Save

Dendroarchaeology of Birnirk and Thule Architectural Timbers (10–13th Centuries)

Preliminary Results on Oxygen Isotope Cross-Dating

In: International Journal of Wood Culture
Authors:
Juliette Taïeb Université Paris 1 Panthéon-Sorbonne, CNRS UMR 7041 ArScAn, Archéologies environnementales Paris France

Search for other papers by Juliette Taïeb in
Current site
Google Scholar
PubMed Close
https://orcid.org/0000-0001-7074-8055
,
Valérie Daux Université Versailles Saint-Quentin, IPSL, CNRS UMR 8212 Laboratoire des Sciences du Climat et de l’Environnement Gif sur Yvette France

Search for other papers by Valérie Daux in
Current site
Google Scholar
PubMed Close
https://orcid.org/0000-0002-8643-260X
, and
Claire Alix Université Paris 1 Panthéon-Sorbonne, CNRS UMR 8096 Archéologie des Amériques Paris France

Search for other papers by Claire Alix in
Current site
Google Scholar
PubMed Close
https://orcid.org/0000-0003-4285-4592
Online Publication Date:
23 Feb 2024
Open Access

Abstract

Along the coast of northwestern Alaska, architectural wood remains are well preserved in the Birnirk and Thule coastal sites of the early 2nd millennium CE. These structural wood elements are unique archives for documenting climatic variations and cultural transformations during this key development period of Inuit culture. Along this treeless Arctic coast, driftwood accumulates from the subarctic forests of interior Alaska. Except for northwestern Alaska, regional tree-ring chronologies are too short (at best 350–400 years) to successfully date archaeological wood remains from Birnirk and Thule coastal sites using conventional dendrochronology. This paper examines the potential of tree-ring derived δ18O signal to annually date eight architectural wood samples from the Rising Whale (KTZ-304) site at Cape Espenberg, northwestern Alaska. We developed a δ18O master chronology, covering the period 935–1157 CE, using five wood samples from the KTZ-304 site. Blind isotope cross-dating of individual series belonging to this δ18O master chronology (one against the other four) showed conclusive dating and a very strong coherence of the isotopic signal. We, then, used the δ18O master chronology to cross-date three other wood samples for which we knew, from previous 14C wiggle-matching procedure, the first measured ring to be in this time interval, within a ± 18 to 30-year precision. Oxygen isotope dendrochronology provided a plausible date for one of the samples (the first measured ring at 1073 CE). This preliminary study encourages us to acquire additional data to extend in time and strengthen the δ18O master chronology of northwestern Alaska (NWAK18O) and help refine our understanding of climate and culture change during the 2nd millennium CE.

Keywords: archaeology; coastal driftwood; dendrochronology; stable isotope; Thule; Western Arctic

1 Introduction

The coasts of northern Alaska contain well-preserved structural wood elements from Birnirk and Thule semi-subterranean houses from the 2nd millennium CE. Archaeologists identify the Birnirk and Thule cultures as direct ancestors of the present-day Iñupiaq culture of northwestern Alaska and Inuit in general (Raghavan et al. 2014; Mason 2020; Unkel et al. 2022). Birnirk and Thule cultures are present in the Bering Strait region during the Medieval Climate Anomaly (MCA) (9th–13th centuries) and the transition to the Little Ice Age (LIA) (13th–15th centuries) (Alix et al. 2015; Mason 2016, 2020). In this treeless tundra environment, people had access to driftwood carried by the main rivers and ocean currents to northern Alaskan beaches (Giddings 1952b; Alix & Brewster 2004; Alix 2009, 2012, 2016). According to the geometry of marine currents and local knowledge, these driftwood logs came mainly from the forest of Interior Alaska via the Yukon and Kuskokwim rivers and that of northwestern Alaska via the Kobuk and Noatak rivers (Fig. 1). To a far lesser extent, in northern Alaska, wood may have also come from northeastern Alaska via northern flowing rivers, such as the Firth River, and northwestern Canada, primarily via the Mackenzie River (Giddings 1943, 1952a; Eurola 1971; Eggertsson 1995; Alix & Koester 2001; Alix 2004, 2005; Alix & Brewster 2004; Hellmann et al. 2017; Taïeb et al. 2022).

Map of Alaska and part of Yukon Territory in Canada, with the location of Rising Whale (KTZ-304) site at Cape Espenberg and the locations of three existing millennial master tree-ring chronologies
Figure 1

Map of Alaska and part of Yukon Territory in Canada, with the location of Rising Whale (KTZ-304) site at Cape Espenberg and the locations of three existing millennial master tree-ring chronologies

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

Coastal architectural wood remains are unique high-resolution chronometric and paleoclimatic archives. In Alaska, they can be used to (i) refine the chronological and temporal framework for archaeological sites and related cultures in a region where dating techniques face methodological limitations (Morrison 1989, 2001; Nash 2000; Mason 2009, 2020; Mason & Bowers 2009; Krus et al. 2019; Taïeb et al. 2023) and (ii) document and characterize climatic variations with a yearly resolution in their regions of origin, where climate proxies are rare and have limited temporal resolution (Mason & Gerlach 1995; Nicolle et al. 2018; Degroot et al. 2021, 2022).

In northern Alaska, conventional dendrochronology (analysis of annual ring-width patterns) successfully dated and determined the origin of coastal archaeological wood in a limited number of cases (Giddings 1943, 1948, 1952a; Oswalt 1951; VanStone 1958; Barber 2003; Griggs et al. 2019; Taïeb et al. 2022, 2023). Long tree-ring master chronologies in the source regions of driftwood are required to cross-date archaeological wood remains, in other words, to synchronize ring-width series patterns (Douglass 1921; Giddings 1952a; Speer 2012; Eggertsson 1994; Hellmann et al. 2016). Today, millennium-long master chronologies are limited to the Kobuk and Firth Rivers in northwestern and northeastern Alaska (978–2002 CE and 1067–2002 CE, respectively) and the Mackenzie River in northwestern Canada (1245–2006 CE) (Fig. 1) (Giddings 1948, 1952a,b; Graumlich & King 1997; D’Arrigo, Mashing, et al. 2005; D’Arrigo, Wilson, et al. 2006; Porter et al. 2013). These time series are based on white spruce (Picea glauca (Moench Voss)), which is the dominant species of the boreal forest in Alaska (Viereck & Little 2007) and the main species identified in driftwood accumulations and coastal archaeological sites (Giddings 1952b; Hopkins & Giddings 1953, 22; Alix 2005, 2016; Griggs et al. 2019). So far, master chronologies of interior Alaskan areas, where a large part of the driftwood originates, are at best 400–350-year long. Therefore, they are too short to cross-date archaeological timbers from Birnirk and Thule coastal sites, and a large portion of sampled structural wood elements can not be cross-dated. When these cross-date with each other, they can only be considered as “floating” — that is, undated — chronologies (Giddings 1948; Taïeb et al. 2022, 2023).

