Preliminary strontium isotope (⁸⁷Sr/⁸⁶Sr) baselines for the Bjäre Peninsula and Halland in southern Sweden

Abstract

During the last decade, the application of strontium isotope analysis (⁸⁷Sr/⁸⁶Sr) has increased rapidly. The understanding of the strontium isoscape and the construction of a strontium isotope baseline in southern Scandinavia are biased toward Denmark and southwestern to eastern Scania. We report the results of new baseline samples in Halland and the Bjäre Peninsula, adding to the rich strontium isotope library that exists for southern Scandinavia. We add nuance to the previous divisions of Scania, in which the Bjäre Peninsula is not well represented. The results show that the Bjäre Peninsula is characterized by relatively low ⁸⁷Sr/⁸⁶Sr values gathered in the preliminary baseline 0.7100 ± 0.0024 (2σ, n = 8), similar to that of southwest Scania and Zealand in Denmark. There is a greater variation and higher values among the Halland samples. Together with previously published faunal samples, a preliminary ⁸⁷Sr/⁸⁶Sr baseline for Halland is 0.7122 ± 0.0055 (2σ, n = 24). We apply these baseline results to archaeological cases, both human and animal, from recent excavations in the Bjäre Peninsula. The results imply that there was a certain degree of mobility and interaction across the landscape in this region of southern Scandinavia in prehistory.

1 Introduction

The archaeological application of strontium isotope (⁸⁷Sr/⁸⁶Sr) analysis has majorly impacted the study of movement, mobility, interaction, and communication in prehistory (Slovak and Paytan, 2012; see Kristiansen, 2022). The basis of the archaeological understanding of mobility is now transferred from being object-based to being based on the research subjects themselves, that is, the remains of the humans and animals inhabiting the past. The uptake of ⁸⁷Sr/⁸⁶Sr into organic tissues (bone, teeth, etc.) is directly related to the local bioavailable strontium ratios. However, for ⁸⁷Sr/⁸⁶Sr analysis to provide meaningful information on migration, local ⁸⁷Sr/⁸⁶Sr “baselines,” that is, reference values, are needed for any geographical area. The understanding of the ⁸⁷Sr/⁸⁶Sr ratio variation in southern Scandinavia has significantly increased since the surge in strontium-related studies in the last few decades. Published baseline studies exist for Scania (e.g., Ladegaard-Pedersen et al., 2021; Boethius et al., 2022), western Sweden (e.g., Blank et al., 2018; Klassen et al., 2020), and Denmark (e.g., Frei et al., 2019). The ⁸⁷Sr/⁸⁶Sr strontium isotope mapping is cumulative, meaning that adding data increases the knowledge of local strontium isotopic variations and bioavailable strontium (e.g., Holt et al., 2021). The majority of ⁸⁷Sr/⁸⁶Sr in Scania is based primarily on samples from the southern and eastern parts. The northwest Scania, including the Bjäre Peninsula, is less sampled and thus less understood.

This article aims to fill this gap in the strontium isoscape map by presenting new baseline data from the Bjäre Peninsula and Halland. We contextualize the data by relating it to the baseline frames that exist for Scania and Halland. Like Ladegaard-Pedersen et al. (2021), we base the discussion on descriptive statistics rather than modeling approaches. We also incorporate earlier published ⁸⁷Sr/⁸⁶Sr reference values from Halland in our analysis (Klassen et al., 2020), presented in Supplementary Table S1. Finally, the aim is to apply this improved baseline archaeologically, with strontium isotope analyses on human and animal remains from recent excavations in Bjäre.

The Bjäre Peninsula constitutes the northwestern outpost of Scania, bordering the region of Halland to the north (Figure 1). Bjäre and parts of Halland lie on the Sveconorwegian orogen. Bjäre is covered by a granitoid to syenotoid migmatitic gneiss (1.7 Ga). This is also the case for Halland, with the addition of granite (1.7–1.0 Ga) and small spots of mainly amphibolite (1.7–0.9 Ga). The northernmost part of Halland consists of granitoid (1.6–1.5 Ga) and metamorphic bedrock, which extends northward above Lake Vänern. We also see areas of granite, syenitoid, and metamorphic equivalents (approximately 1.36–1.27 Ga) and gabbro, diorite, ultrabasic rock, dolerite (1.6–1.3 Ga), and metamorphic rocks. The intersection between Bjäre and Halland, that is, the bay interconnecting them, is Cretaceous sedimentary cover rock consisting of limestone, sandstone, and clay.

