laboratory, Department of Physics, University of Arizona, Tucson, AZ 85721, USA
Mihály Molnár
Affiliation:
Isotope Climatology and Environmental Research Centre (ICER), Institute for Nuclear Research, H-4026 Debrecen, Hungary
Todd E Lange
Affiliation:
University of Arizona AMS Laboratory, Department of Physics, University of Arizona, Tucson, AZ 85721, USA
Jessica A Nordby
Affiliation:
University of Arizona AMS Laboratory, Department of Physics, University of Arizona, Tucson, AZ 85721, USA
A J Timothy Jull
Affiliation:
University of Arizona AMS Laboratory, Department of Physics, University of Arizona, Tucson, AZ 85721, USA Isotope Climatology and Environmental Research Centre (ICER), Institute for Nuclear Research, H-4026 Debrecen, Hungary Department of Geosciences, University of Arizona, Tucson, AZ 85721, USA
Tamás Varga
Affiliation:
Isotope Climatology and Environmental Research Centre (ICER), Institute for Nuclear Research, H-4026 Debrecen, Hungary
Christoffer Karoff
Affiliation:
Institute for Geoscience, Aarhus University, Høegh-Guldbergs Gade 2, DK-8000 Aarhus C, Denmark Stellar Astrophysics Centre, Department of Physics and Astronomy, Aarhus University, Ny Munkegade 120, DK-8000 Aarhus C, Denmark
Mads F Knudsen
Affiliation:
Institute for Geoscience, Aarhus University, Høegh-Guldbergs Gade 2, DK-8000 Aarhus C, Denmark
*
*Corresponding author. Email: sgbn@geo.au.dk
Rights & Permissions [Opens in a new window]

Abstract

A laboratory intercomparison project was carried out on 20 annually resolved late-wood samples from the Danish oak record. The project included the following three laboratories: (1) the University of Arizona AMS Laboratory, University of Arizona, USA (AA); (2) HEKAL AMS Laboratory, MTA Atomki, Hungary (DeA); and (3) Aarhus AMS Centre (AARAMS), Aarhus University, Denmark (AAR). The large majority of individual data points (96%) lie within ±2σ of the weighted mean. Further assessment of the accuracy associated with the individual laboratories showed good agreement, indicating that consistent and reliable 14C measurements well in agreement with each other are produced at the three laboratories. However, the quoted analytical uncertainties appear to be underestimated when compared to the observed variance of differences from the geometric mean of the samples. This study provides a general quality check of the single-year tree-ring 14C measurements that are included in the new calibration curve.

Keywords

IntCalintercomparisonradiocarbon datingsingle-year measurementtree ring

Information

Type
Research Article
Information
Radiocarbon , Volume 63 , Issue 5 , October 2021 , pp. 1445 - 1452
Check for updates
Copyright
© The Author(s), 2021. Published by Cambridge University Press for the Arizona Board of Regents on behalf of the University of Arizona

INTRODUCTION

In the radiocarbon (14C) community, there is a long tradition for proficiency testing of laboratory protocols and 14C analysis in terms of accuracy and measurement variability (e.g. Gulliksen and Scott Reference Gulliksen and Scott1995; Naysmith et al. Reference Naysmith, Scott, Cook, Heinemeier, van der Plicht, Van Strydonck, Ramsey, Grootes and Freeman2007; Scott et al. Reference Scott, Cook and Naysmith2010). Because the international 14C calibration curve is the backbone for almost all 14C dating world-wide, it is of utmost importance to carefully test the accuracy and analytical precision of tree-ring samples submitted as datasets for an updated calibration curve (see Wacker et al. Reference Wacker, Scott, Bayliss, Baillie, Brown, Bard, Bollhalder, Friedrich, Capano and Cherkinsky2020). Two of the three labs in this study (Arizona and Aarhus) have taken part in the large intercomparison study organized by the Swiss AMS group (Wacker et al. Reference Wacker, Scott, Bayliss, Baillie, Brown, Bard, Bollhalder, Friedrich, Capano and Cherkinsky2020), but the present study aimed specifically at demonstrating laboratory performance on single-year wood from the Danish oak record. This could eventually help secure consistent 14C measurements and also strengthen cooperation among laboratories internationally. This is crucial for comparison of data, and for achieving reliable datasets used for future calibration curves.