Therefore, our aim is to develop annually dated tree-ring chronologies that can be used for (i) dating archaeological wood from coastal Inuit sites and (ii) analysing climatic variations during the MCA and LIA. Recent papers demonstrated the potential for oxygen isotope (δ18O) dendrochronology to date living trees and archaeological wood remains at a yearly resolution and develop tree-ring master chronologies where conventional dendrochronology is limited (Roden 2008; Xu et al. 2013; Baker et al. 2015; Yamada et al. 2018; Loader et al. 2019, 2021, 2022). Furthermore, δ18O dendrochronology is based on the same principles as conventional dendrochronology but relies on cross-dating δ18O rather than tree-ring width (TRW) variations, the δ18O signal being often strongly consistent between trees and coherent at larger spatial scales than the ring-width signal (Roden 2008; Xu et al. 2013; Loader et al. 2019, 2022).

The present study complements another one in which we explored 14C wiggle-matching as a high-resolution chronometric approach to precisely date “floating” sequences of archaeological Birnirk and Thule structural wood remains (Taïeb et al. 2023). Here, based on a methodology developed by Loader et al. (2019), we explore the potential of the δ18O cross-dating approach for dating, at an annual resolution, floating tree-ring series from Birnirk and Thule wood samples. In Alaska, the δ18O values of annual and monthly precipitation show a large-scale geographic homogeneity (West et al. 2010). The geographical gradient of the isotopic signal is perceptible from south to northeast Alaska (Isomap, ©Bowen 2003–2022, Waterisotopes Database (2023) http://waterisotopesDB.org (accessed 2 June 2023). Therefore, we tested the capacity of the δ18O chronology developed in northwestern Alaska to cross-date the δ18O series of archaeological wood of unknown geographical origins, which could not be dated with conventional dendrochronology. We hypothesized that the δ18O signal for trees growing in different regions, but influenced by the same mode of climate variability, was sufficiently close. To test this hypothesis, we first evaluated the coherence of the δ18O signal using five wood samples dated by conventional dendrochronology. We, then, built a δ18O reference chronology with these five individual series (NWAK18O hereafter) and, finally, tested the δ18O–based dating approach on three “floating” wood series.

2 Materials and Methods

2.1 Archaeological Site and Wood Sample Selection

The wood samples used to develop NWAK18O and test the δ18O–based dating approach were selected from structural remains of a semi-subterranean house (F-12) at the Rising Whale (KTZ-304) site, Cape Espenberg, in the Bering Land Bridge National Preserve (BELA) (Figs 1 and 2; Fig. A1 in the Appendix) (Alix et al. 2018). Dated with radiocarbon and dendrochronology, the feature is a late Birnirk house occupied in the early 13th century CE (Alix et al. 2018, 2020b). The multi-rooms house has a wood structure with occasional whale bones and most likely a sod cover. Field observations and architectural analysis (including tree-ring analysis) suggest that the frame of the house represents one main building episode (Fig. 2) (Alix et al. 2018, 2020a,b).

Excavation map of the floor of the Birnirk house F-12, Rising Whale site (KTZ-304), Cape Espenberg (field recording C. Alix, CAD S. Eliès and C. Alix). Each type of wooden architectural element is assigned a colour: shades of blue, brown, green and grey-brown for the roof logs, wall logs, wood bench and wooden floor, respectively
Figure 2

Excavation map of the floor of the Birnirk house F-12, Rising Whale site (KTZ-304), Cape Espenberg (field recording C. Alix, CAD S. Eliès and C. Alix). Each type of wooden architectural element is assigned a colour: shades of blue, brown, green and grey-brown for the roof logs, wall logs, wood bench and wooden floor, respectively

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

Cross-sections were collected from structural logs and planks during field excavations in 2016 and 2017 (Fig. A1 in the Appendix; Alix et al. 2017, 2020a,b; Alix & Mason 2018). We analysed the ring-width series of sixty-eight Picea glauca cross-sections following standard dendrochronological methods (e.g. Cook & Kairiukstis 1990; Speer 2012). Based on TRW patterns, we grouped the sixty-eight cross-sections into distinct sequences: five sequences of two to four individuals and one sequence of sixteen individuals (hereafter F12_C2). We successfully cross-dated seventeen individual TRW series, including those grouped as F12_C2, by using the TRW Kobuk River master chronology (Giddings 1952a; Graumlich & King 1997; D’Arrigo et al. 2005). However, 51 of the 68 individual TRW series could not be dated using conventional dendrochronology. Five wood cross-sections used to build the F12_C2 sequence were selected to develop an exploratory δ18O chronology, NWAK18O (Table 1; Fig. A1 in the Appendix). The selection criteria included: (i) overlap of periods and (ii) high state of preservation. Hereafter, the five series dated at a yearly resolution by conventional dendrochronology and used to develop NWAK18O are individually referred to as δ18O–ref series.

Selected samples used for geochemical and oxygen isotope analyses (reference and test δ18O individual series)
Table 1

Selected samples used for geochemical and oxygen isotope analyses (reference and test δ18O individual series)

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

To evaluate the potential of the δ18O cross-dating approach, we selected three additional undated wood cross-sections. The TRW chronologies of these cross-sections do not cross-date with each other and are part of three distinct tree-ring sequences (Table 1; Fig. A1 in the Appendix). We selected these cross-sections from a set of eight archaeological wood samples previously placed in calendar time using radiocarbon dating and wiggle-matching, with an uncertainty of a few decades (Table 1) (Taïeb et al. 2023). We selected these three samples on the following criteria: (i) they could neither be dated with dendrochronology nor provenanced, (ii) they were contemporaneous with NWAK18O based on wiggle-matching dating, (iii) their length (>70 rings) ensured a minimum overlap of fifty years with NWAK18O, and (iv) they were well preserved. Hereafter, the three δ18O series corresponding to these undated wood cross-sections are referred to as the δ18O–test series.