To validate recently produced 14C measurements on single-year late-wood tree rings, which have been submitted to the new calibration curve, an intercomparison project including the three laboratories HEKAL AMS Laboratory, MTA Atomki, Hungary (DeA), the University of Arizona AMS Laboratory, University of Arizona, USA (AA), and Aarhus AMS Centre (AARAMS), Aarhus University, Denmark (AAR), was performed. The samples used for this intercomparison derive from the Danish oak record, which also provided samples for the updated radiocarbon calibration curve (Reimer et al. Reference Reimer, Austin, Bard, Bayliss, Blackwell, Ramsey, Butzin, Cheng, Edwards and Friedrich2020). The samples were first analyzed at AARAMS and then distributed to HEKAL and the Arizona AMS Laboratory with no prior information on their calibrated or 14C age. The 14C data produced by AARAMS were published by Fogtmann-Schulz et al. (Reference Fogtmann-Schulz, Østbo, Nielsen, Olsen, Karoff and Knudsen2017, Reference Fogtmann-Schulz, Kudsk, Trant, Baittinger, Karoff, Olsen and Knudsen2019) and Kudsk et al. (Reference Kudsk, Philippsen, Baittinger, Fogtmann-Schulz, Knudsen, Karoff and Olsen2019). Several 14C data produced at AA and HEKAL have also been accepted for the new calibration curve (e.g., Jull et al. Reference Jull, Panyushkina, Miyake, Masuda, Nakamura, Mitsutani, Lange, Cruz, Baisan, Janovics, Varga and Molnár2018a; Manning et al. Reference Manning, Griggs, Lorentzen, Ramsey, Chivall, Jull and Lange2018; Pearson et al. Reference Pearson, Brewer, Brown, Heaton, Hodgins, Jull, Lange and Salzer2018).

METHODS

Twenty late-wood samples from two pieces of Danish oak were isolated at AARAMS and sent to the University of Arizona Laboratory, Tucson (Arizona), and HEKAL, Debrecen. The two pieces of wood originate from Ravning Enge and Haderslev Fjord, Denmark, and span the periods 693–702 and 1011–1020 CE, respectively. Oak trees grow by producing annual tree rings composed of an early- (EW) and a late-wood (LW) component. In the beginning of the growing season (spring and early summer), EW is produced and succeeded by LW, which is produced in the late summer (Speer Reference Speer2010). The EW fraction is partially produced from a reserve stored in the tree the previous growing season and may thus contain 14C atoms from the previous year (Pilcher Reference Pilcher1995; Speer Reference Speer2010). In this intercomparison project, only the LW fraction is used as this is the most reliable wood-component for 14C analysis (Kudsk et al. Reference Kudsk, Olsen, Nielsen, Fogtmann-Schulz, Knudsen and Karoff2018). Details on wood cutting, EW and LW separation, and dendrochronology can be found in Fogtmann-Schulz et al. (Reference Fogtmann-Schulz, Kudsk, Trant, Baittinger, Karoff, Olsen and Knudsen2019) and Kudsk et al. (Reference Kudsk, Philippsen, Baittinger, Fogtmann-Schulz, Knudsen, Karoff and Olsen2019). The samples were not homogenized. The possible carry-over effect from the previous growing season is expected to be undetectable in the measured data since the data derive from a time period where the calibration curve is relatively flat, and because the samples solely consist of LW. At all three laboratories, cellulose was extracted from the LW samples prior to 14C analysis, using each laboratory’s own standard extraction method.