2.2 TRW and δ18O Determination

The selected cross-sections were kept in a freezer following their excavation and freeze-dried before processing (24 h at –25°C). The radii were cut out of the cross-sections with a mechanical bandsaw, manually sanded (grain size of 50–600 grit), and measured to a 0.001 mm resolution with a Velmex measurement table. We used TSAP-WIN and COFECHA software programs to calibrate and verify the tree-ring measurements (Holmes 1983; RinnTech 2000).

The isotopic analyses were conducted at the Laboratoire des Sciences du Climat et de l’Environnement (LSCE, Gif sur Yvette, France). For each radius, the rings were individually split using a scalpel under a binocular magnifier (×10/×20). Each ring was placed in a bag made from a Polytetrafluoroethylene (PTFE) filter (47 mm diameter, 10 µm porosity), closed by a PTFE ribbon to which a PTFE label was attached. We extracted wood α-cellulose from the whole ring of each annual growth ring following the Jayme-Wise method of delignification with an acidified sodium chlorite solution (NaClO2), purification with sodium hydroxide (NaOH), and neutralization with acetic acid over three days (Leavitt & Danzer 1993; Boettger et al. 2007). After drying in an oven for two days, the cellulose bags were transferred to individual Eppendorf tubes with deionized water and placed in a refrigerator overnight. Then, α-cellulose was homogenized ultrasonically with a sonotrode apparatus (max. 2 min/sample) and freeze-dried (24 h at –25°C).

The cellulose of each ring was, then, weighted in silver capsules (between 10–20 µg/sample), and the isotopic composition of oxygen of α-cellulose was determined by pyrolysis (1400°C) with a High-Temperature Conversion Elemental Analyzer (TC-EA, Thermo Scientific) associated with a mass spectrometer (IRMS, IsoPrime). Each cellulose sample was measured at least twice and up to four times. A reference material (Whatmann CC31), which was used for an intercomparison between nine European laboratories (Boettger et al., 2007), was included in the analysis sequences. The δ18O of CC31 (set by consensus between the nine laboratories) is 31.85‰. A one-point calibration was performed using these reference values. The long-term analytical precision based on repeated measurements of CC31 cellulose was within 0.20‰. The study precision based on the aggregates of repeated measures of study samples is equal to 0.15‰ (see Brooks et al. 2022, for calculation details and equation).

The oxygen isotopic composition is expressed as δ18O per mill (‰). The δ18O of each sample is calculated by reference to the Vienna Standard Mean Ocean Water (VSMOW) as:

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

where Rsample is the oxygen isotope ratio (18O/16O) of a sample cellulose.

2.3 Wood Dating Using Oxygen Isotopes

To test the potential of oxygen isotope dendrochronology for archaeological wood remains from northwestern Alaska, we followed the methodology developed for archaeological oak wood remains in Central England (Loader et al. 2019) and replicated on New Zealand matai trees (Loader et al. 2022). A detailed and explicit presentation of these calculations, including their rationale, can be found in these two publications.

First, the δ18O individual chronologies were filtered to eliminate the low-frequency variability and highlight the interannual variations. We tested nine filters from the five main filter categories (Table A1 in the Appendix) as discussed by Loader et al. (2019). The ends of the δ18O individual chronologies were mirrored to deal with end effects. Finally, we selected a 9-year rectangular filter, which produced the chronology with the lowest first-order autocorrelation (AR1; Table A1 in the Appendix). Data were indexed by subtracting the 9-year filter. In the following, we add the subscript “i” to the name of a series to indicate it is indexed (e.g. δ18O–refi).

Second, the indexed series were cross-dated isotopically and evaluated using various statistical indices to assess the match probability of the dates and avoid spurious correlations. In the Results section, we include the correlation coefficient r, the Student’s t-value, the degrees of freedom (df; adjusted for filtering), the inverse probability (1/P ; the probability that a match of equivalent value occurs at an incorrect position), and the isolation factor (IF; the ratio of the best to the second-best probabilities; Loader et al. 2022). We calculated the degrees of freedom using a corrected sample size Ncor , which is an adjustment that considers any remaining AR1 and the statistical cost of the filter. The Ncor values reported here are related to the series length at the position of the strongest association. We used two strict thresholds as validation of the match probability: 1/P≥100 and IF≥10 (Loader et al. 2019).

We evaluated the dating procedure on the five δ18O–refi series. Each series was blindly cross-dated with a mean chronology obtained by averaging the other four. The minimum overlap between the individual series and mean chronologies was 82 years (Table 2). Then, we conducted the oxygen isotope cross-dating between NWAK18O and the three δ18O–testi series. This cross-dating was performed on both the total period of NWAK18O (935–1157 CE) and the better-replicated section (BRS) of NWAK18O (n>3; 1040–1137 CE). We assessed the match probability for a minimum common overlap period of 50 years (Loader et al., 2019). The results were assessed based on the calculated match probability and statistics, visual year-to-year congruence between individual series and NWAK18O, and consistency of potential dates with wiggle-matching calendar intervals (Table 1).

Results of the cross-dating of the five individual series dated by dendrochronology
Table 2

Results of the cross-dating of the five individual series dated by dendrochronology

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

3 Results and Discussion

3.1 Evaluating the Approach with the Five δ18O–Ref Series

The mean δ18O values of the five δ18O–ref series range from 23.23‰ to 23.99‰. The δ18O–refi correlate well with each other with a mean standard deviation of 0.85 and a mean Pearson coefficient of the pairwise correlations (R̄) of 0.62 over the entire analysis period (935–1157 CE) (Cook et al. 2000: 81). The running expressed population signal (EPS) (Wigley et al. 1984; Speer 2012) was higher than 0.85 for the period 1048–1118 CE where the sample depth for each year was at least 4. The cross-dating of the five tree-ring series corresponding to the five δ18O–ref series show Baillie Pilcher t-values (TVBP) between 2.3 and 6.4 (Table 2; Baillie & Pilcher 1973; Sander & Levanic 1996) and a medium to strong inter-series signal. The cross-correlation of the five δ18O–refi series show t-values between 10.77 and 2.15 (Table 2), thus, presenting a stronger inter-series coherence than the ring-widths.

The blind cross-dating of the five δ18O–refi series (one against the other four) yielded the same dates as conventional dendrochronology (Tables 1 and 3). The correlation coefficients between one δ18O–refi series and the mean of the other four were between 0.39 and 0.85, and the IF between 26.71 and over 1000 (Table 3). These results indicate the medium-to-high strength of cross-dating matches (Table 3). For 12w62-04, the match statistics were low even though the known date of the first year of growth was accurate.

The strong coherence between the δ18O–refi series supports averaging them into an NWAK18O chronology, the δ18O master chronology for northwestern Alaska (Fig. 3). NWAK18O is 223 years long and extends from 935 to 1157 CE. The sample depth ranged from one to five trees. NWAK18O is almost normally distributed, with close mean and median values (0.05 and 0.02), low skewness (–0.04), and low excess Kurtosis (0.37).

Results of blind cross-dating of one δ18O–refi series against the other four
Table 3

Results of blind cross-dating of one δ18O–refi series against the other four

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

Northwestern Alaska δ18O master chronology and individual series. (A) The five individual δ18O–refi series positioned in calendar time. (B) NWAK18O chronology (935–1157 CE) (black line) and sample depth (grey dotted line)
Figure 3

Northwestern Alaska δ18O master chronology and individual series. (A) The five individual δ18O–refi series positioned in calendar time. (B) NWAK18O chronology (935–1157 CE) (black line) and sample depth (grey dotted line)

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

The blind dating of the dendro-dated δ18O series from northwestern Alaska demonstrated that the δ18O–based cross-dating method was very efficient for dating archaeological timbers from coastal sites when they came from trees with the same geographical origin. Each dating match probability exceeded the statistical thresholds of validation indicated by the authors, supporting the hypothesis of a strong δ18O signal, at least, at the regional scale (Table 3). The length of the chronology and the number of trees per year do not seem to be an obstacle for cross-dating contemporary δ18O series from northwestern Alaska trees. This supports the hypothesis of very strong inter-tree isotopic signal coherency at the regional scale.

3.2 Evaluating the Approach with the δ18O–Test Series

The cross-dating of the three floating δ18O–test series with NWAK18O yielded low statistics for 12w128-03 and 12w51-22. The match probabilities did not pass the statistical thresholds of validity, either on the whole chronology (935–1157 CE) or on the BRS chronology (n>3; 1040–1137 CE) (Table 4). The dates obtained for the first ring of the series, respectively 937 and 1074 CE for the entire chronology and 948 and 1065 CE for the BRS chronology, did not fall in their wiggle-matching intervals, 1034–1052 and 1087–1112 CE, respectively (Tables 1 and 4). Visually, some pointer years were common to both the series and NWAK18O, but the overall congruency was low (Fig. 4A, B). Given the low statistics and disagreement with the wiggle-matching time intervals, we did not validate these results.

Results of the cross-dating of NWAK18o chronology and floating δ18O–testi series
Table 4

Results of the cross-dating of NWAK18o chronology and floating δ18O–testi series

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

Comparisons of the whole NWAK18O chronology (grey) and δ18O–testi series (black). (A) 12w128-03, beginning in 937 CE (undated, fails statistics and 14C wiggle-matching disagreement). (B) 12w51-22 beginning in 1074 CE (undated, fails statistics and 14C wiggle-matching disagreement). (C) 12w62-05 beginning in 1073 CE (plausible, low statistics but 14C wiggle-matching agreement)
Figure 4

Comparisons of the whole NWAK18O chronology (grey) and δ18O–testi series (black). (A) 12w128-03, beginning in 937 CE (undated, fails statistics and 14C wiggle-matching disagreement). (B) 12w51-22 beginning in 1074 CE (undated, fails statistics and 14C wiggle-matching disagreement). (C) 12w62-05 beginning in 1073 CE (plausible, low statistics but 14C wiggle-matching agreement)

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

For the 12w62-05 series, the statistics were moderately good, and the match probabilities stronger than those of the other two (Table 4). This match probability with the whole NWAK18O chronology relies on an 85-year overlap, an r = 0.37, and a t = 3.42, which is close to the performance of sample 12w62-04 that was successfully dated using TRW standard dendrochronology and δ18O. The inverse probability was close to 1,000; however, the IF was less than 10 (the validity threshold proposed by Loader et al. 2019). When the δ18O cross-dating was calculated with the BRS NWAK18O chronology, we obtained the same match probability date with statistics slightly stronger for r (0.40 instead of 0.37) and IF (7.65 instead of 4.61) but slightly lower for t (3.18 instead of 3.42) and the inverse probability (404.76 instead of 979.84) (Table 4). The number of trees for each year of the chronology likely played a role in the strength of the correlations obtained. The date of the first year of growth at 1073 CE is consistent with the wiggle-matching interval of 1071–1102 CE (Tables 1 and 4). The first year of growth at 1073 CE for this series provides a date of 1192 CE for the last measured growth ring, which is consistent with the youngest timber date for F-12 house (Alix et al. 2020a,b). Some pointer years are common between the 12w62-05 series and NWAK18O, and there is overall consistency of the two signals, especially between 1073 and 1121 CE and between 1138 and 1157 CE (Fig. 4C). The agreement with the wiggle-matching interval and elements of similarity advocates 1073 CE as the first year of growth of sample 12w62-05. However, this result must be confirmed by further analyses.

Overall, none of the results are convincing for the three δ18O–test series of unknown geographical origins and need to be refined. Only one of the three samples showed an agreement. The reference chronology, NWAK18O, is 223 years old with a sample depth of only 1 to 5 trees per year. As a result, when the sample depth is less than four trees, the chronology may not be representative of the tree population (EPS=0.85). Based on the wiggle-matching interval in which the first measured ring was most likely dated (Table 1), series 12w51-22 (wiggle-matching: 1087–1112 CE) was located in the younger part of NWAK18O where the series is of less quality (sample depth lower than four trees). This may explain why we found no significant matches with NWAK18O. The series 12w128-03 (1034–1052 CE) and 12w62-05 (1071–1102 CE) are contemporaneous with more robust portions of NWAK18O (four to five trees) over a longer period. The poor agreement between 12w128-03 and NWAK18O may indicate that the origin of these trees is far beyond the geographic area in which NWAK18O is representative. This suggests that the samples are from trees that grew in regions influenced by a different mode of climate variability than that in northwestern Alaska.