Arizona AMS Laboratory (AA)

α-cellulose extraction was performed using the method of Lange et al. (Reference Lange, Nordby, Murphy, Hodgins and Pearson2019). Wood samples were first subjected to a standard acid-base-acid extraction: 3N hydrochloric acid (HCl) at 70°C overnight, type-1 water washes, 1.2N sodium hydroxide (NaOH) at 70°C overnight with additional changes of the NaOH solution and incubation until the wash solution remained clear, type-1 water washes, followed by 3N HCl at 70°C overnight, followed by water washes. The wood was then placed in acidified sodium chlorite (0.3N NaClO2, in 60·10–3 mN HCl), and incubated at 75°C with solution changes until the wood was white in color. This was followed by another round of type-1 water washes. Extraction continued with 17.5% NaOH at room temperature for 2 hr, rinsing to a neutral pH with type-1 water, a final acid wash with 3N HCl for 1 hr, and a finishing rinsing to a neutral pH with type-1 water. Holo-cellulose purified from Brown Coal (Tertiary Age, from S. Levitt, Laboratory for Tree-Ring Research, University of Arizona) was used as process blank. Combustion, graphitization, and AMS measurement on a 3MV NEC 9SDH Pelletron AMS running at 2.5MV were carried out as described by Jull et al. (Reference Jull, Burr, Beck, Hodgins, Biddulph, Gann, Hatheway, Lange and Lifton2006).

Hertelendi Laboratory of Environmental Studies, HEKAL (DeA)

Holo-cellulose was extracted using a base-acid-base-acid-bleaching procedure (BABAB) (method fully described in Molnár et al. Reference Molnár, Janovics, Major, Orsovszki, Gönczi, Veres, Leonard, Castle, Lange, Wacker, Hajdas and Jull2013a). Samples were alternately suspended in 4% (1N) NaOH and 4% HCl two times, and lastly treated with 5% NaClO2 acidified with 4% HCl. All steps were performed at 75°C.

HEKAL has run IAEA-C9 (fossil wood) samples together with the tree-ring samples to monitor the process blank level and other exogenous contamination. The cellulose was combusted in sealed glass tubes containing MnO2, and then converted to CO2, cryogenically purified, and graphitized by a sealed-tube graphitization method (Rinyu et al. Reference Rinyu, Molnár, Major, Nagy, Veres, Kimák, Wacker and Synal2013; Janovics et al. Reference Janovics, Futó and Molnár2018). The graphite targets were analyzed in the EnvironMICADAS AMS system at HEKAL AMS Laboratory, ATOMKI, Debrecen, Hungary (Molnár et al. Reference Molnár, Rinyu, Veres, Seiler, Wacker and Synal2013b).

Aarhus AMS Centre, AARAMS (AAR)

α-cellulose was extracted from wood samples using a combination of the methods proposed by Loader et al. (Reference Loader, Robertson, Barker, Switsur and Waterhouse1997) and Southon and Magana (Reference Southon and Magana2010), summarized in Fogtmann-Schultz et al (Reference Fogtmann-Schulz, Kudsk, Adolphi, Karoff, Knudsen, Loader, Muscheler, Trant, Østbø and Olsen2020). The wood samples were first bleached at 70°C with 1N NaClO2 acidified with 1N HCl, then treated at room temperature with 17% NaOH to dissolve hemicelluloses. Lastly, samples were treated with 1N HCl at room temperature to remove any contamination from atmospheric CO2. Between each part of the pretreatment, the samples were rinsed with Milli-Q water to neutral pH. Kauri wood from Marine Isotope Stage 7 provided by P. Reimer at Queen’s University Belfast was pretreated along with the wood samples, and used as a process blank. Subsequently, the α-cellulose was sealed in quartz tubes containing copper oxide (CuO) and combusted to CO2 at 900°C, cryogenically purified, and then reduced to graphite using hydrogen and iron as catalyst. All samples were analyzed using the 1MV High Voltage Engineering AMS system at the Aarhus AMS Centre, Aarhus University, Denmark (Olsen et al. Reference Olsen, Tikhomirov, Grosen, Heinemeier and Klein2017).