The validity area of the reference series NWAK18O may not be wide enough (geographically). This is not what we anticipated from the distribution of δ18O in precipitation depicting a gradual southwest-to-northeast 18O progressive depletion (West et al. 2010; Isomap ©Bowen 2003–2022), which should be consistent with high correlations of δ18O chronologies over large distances. Thus, the distribution of δ18O in precipitation may be more heterogenous than expected or the precipitation δ18O signal may be transformed by trees in such a way that the distribution of δ18O in tree-rings may not be as spatially consistent as that of precipitation. This should be further investigated in the future.

4 Conclusion

This preliminary study indicates a strong and regionally consistent δ18O signal at the regional scale between dendro-dated series from northwestern Alaska during the early 2nd millennium CE. The blind dating of each individual series contributing to NWAK18O with the mean of the other four provided the same dating results as the conventional TRW dendrochronology. With medium-to-high cross-dating strength, blind dating supports averaging the individual series into one NWAK18O master chronology. The results of the three cross-dating floating δ18O–test series from unknown geographic origins with NWAK18O are unconvincing for two samples which suggests that they are from regions where the isotopic signal differs from that of northwestern Alaska. The likely date of δ18O–test series 12w62-05 is plausible. However, the low strength of cross-dating advocates the acquisition of new data to extend and strengthen the NWAK18O chronology and possibly increase the strength of cross-dating matches. We used an approach based on multiple proxies using dendro-dated timbers to evaluate the performance of oxygen isotope cross-dating. To obtain secure and independent dating results using oxygen isotope (δ18O) dendrochronology, it is necessary to develop a longer and well-replicated chronology.

A longer and more robust chronology could, then, be used to date other archaeological wood samples, thus contributing to refining the chronological and climatic framework of the 2nd millennium CE in Alaska that saw the emergence of the Inuit culture. Oxygen isotope cross-dating, with a longer reference chronology, may help overcome the absence of sufficiently long reference ring-width chronologies for dating Birnirk and Thule structural wood elements originating from various regions, where 18O progressive depletion shows a gradual southwest-to-northeast distribution.

Stable isotope dendrochronology is a developing tool complementary to conventional dendrochronology that can help date additional archaeological wood in the future. It is an expensive and time-consuming method that can only be used as a complement to conventional dendrochronology, not as a replacement tool. At this stage, it is important to strengthen and lengthen the NWAK18O, establish a gradient of δ18O isotopic signal in western Alaska and record its spatio-temporal variability. This will help understand the factors contributing to this variability. In the future, further development of NWAK18O and its contribution to the annual dating of Birnirk and Thule architectural timbers from coastal sites may contribute to a broader geographic identification of multiple driftwood source regions.

Acknowledgements

The Cape Espenberg tree-ring samples were collected as part of the Cape Espenberg Birnirk Project (CEBP), PIs: C. Alix and O.K. Mason. Our deepest thanks go to the communities of Shishmaref, Deering, and Kotzebue for their participation in and contribution to the consultation process during the preparation stages of the project. We especially thank the families with ancestral links and ties to Cape Espenberg for supporting our archaeological efforts and for their collaboration. The Cape Espenberg archaeology project (CEBP Project) was funded by a collaborative grant (ARC-1523160, ARC-1523205 and ARC-1523059) from the Office of Polar Programs Arctic Social Sciences at the National Science Foundation (NSF) and received funding from the Archaeology Commission of the French Ministry of Foreign Affairs. This research was part of J. Taïeb doctoral research which was supported by a 3-year PhD fellowship from the University of Paris 1 Pantheon-Sorbonne and two grants from the World Wood Day Foundation and International Wood Culture Society for the project “Let the Wood Speak: Dendro-Archaeology, Climate and Culture in Northwestern Alaska at the Beginning of the Second Millennium AD”. Radiocarbon dating for 14C wiggle-matching analyses was funded by CEA-Saclay and conducted at the Laboratoire de Mesure du Carbone 14 (CEA-Saclay, Gif-sur-Yvettes, France). We are grateful to Glenn P. Juday, Emeritus Professor at the University of Alaska Fairbanks (UAF), for his advice on conventional tree-ring analyses, as well as Nancy Bigelow at the Alaska Quaternary Center at UAF and Owen K. Mason (INSTAAR, University of Colorado-Boulder and co-lead PI of the CEBP Project) for their continued support. We also thank Ryan Jess and Mike Lorrain at UAF who prepared and measured parts of the Cape Espenberg samples. Finally, we would like to thank Monique Pierre, Michel Stievenard, and Tiphaine Penchenat from the dendro team of the Laboratoire des Sciences du Climat et de l’Environnement (LSCE) for their valuable help with the various geochemical analyses. The CEBP Project was conducted in accordance with the Memorandum Of Agreement between the National Park Service, the NSF, and the Alaska State Historic Preservation Officer regarding Data Recovery and Intentional Excavation of sites KTZ-304, KTZ-094 and KTZ-157 at Cape Espenberg Bering Land Bridge National Preserve, and the NAGPRA Plan Of Action for Intentional Excavation of Human Skeletal Remains from House-Pit Features at KTZ-304 and Surface Remains from KTZ-094 and KTZ-157 at Cape Espenberg Bering Land Bridge National Preserve (Agreement). It complies with the National Park Service research requirements and receives Research Permit and Reporting System Permits #BELA2–16_SCI_0002, BELA2017_SCI_0003, and BELA2018_SCI_0005 for the project’s three field seasons.

References

  • Alix C. 2004. Bois flotté et archéologie de l’Arctique : contribution à la préhistoire récente du détroit de Béring. Études Inuit/Inuit Studies 28 (1): 109–132. DOI: 10.7202/012642ar.

    • Search Google Scholar
    • Export Citation

  • Alix C. 2005. Deciphering the impact of change on the driftwood cycle: contribution to the study of human use of wood in the Arctic. Glob. Planet Change 47: 83–98. DOI: 10.1016/j.gloplacha.2004.10.004.

    • Search Google Scholar
    • Export Citation

  • Alix C. 2009. Persistence and change in Thule wood use. In: Mason OK, McGheen R (eds.), The Northern World, AD 900–1400: The Dynamics of Climate, Economy, and Politics in Hemispheric Perspective: 179–205. University of Utah Press, Salt Lake City, UT.

    • Search Google Scholar
    • Export Citation

  • Alix C. 2012. Introduction: Arctic peoples and wood. Études Inuit/Inuit Studies 36 (1): 5–22. DOI: 10.7202/1015704ar.

  • Alix C. 2016. A critical resource: wood use and technology in the North American Arctic. In: Mason OK, Friesen TM (eds.), The Oxford handbook of the prehistoric Arctic: 109–129. Oxford University Press, New York, NY. DOI: 10.1093/oxfordhb/9780199766956.013.12.

    • Search Google Scholar
    • Export Citation

  • Alix C, Brewster K. 2004. Not all driftwood is created equal: wood use and value along the Yukon and Kuskokwim Rivers, Alaska. Alaska J. Anthropol. 2 (1): 2–19.

    • Search Google Scholar
    • Export Citation

  • Alix C, Koester D. 2001. Driftwood in the North Pacific. Report of a driftwood sampling project undertaken on the Harriman Expedition retraced. University of Alaska, Fairbanks, AK.

    • Search Google Scholar
    • Export Citation

  • Alix C, Mason OK, Bigelow NH, Anderson SL, Rasic J, Hoffecker JF. 2015. Archéologie du Cap Espenberg où la question du Birnirk et de l’origine du Thulé dans le nord-ouest de l’Alaska. Nouv. Archéol. 141: 13–19. DOI: 10.4000/nda.3065.

    • Search Google Scholar
    • Export Citation

  • Alix C, Mason OK, Norman LEY, Bigelow NH, Maio C. 2017. Birnirk prehistory and the emergence of Inupiaq culture in Northwestern Alaska, Archaeological and anthropological perspectives. Field investigation at Cape Espenberg, 2016. Annual Report to the National Park Service, US Department of the Interior, Fairbanks, AK.

    • Search Google Scholar
    • Export Citation

  • Alix C, Mason OK, Norman LEY. 2018. Whales, wood and baleen in Northwestern Alaska — reflection on whaling through wood and boat technology at the rising whale site. In: Kim HY (ed.), Korean Prehistoric Art. Whale on the Rock II: 11–68. Ulsan Petroglyph Museum, Ulsan.

    • Search Google Scholar
    • Export Citation

  • Alix C, Taïeb J, Mason OK, Juday GP. 2020a. Tree-ring research at Cape Espenberg, New insights on time and architecture. Paper presented at the 47th Annual Meeting of the Alaska Anthropological Association, Fairbanks, AK.

  • Alix C, Mason OK, Anderson S, Bigelow NH, Krus A, et al. 2020b. Projet Cap Espenberg, Aux origines de la culture Inupiaq dans le nord-ouest de l’Alaska, Bilan du programme quadriennal 2016–2019 et proposition de renouvellement 2021–2024. Report for the Ministry of Europe and Foreign Affairs, Paris.

    • Search Google Scholar
    • Export Citation

  • Baillie MGL, Pilcher JR. 1973. A simple cross-dating program for tree-ring research. Tree-Ring Bull. 33: 7–14.

  • Baker JCA, Hunt SFP, Clerici SJ, Newton RJ, Bottrell SH, Leng MJ, Heaton THE, et al. 2015. Oxygen isotopes in tree rings show good coherence between species and sites in Bolivia. Glob. Planet. Change 133: 298–308. DOI: 10.1016/j.gloplacha.2015.09.008.

    • Search Google Scholar
    • Export Citation

  • Barber VA. 2003. Past Climates of Northern Alaska from Tree Rings. In: Mason OK (ed.), Uivvaq Heritage Project — Field season 2002 Final report: 59–68. GeoArch Alaska, Anchorage, AK.

    • Search Google Scholar
    • Export Citation

  • Boettger T, Haupt M, Knöller K, Weise S, Waterhouse J, et al. 2007. Wood cellulose preparation methods and mass spectrometric analyses of δ 13 C, δ 18 O, and nonexchangeable δ 2 H values in cellulose, sugar, and starch: an interlaboratory comparison. Anal. Chem. 79: 4603–4612. DOI: 10.1021/ac0700023.

    • Search Google Scholar
    • Export Citation

  • Brooks JR, Rugh WD, Werner RA. 2022. Tree-Ring stable isotope measurements: the role of quality assurance and quality control to ensure high quality data. In: Siegwolf RTW, Brooks JR, Roden J, Saurer M (eds.), Stable isotopes in tree rings: inferring physiological, climatic and environmental responses: 191–213. Springer International, Cham. DOI: 10.1007/978-973-030-92698-4_6.

    • Search Google Scholar
    • Export Citation

  • Cook ER, Kairiukstis LA (eds). 1990. Methods of Dendrochronology: Applications in the Environmental Sciences. Kluwer Academic. Dordrecht. DOI: 10.1007/978-994-015-7879-0.

    • Search Google Scholar
    • Export Citation

  • Cook ER, Buckley BM, D’Arrigo RD, Peterson MJ. 2000. Warm-season temperatures since 1600 BC reconstructed from Tasmanian tree rings and their relationship to large-scale sea surface temperature anomalies. Climate Dynam. 16 (2): 79–91. DOI: 10.1007/s003820050006.

    • Search Google Scholar
    • Export Citation

  • D’Arrigo RD, Mashig E, Frank D, Wilson R, Jacoby GC. 2005. Temperature variability over the past millennium inferred from Northwestern Alaska tree rings. Climate Dynam. 24: 227–236. DOI: 10.1007/s00382-00004-0502-0501.

    • Search Google Scholar
    • Export Citation

  • D’Arrigo RD, Wilson R, Jacoby GC. 2006. On the long-term context for twentieth Century warming. J. Geophys. Res. 111: 1–12.

  • Degroot D, Anchukaitis KJ, Bauch M, Burnham J, Carnegy F, et al. 2021. Towards a Rigorous understanding of societal responses to climate change. Nature 591 (7851): 539–550. DOI: 10.1038/s41586-41021-03190-03192.

    • Search Google Scholar
    • Export Citation

  • Degroot D, Anchukaitis KJ, Tierney JE, Riede F, Manica A, Moesswilde E, Gauthier N. 2022. The history of climate and society: a review of the influence of climate change on the human past. Environ. Res. Lett. 17(10): 103001. DOI: 10.1088/1748-9326/ac8faa.

    • Search Google Scholar
    • Export Citation

  • Douglass AE. 1921. Dating our prehistoric ruins. J. Am. Mus. Nat. Hist. 21(1): 27–30.

  • Eggertsson O. 1994. Mackenzie River driftwood: a dendrochronological study. Arctic 47(2): 128–136.