All laboratories use international secondary standards, such as IAEA-C3, IAEA-C7 and IAEA-C8, as well as wood-blank materials, including charcoal, oak (see e.g., Jull et al Reference Jull, Pearson, Taylor, Southon, Santos, Kohl, Hajdas, Molnar, Baisan, Lange, Cruz, Janovics and Major2018b; Fogtmann-Schultz et al. Reference Fogtmann-Schulz, Kudsk, Trant, Baittinger, Karoff, Olsen and Knudsen2019; Kudsk et al. Reference Kudsk, Philippsen, Baittinger, Fogtmann-Schulz, Knudsen, Karoff and Olsen2019) and Brown Coal. At AARAMS, 14C ages are calculated using an in-house graphical normalization program in Matlab, whereas Arizona uses the in-house software and methods described in Donahue et al. (Reference Donahue, Linick and Jull1990) and Burr et al. (Reference Burr, Donahue, Tang, Beck, McHargue, Biddulph, Cruz and Jull2007). HEKAL uses the standard BATS software for sample calculation and data reduction (Wacker et al. Reference Wacker, Christl and Synal2010). Assessment of secondary standards and blanks for the AARAMS analysis can be found in Fogtmann-Schultz et al. (Reference Fogtmann-Schulz, Kudsk, Trant, Baittinger, Karoff, Olsen and Knudsen2019) and Kudsk et al. (Reference Kudsk, Philippsen, Baittinger, Fogtmann-Schulz, Knudsen, Karoff and Olsen2019), while for Arizona, Brown Coal holocellulose process blanks were measured with a F14C value of 0.0017 ± 0.0004 (n=8). The geometric mean and scatter of the processed IAEA-C9 samples (F14C = 0.0016 ± 0.0003, n=3) and IAEA-C3 samples (F14C = 1.2929 ± 0.0019, n=3) from HEKAL were in good agreement with their consensus values. Other intercomparisons of HEKAL wood preparation and AMS 14C analyses were recently reported by Jull et al. (Reference Jull, Pearson, Taylor, Southon, Santos, Kohl, Hajdas, Molnar, Baisan, Lange, Cruz, Janovics and Major2018b). For all three laboratories, the uncertainty of the primary standard (OX-II) and blank samples is included in the analytical result using propagation of errors.

RESULTS AND DISCUSSION

The average cellulose extraction yield of the three laboratories is very similar, amounting to 39.5%, 39.7%, and 38.5% for the Arizona AMS Laboratory, HEKAL, and AARAMS, respectively. For each year, we calculated the weighted average 14C age, and a reduced χ2 test is used to assess the validity of the weighted average values. The deviations from the weighted average in units of standard deviations (z-scores) are calculated for each laboratory and plotted in Figure 2, along with the deviation mean for each individual laboratory. One sample, AAR-27731, differs more than 6σ from the weighted average values and is regarded as an outlier, and it will therefore not be discussed further. All data are presented in Table 1 and in Figure 1.

Table 1

First column shows the sample calendar year in CE. Columns with Lab ID denote ID of the particular measurement and laboratory code. 14C age (BP) denote the measured carbon 14C age in years before present (BP). Z-score is the deviation in units of standard deviations from the calculated weighted average of the particular year, which is given in the column weighted mean (μw). Reduced χ2 statistics are provided as the χ2 value of the sample, which is to be compared with the limiting χ2 value calculated at 95% confidence interval. The geometric mean (μ) and the individual sample difference from the geometric mean (Δμ) are furthermore shown. AAR data are previously published in Kudsk et al. (Reference Kudsk, Philippsen, Baittinger, Fogtmann-Schulz, Knudsen, Karoff and Olsen2019). Outliers are marked with *.

Figure 1

Measured 14C ages from each laboratory. Figure A shows results of the wood piece from Ravning Enge (693–702 CE) and Figure B of the piece from Haderslev Fjord (1011–1020 CE).

Overall, the 14C results from the three laboratories agree within statistical error (Figure 1). Three samples have z-scores equal to or higher/lower than ±2σ, corresponding to 4.4% of the total samples (Table 1). A total of 5% of the samples may be expected to be outside the ±2σ range if normally distributed. The reduced χ2 values range from 0 to 3.3, and three samples year (697 CE; χ2: 3.3≤3.0, 698 CE; χ2: 3.2≤3.0 and CE 700; χ2: 2.9≤2.6) fail the reduced χ2 test. This may indicate that the 14C age variability of these three samples is too large, and thus that the individual measurements are not statistically agreeing (Table 1). However, at the 95% significance level, 5% of the samples are expected to fail the χ2 test (a total of 5.0% failed). If tested at the 97% significance level, all samples pass the χ2 test. Thus, no obvious outliers can be observed in the dataset apart from AAR-27731 (Figure 2). The average value of the reduced χ2 values is 1.6 with a standard deviation of 1.0. Normally distributed samples would display a reduced χ2 value around 1 with a χ2 standard deviation of 1.0. In the present dataset, 40% of an expected 70% of the samples show reduced χ2 values lower than or equal to 1.2, suggesting that even though they pass the reduced χ2 test the sample variance for individual years is higher than expected in 30% of the cases. This may indicate that the analytical errors most likely are underestimated.

Figure 2

z-scores of the measurements from the three laboratories. Horizontal colored lines indicate the mean of the z-scores for each laboratory. Figure A shows z-scores for the wood piece from Ravning Enge (693-702 CE) and Figure B of the piece from Haderslev Fjord (1011–1020 CE). The outlier AAR-27731 has been omitted.

The average analytical error for samples from Arizona (AA) is ±28 years (±3.4‰), while HEKAL and AARAMS both yielded ±16 14C years (±2.0‰) (Table 1). To evaluate the mean measurement uncertainty, we calculate the geometric mean (μ) of the results, and subsequently the mean and standard deviation of the difference (Δμ) from the geometric mean (Table 1). For the Arizona AMS Laboratory, the standard deviation of the differences from the geometric mean is ±23 14C years, which is lower than the quoted analytical error. For HEKAL and AARAMS, the standard deviation of the differences from the geometric mean is ±18 and ±20 14C years, respectively (Table 1). This may suggest that HEKAL and AARAMS underestimate the analytical errors with 2 and 4 14C years, respectively. However, if the analysis only considers HEKAL and AARAMS, then the differences from the geometric mean would be ±16 and ±17 14C years, respectively, and thus in better agreement with the quoted analytical uncertainties for both laboratories. Due to the limited number of measurements of each sample, it is difficult to assess the specific cause behind the missing error, and if the high variability is associated with one or more laboratories. The missing error could originate from the data reduction calculations, but it could also relate to an unknown variability in sample preparation or graphitization protocols.

It is clear, however, that the average reduced χ2 value of 1.6 suggests that the errors generally are underestimated. If we use the average reduced χ2 value to revise the common error of the complete dataset, then errors should increase by a factor of $$\sqrt {1.6} $$ , or 1.26. This would correspond to mean errors of 35, 20, and 20 14C years for Arizona, HEKAL, and AARAMS, respectively. The mean values of the difference from the geometric mean of 6, –3, and –2 14C years, along with mean z-score values of 0.2, –0.1, and 0.1, indicate that all three laboratories accurately measure the 14C age with no systematic errors (Table 1). The standard deviation of all z-scores from each laboratory is around ±1, further indicating that the quoted mean values are normally distributed within errors (Figure 2).