  • Eggertsson O. 1995. Origin of the Arctic driftwood — a dendrochronological study. Oceanogr. Lit. Rev. 7 (42): 543–544.

  • Eurola S. 1971. The driftwoods of the Arctic Ocean. Rep. Kevo Subarct. Res. Stat. 7: 74–80.

  • Giddings JL. 1943. A plan for mapping Arctic Sea Currents. Geogr. Rev. 33: 326–327.

  • Giddings JL. 1948. Chronology of the Kobuk-Kotzebue sites. Tree-Ring Bull. 14(4): 26–32.

  • Giddings JL. 1952a. Driftwood and problems of the Arctic Sea currents. Trans. Am. Philos. Soc. 96: 129–142.

  • Giddings JL. 1952b. The Arctic Woodland culture and the Kobuk River. University Museum Monograph, University of Pennsylvania, Philadephia, PA.

    • Search Google Scholar
    • Export Citation

  • Graumlich L, King J. 1997. Late Holocene climatic variation in Northwestern Alaska as reconstructed from tree-ring. Final Unpublished Report to the Cooperative Research with the National Park Services. University of Arizona, Tucson, AZ.

    • Search Google Scholar
    • Export Citation

  • Griggs C, Kocik C, Urban T, Manning S. 2019. Dendrochronology of Swift Water Place and other tree-ring samples from North West Alaska. In: Anderson DD, Anderson WW (eds.), Life at Swift Water Place: Northwest Alaska at the threshold of European contact: 129–153. University of Alaska, Fairbanks, AK.

    • Search Google Scholar
    • Export Citation

  • Hellmann L, Agafonov L, Churakova (Sidorova) O, Düthorn E, Eggertsson Ó, et al. 2016. Regional coherency of boreal forest growth defines Arctic driftwood provenancing. Dendrochronologia 39: 3–9. DOI: 10.1016/j.dendro.2015.12.010.

    • Search Google Scholar
    • Export Citation

  • Hellmann L, Tegel W, Geyer J, Kirdyanov A, Nikolaev AN, et al. 2017. Dendro-provenancing of Arctic driftwood. Quater. Sci. Rev. 162: 1–11. DOI: 10.1016/j.quascirev.2017.02.025.

    • Search Google Scholar
    • Export Citation

  • Holmes RL. 1983. Computer-assisted quality control in tree-ring dating and measurement. Tree-Ring Bull. 43: 69–78.

  • Hopkins DM, Giddings JL. 1953. Geological background of the Iyatayet archeological site, Cape Denbigh, Alaska. Smithsonian Miscell. Coll. 121(11): 1–33. Available online at http://repository.si.edu/xmlui/handle/10088/23767.

    • Search Google Scholar
    • Export Citation

  • Krus A, Jensen A, Hamilton W, Sayle K. 2019. A context-appropriate approach to marine 14C calibration: ΔR and bayesian framework for the Nuvuk cemetery, Point Barrow, Alaska. Radiocarbon 61(3): 733–747. DOI: 10.1017/RDC.2019.20.

    • Search Google Scholar
    • Export Citation

  • Leavitt SW, Danzer SR. 1993. Method for batch processing small wood samples to holocellulose for stable-carbon isotope analysis. Anal. Chem. 65 (1): 87–89. DOI: 10.1021/ac00049a017.

    • Search Google Scholar
    • Export Citation

  • Loader NJ, McCarroll D, Miles D, Young GHF, Davies D, Bronk Ramsey C. 2019. Tree ring dating using oxygen isotopes: a master chronology for Central England. J. Quater. Sci. 34 (6): 475–490. DOI: 10.1002/jqs.3115.

    • Search Google Scholar
    • Export Citation

  • Loader NJ, McCarroll D, Miles D, Young GHF, Davies D, Bronk Ramsey C, Williams M, Fudge M. 2021. Dating of non-oak species in the United Kingdom historical buildings archive using stable oxygen isotopes. Dendrochronologia 69: 125862. DOI: 10.1016/j.dendro.2021.125862.

    • Search Google Scholar
    • Export Citation

  • Loader NJ, Boswijk G, Young GHF, Hogg AG, McCarroll D. 2022. Developing tree ring chronologies from New Zealand Matai (Prumnopitys taxifolia (D.Don) Laub.) for archaeological dating: stable isotope dendrochronology. Dendrochronologia 76: 126030. DOI: 10.1016/j.dendro.2022.126030.

    • Search Google Scholar
    • Export Citation

  • Mason OK. 2009. Flight from the Bering Strait: Did Siberian Punuk/Thule military cadres conquer northwest Alaska? In: Mason, OK McGheen R (eds.), The Northern World, AD 900–1400: the dynamics of climate, economy, and politics in hemispheric perspective: 76–128. University of Utah Press, Salt Lake City, UT.

    • Search Google Scholar
    • Export Citation

  • Mason OK. 2016. Thule origins in the old Bering Sea culture: the interrelationship of Punuk and Birnirk cultures. In: Mason OK, Friesen TM, The Oxford handbook of the prehistoric Arctic: 489–513. Oxford University Press, New York, NY.

    • Search Google Scholar
    • Export Citation

  • Mason OK. 2020. The Thule migrations as an analog for the early peopling of the Americas: evaluating scenarios of overkill, trade, climate forcing, and scalar stress. PaleoAmerica 6(4): 308–356. DOI: 10.1080/20555563.2020.1783969.

    • Search Google Scholar
    • Export Citation

  • Mason OK, Bowers PM. 2009. The origin of the Thule culture is always elsewhere: early Thule within Kotzebue Sound, “cult-de-sac” or nursery? In: Grønnow B (ed.), On track of the Thule culture from Bering Strait to East Greenland: 25–44. Collected Publications from the National Museum, Studies in Archaeology and History, Copenhagen.

    • Search Google Scholar
    • Export Citation

  • Mason OK, Gerlach SC. 1995. Chukchi hot spots, paleo‐polynyas, and caribou crashes: climatic and ecological dimensions of North Alaska prehistory. Arctic Anthropol. 32 (1): 101–130.

    • Search Google Scholar
    • Export Citation

  • Morrison D. 1989. Dating Thule culture. Arctic Anthropol. 26(2): 48‑77.

  • Morrison D. 2001. Radiocarbon dating the Birnirk-Thule Transition. Anthropol. Pap. Univ. Alaska NS 1(1): 73–85.

  • Nash S. 2000. James Louis Giddings’ archaeological tree-ring dating in the American Arctic: a forgotten legacy. Arctic Anthropol. 37: 60–78.

    • Search Google Scholar
    • Export Citation

  • Nicolle M, Debret M, Massei N, Colin C, deVernal A, Divine D, Werner JP, Hormes A, Korhola A, Linderholm HW. 2018. Climate variability in the Subarctic area for the last 2 millennia. Climate Past. 14 (1): 101–116. DOI: 10.5194/cp-14-101-2018.

    • Search Google Scholar
    • Export Citation

  • Oswalt WH. 1951. The origin of driftwood in Hooper Bay, Alaska. Tree-Ring Bull. 18: 6–8.

  • Porter TJ, Pisaric MFJ, Kokelj SV, deMontigny P. 2013. A ring-width-based reconstruction of June-July minimum temperatures since AD 1245 from white spruce stands in the Mackenzie Delta region, northwestern Canada. Quarter. Res. 80(2): 167–179. DOI: 10.1016/j.yqres.2013.05.004.

    • Search Google Scholar
    • Export Citation

  • Raghavan M., Degiorgio M, Albrechtsen A, Moltke I, Skoglund P et al. 2014. The genetic prehistory of the New World Arctic. Science 345(6200): 1–8.

    • Search Google Scholar
    • Export Citation

  • RinnTech. 2000. TSAP-WIN User Reference manual. RinnTech, Heidelberg.

  • Roden J. 2008. Cross-dating of tree ring δ18O and δ13C time series. Chem. Geol. 252(1): 72–79. DOI: 10.1016/j.chemgeo.2008.01.007.

    • Search Google Scholar
    • Export Citation

  • Sander C, Levanic T. 1996. Comparison of T-values calculated in different dendrochronological programmes. Dendrochronologia 14: 269–272.

    • Search Google Scholar
    • Export Citation

  • Speer JH. 2012. Fundamentals of tree-ring research. University of Arizona Press, Tucson, AZ.

  • Taïeb J, Alix C, Juday GP, Jensen AM, Daux V, Petit C. 2022. Dating coastal archaeological wood from Pingusugruk (15th–17th CE), Northern Alaska: preliminary results and methodological perspectives in dendrochronology. Int. J. Wood Cult. 2(1–3): 89–116. DOI: 10.1163/27723194-bja10006.

    • Search Google Scholar
    • Export Citation

  • Taïeb J, Daux V, Alix C, Hatté C. 2023. Contributions of 14C wiggle-matching for dendroarchaeological studies of coastal Birnirk and Thule sites in northern Alaska. In: Bacoup P, Taïeb, J. (eds.), Actes des rencontres internationales « Journées bois. Échanges interdisciplinaires sur le bois et les sociétés » organisées les 18–19 octobre 2021 à l’INHA, Paris, Archéologie, société et environnement 3(1): 55–77. DOI: 10.21494/ISTE.OP.2023.1025

  • Unkel SL, Norman LEY, Tackney J, Krus AM, Jensen AM, Alix C, Mason OK, O’Rourke DH. 2022. Genetic Analysis of Birnirk Inuit from the Alaskan North Slope. Arctic 75(1): 121–132. DOI: 10.14430/arctic74916.

    • Search Google Scholar
    • Export Citation

  • VanStone JW. 1958. The origin of driftwood on Nunivak Island, Alaska. Tree-Ring Bull. 22: 12–15.

  • Viereck LA, Little EL. 2007. Alaska trees and shrubs. University of Alaska Press, Fairbanks, AK.

  • West J, Bowen GJ, Dawson TE, Tu K. 2010. Isoscapes: understanding movement, pattern, and process on earth through isotope mapping. Springer, New York, NY. DOI: 10.1007/978-990-481-3354-3.

    • Search Google Scholar
    • Export Citation

  • Wigley TML, Briffa KR, Jones PD. 1984. On the average value of correlated time series, with applications in dendroclimatology and hydrometeorology. J. Climate Appl. Meteorol. 23: 201–213.

    • Search Google Scholar
    • Export Citation

  • Xu J, Pan B, Liu T, Hajdas I, Zhao B, Yu H, Liu R, Zhao P. 2013. Climatic impact of the millennium eruption of Changbaishan volcano in China: new insights from high-precision radiocarbon wiggle-match dating. Geophys. Res. Lett. 40(1): 54–59. DOI: 10.1029/2012GL054246.

    • Search Google Scholar
    • Export Citation

  • Yamada R, Kariya Y, Kimura T, Sano M, Li Z, Nakatsuka T. 2018. Age determination on a catastrophic rock avalanche using tree-ring oxygen isotope ratios — the scar of a historical gigantic earthquake in the Southern Alps, Central Japan. Quarter. Geochronol. 44: 47–54. DOI: 10.1016/j.quageo.2017.12.004.

    • Search Google Scholar
    • Export Citation

Appendix

Left: Three of the eight Picea glauca cross-sections presented here are from the architectural elements of Birnirk house F-12 (KTZ-304). Right: North room of F-12 at the end of the 2017 excavation (C. Alix)
Figure A1

Left: Three of the eight Picea glauca cross-sections presented here are from the architectural elements of Birnirk house F-12 (KTZ-304). Right: North room of F-12 at the end of the 2017 excavation (C. Alix)

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029

Categories of tested filters and first-order autocorrelation AR1 of the NWAK18o chronology according to the different filters tested
Table A1

Categories of tested filters and first-order autocorrelation AR1 of the NWAK18o chronology according to the different filters tested

Citation: International Journal of Wood Culture 4, 1 (2024) ; 10.1163/27723194-bja10029
Title:
Dendroarchaeology of Birnirk and Thule Architectural Timbers (10–13th Centuries)
Article Type:
Research Article
DOI:
https://doi.org/10.1163/27723194-bja10029
Language:
English
Pages:
17–38
Keywords:
archaeology; coastal driftwood; dendrochronology; stable isotope; Thule; Western Arctic
In:
International Journal of Wood Culture
In:
Volume 4: Issue 1
Received:
16 Mar 2023
Accepted:
05 Feb 2024
Publisher:
Brill
E-ISSN:
2772-3194
Print ISSN:
2772-3186
Subjects:
General, Biology, General, Art History, Materials Sciences

Content Metrics

All Time Past 365 days Past 30 Days
Abstract Views 0 0 0
Full Text Views 932 752 33
PDF Views & Downloads 500 256 8