This study shows that the three laboratories, using different chemical pretreatments, instruments, and data-handling procedures, are able to produce consistent and reliable 14C measurements that are in good agreement with each other. Further, the statistical outcome from the high-precision datasets from HEKAL and AARAMS, obtained using different instruments, are very similar (Table 1). Although the laboratories have used different cellulose extraction methods, there appears to be no systematic influence on the 14C measurements or the cellulose yield, which is in agreement with previous work (e.g., Němec et al. Reference Němec, Wacker, Hajdas and Gäggeler2010). This small intercomparison study was conducted to evaluate the laboratory procedures for producing single-year 14C data for the international calibration curve. The IntCal working group assesses the quality (analytical errors and agreement among different datasets) of all datasets that are considered for inclusion in a forthcoming calibration curve. The discussion presented here therefore highlights a need for further reducing the scatter and improving estimates of the associated uncertainties.

CONCLUSION

An intercomparison project involving the three laboratories: (1) University of Arizona AMS Laboratory, Arizona; (2) HEKAL, Debrecen; and (3) AARAMS, Aarhus, has been carried out on two pieces of Danish oak covering a total of 20 years. Using reduced χ2 statistics and individual sample differences from the geometric mean, it can be concluded that the three laboratories are able to produce accurate and reliable 14C measurements. The quoted analytical uncertainties appear to be underestimated when compared to the observed variance of differences from the geometric mean. Furthermore, no systematic effects related to the sample pretreatment, type of instrument, modes of operation, and data handling could be observed.

ACKNOWLEDGMENTS

We are grateful for the support of the technical staff at all three laboratories. Sabrina Gjødvad Kaiser Kudsk, Mads Faurschou Knudsen and Christoffer Karoff sincerely acknowledge the financial support from the Villum Foundation. This work was partly supported by the European Union and the State of Hungary, co-financed by the European Regional Development Fund in the project of GINOP-2.3.2-15-2016-00009 “ICER”.

References

REFERENCES

Burr, GS, Donahue, DJ, Tang, Y, Beck, JW, McHargue, L, Biddulph, D, Cruz, R, Jull, AJT. 2007. Error analysis at the NSF-Arizona AMS facility. Nuclear Instruments and Methods in Physics Research B 259:149–153.CrossRefGoogle Scholar
Donahue, DJ, Linick, TW, Jull, AJT. 1990. Isotope-ratio and background correction for accelerator ass spectrometry radiocarbon measurements. Radiocarbon 32(2):135–142.CrossRefGoogle Scholar
Fogtmann-Schulz, A, Østbo, SM, Nielsen, SGB, Olsen, J, Karoff, C, Knudsen, MF. 2017. Cosmic ray event in 994 CE recorded in radiocarbon from Danish oak. Geophysical Research Letters 44(16):8621–8628.CrossRefGoogle Scholar
Fogtmann-Schulz, A, Kudsk, SGK, Trant, PLK, Baittinger, C, Karoff, C, Olsen, J, Knudsen, MF. 2019. Variations in solar activity across the Sporer minimum based on radiocarbon in Danish oak. Geophysical Research Letters 46(15):8617–8623.CrossRefGoogle Scholar
Fogtmann-Schulz, A, Kudsk, SGK, Adolphi, F, Karoff, C, Knudsen, MF, Loader, NJ, Muscheler, R, Trant, PLK, Østbø, SM, Olsen, J. 2020. Batch processing of tree ring samples for radiocarbon analysis. Radiocarbon 63(1):77–89.CrossRefGoogle Scholar
Gulliksen, S, Scott, M. 1995. Report of the TIRI workshop, Saturday 13 August 1994. Radiocarbon 37(2):820–821.CrossRefGoogle Scholar
Kudsk, SGK, Olsen, J, Nielsen, L, Fogtmann-Schulz, A, Knudsen, MF, Karoff, C. 2018. What is the carbon origin of early-wood? Radiocarbon 60(5):1457–1464.CrossRefGoogle Scholar
Kudsk, SGK, Philippsen, B, Baittinger, C, Fogtmann-Schulz, A, Knudsen, MF, Karoff, C, Olsen, J. 2019. New single-year radiocarbon measurements based on Danish oak covering the periods AD 692–790 and 966–1057 AD. Radiocarbon 62(4):969–987.CrossRefGoogle Scholar
Janovics, R, Futó, I, Molnár, M. 2018. Sealed tube combustion method with MnO2 for AMS 14C Measurement. Radiocarbon 60(5):1347–1355.CrossRefGoogle Scholar
Jull, AJT, Burr, GS, Beck, JW, Hodgins, GWL, Biddulph, DL, Gann, J, Hatheway, AL, Lange, TE, Lifton, NA. 2006. Application of accelerator mass spectrometry to environmental and paleoclimate studies at the University of Arizona. Elsevier, Radioactivity in the Environment 8:3–23.CrossRefGoogle Scholar
Jull, AJT, Panyushkina, I, Miyake, F, Masuda, K, Nakamura, T, Mitsutani, T, Lange, TE, Cruz, RJ, Baisan, C, Janovics, R, Varga, T, Molnár, M. 2018a. More rapid 14C excursions in the tree-ring record: a record of different kind of solar activity at about 800 BC? Radiocarbon 60(4):1237–1248.CrossRefGoogle Scholar
Jull, AJT, Pearson, CL, Taylor, RE, Southon, JR, Santos, GM, Kohl, CP, Hajdas, I, Molnar, M, C Baisan, C, Lange, TE, Cruz, R, Janovics, R, Major, I. 2018b. Radiocarbon dating and intercomparison of some early historical radiocarbon samples. Radiocarbon 60(2):535–548.CrossRefGoogle Scholar
Lange, TE, Nordby, JA, Murphy, PLO, Hodgins, GWL, Pearson, CL. 2019. A detailed investigation of pretreatment protocols for high precision radiocarbon measurements of annual tree-rings. Nuclear Instruments and Methods in Physics Research B 455:230–233.CrossRefGoogle Scholar
Loader, NJ, Robertson, I, Barker, AC, Switsur, VR, Waterhouse, JS. 1997. An improved technique for the batch processing of small wholewood samples to α-cellulose. Chemical Geology 136:313–317.CrossRefGoogle Scholar
Manning, SW, Griggs, C, Lorentzen, B, Ramsey, CB, Chivall, D, Jull, AJT, Lange, TE. 2018. Fluctuating radiocarbon offsets observed in the southern Levant and implications for archaeological chronology debates. Proceedings of the National Academy of Sciences of the United States of America 115(24):6141–6146.CrossRefGoogle ScholarPubMed
Molnár, M, Janovics, R, Major, I, Orsovszki, J, Gönczi, R, Veres, M, Leonard, A G, Castle, S M, Lange, T E, Wacker, L, Hajdas, I, Jull, AJT. 2013a. Status report of the new AMS 14C sample preparation lab of the Hertelendi Laboratory of Environmental Studies (Debrecen, Hungary). Radiocarbon 55(2–3):665–676.CrossRefGoogle Scholar
Molnár, M, Rinyu, L, Veres, M, Seiler, M, Wacker, L, Synal, H-A. 2013b. EnvironMICADAS: a mini 14C AMS with enhanced gas ion source interface in the Hertelendi Laboratory of Environmental Studies (HEKAL), Hungary. Radiocarbon 55(2–3):338–344.CrossRefGoogle Scholar
Naysmith, P, Scott, EM, Cook, GT, Heinemeier, J, van der Plicht, J, Van Strydonck, M, Ramsey, CB, Grootes, PM, Freeman, S. 2007. A cremated bone intercomparison study. Radiocarbon 49(2):403–408.CrossRefGoogle Scholar
Němec, M, Wacker, L, Hajdas, I, Gäggeler, H. 2010. Alternative methods for cellulose preparation for AMS measurement. Radiocarbon 52(3):1358–1370.CrossRefGoogle Scholar
Olsen, J, Tikhomirov, D, Grosen, C, Heinemeier, J, Klein, M. 2017. Radiocarbon analysis on the new AARAMS 1MV Tandetron. Radiocarbon 59(3):905–913.CrossRefGoogle Scholar
Pearson, CL, Brewer, PW, Brown, D, Heaton, TJ, Hodgins, GW, Jull, AT, Lange, T, Salzer, MW. 2018. Annual radiocarbon record indicates 16th century BCE date for the Thera eruption. Science Advances 4(8):eaar8241.CrossRefGoogle ScholarPubMed
Pilcher, JR. 1995. Biological considerations in the interpretation of stable isotope ratios in oak tree-rings. In: Frinzel B, Stauffer B, Weiss MM, editors. Paläoklimaforschung/Paleoclimate Research 15. European Science Foundation, Strasbourg, France (1993). p 157–161.Google Scholar
Reimer, PJ, Austin, WEN, Bard, E, Bayliss, A, Blackwell, PG, Ramsey, CB, Butzin, M, Cheng, H, Edwards, RL, Friedrich, M, et al. 2020. The IntCal20 Northern Hemisphere radiocarbon age calibration curve (0–55 cal kBP). Radiocarbon 62(4):725–757. doi: 10.1017/RDC.2020.41.CrossRefGoogle Scholar
Rinyu, L, Molnár, M, Major, I, Nagy, T, Veres, M, Kimák, Á, Wacker, L, Synal, H-A. 2013. Optimization of sealed tube graphitization method for environmental C-14 studies using MICADAS, Nuclear Instruments and Methods in Physics Research B 294:270–275.CrossRefGoogle Scholar
Scott, EM, Cook, GT, Naysmith, P. 2010. The Fifth International Radiocarbon Intercomparison (VIRI): an assessment of laboratory performance in Stage 3. Radiocarbon 52(3):859–865.CrossRefGoogle Scholar
Southon, JR, Magana, AL. 2010. A comparison of cellulose extraction and ABA pretreatment methods for AMS 14C dating of ancient wood. Radiocarbon 52(3):1371–1379.CrossRefGoogle Scholar
Speer, JH. 2010. Fundamentals of tree-ring research. University of Arizona Press. 509 p. Google Scholar
Wacker, L, Christl, M, Synal, H-A. 2010. BATS: a new tool for AMS data reduction. Nuclear Instruments and Methods in Physics Research B 268(7–8):976–979.CrossRefGoogle Scholar
Wacker, L, Scott, EM, Bayliss, A, Baillie, M, Brown, D, Bard, E, Bollhalder, S, Friedrich, M, Capano, M, Cherkinsky, A, et al. 2020. Findings from an in-depth annual tree-ring radiocarbon intercomparison. Radiocarbon 62(4):873–882.CrossRefGoogle Scholar
Figure 0

Table 1 First column shows the sample calendar year in CE. Columns with Lab ID denote ID of the particular measurement and laboratory code. 14C age (BP) denote the measured carbon 14C age in years before present (BP). Z-score is the deviation in units of standard deviations from the calculated weighted average of the particular year, which is given in the column weighted mean (μw). Reduced χ2 statistics are provided as the χ2 value of the sample, which is to be compared with the limiting χ2 value calculated at 95% confidence interval. The geometric mean (μ) and the individual sample difference from the geometric mean (Δμ) are furthermore shown. AAR data are previously published in Kudsk et al. (2019). Outliers are marked with *.

Figure 1

Figure 1 Measured 14C ages from each laboratory. Figure A shows results of the wood piece from Ravning Enge (693–702 CE) and Figure B of the piece from Haderslev Fjord (1011–1020 CE).

Figure 2

Figure 2 z-scores of the measurements from the three laboratories. Horizontal colored lines indicate the mean of the z-scores for each laboratory. Figure A shows z-scores for the wood piece from Ravning Enge (693-702 CE) and Figure B of the piece from Haderslev Fjord (1011–1020 CE). The outlier AAR-27731 has been omitted.