Figure 1

2020:414; Supplementary Table S1). Geological maps from “Jordarter 1:25,000-1:100,000” and “Berggrund 1:1 miljon” © Sveriges geologiska undersökning. Lower right: Map of southern Scandinavia, with highlighted areas of interest, including the strontium isotope areas (A1-4) suggested by Ladegaard-Pedersen et al. (2021). Maps made with ArcGIS vs 3.0.0 © ESRI.

The similarity in bedrock across Halland and Bjäre and inland northern Scania indicates that most of these areas, disregarding any sea spray or marine effects, should hypothetically contain similar ⁸⁷Sr/⁸⁶Sr ratios. However, soil variation is often a more important consideration of bioavailable strontium (Wilhelmson, 2017, p. 199). The soil types are more heterogeneous in Bjäre, dominated by till with spots of exposed bedrock across the peninsula, while along the south coast and the “tip” of the peninsula are postglacial sand/gravel deposits (Figure 1). Halland, by comparison, is dominated by postglacial sand, with spots of fluvioglacial deposits, exposed bedrock, and clay/silt.

2 Material and methods

Leaves from two different plants from seven locations in Halland and three in Bjäre (Table 1) were sampled for the ⁸⁷Sr/⁸⁶Sr analysis. Three leaves were each collected from Lokal 18, Lokal 21, and Ängelbäcksstrand on Bjäre (Figure 1). We chose both coastal and inland locations to capture possible sea-spray effects or any marine variation (Figure 1). The preliminary strontium baseline is established on descriptive statistics of the isotopic results, as well as earlier published baseline ⁸⁷Sr/⁸⁶Sr values from Klassen et al. (2020). We have excluded samples from Klassen et al. (2020, p. 414), which we consider possibly unrepresentative of the local signal, such as modern minks, which probably were feralized from mink industries with an unknown food supply. Furthermore, we do not include humans and domestic animals, for example, pigs, sheep, and goats, that could hypothetically have a high degree of mobility (e.g., Meiri et al., 2017). This is a more conservative approach than the contextual one put forward by Klassen et al. (2020); however, we believe it is more robust in establishing the local ⁸⁷Sr/⁸⁶Sr baseline.

2.1 Strontium isotope analysis

The samples were processed in the isotope geochemistry clean laboratory at the Department of Earth, Marine and Environmental Sciences, University of North Carolina at Chapel Hill. Plant material (including leaves/twigs and grass) was dried in an oven at approximately 50°C overnight and then ashed in a Thermo^® box (muffle) furnace at 550°C for 4 h. Bone and enamel were abraded with a Dremel, whereupon the best-preserved section was selected and lightly crushed. Approximately 8–10 mg of the sample was dissolved in distilled 3.5 M HNO_3. Strontium was isolated using ion-exchange column chromatography with EiChrom Sr-Spec™ resin using 3.5 M HNO₃ to wash the columns and quadruple distilled water to elute the strontium. H₃PO₄ was added to the sample and evaporated to dryness. The Sr samples were loaded on Re filaments with TaF₅ and analyzed in triple-dynamic multi-collector mode with ⁸⁸Sr = 3 V (10¹¹ Ω resistor) on either a VG Sector-54 thermal ionization mass spectrometer (TIMS) or IsotopX Phoenix X62 TIMS. All data are normalized to ⁸⁶Sr/⁸⁸Sr = 0.1194, assuming exponential fractionation behavior. Data are reported relative to a value for SRM NBS-987 of ⁸⁷Sr/⁸⁶Sr = 0.710250, with long-term external error of 0.000020 (2σ).

3 Results and analysis

The ⁸⁷Sr/⁸⁶Sr results from the plants (Table 1) are summarized for each area as the sample mean (x) ± 2σ (Wilhelmson, 2017, p. 198), generating a preliminary single-proxy baseline for Bjäre of 0.7100 ± 0.0024 (2σ, n = 8) and Halland 0.7124 ± 0.0058 (2σ, n = 15). The ⁸⁷Sr/⁸⁶Sr ratios of plants show that Bjäre provided lower values than Halland. The ⁸⁷Sr/⁸⁶Sr variation is narrower in Bjäre (0.7082–0.7115) compared to Halland (0.7087–0.7177), which could be due to relatively fewer samples or a more homogeneous makeup of bioavailable strontium. The lower ⁸⁷Sr/⁸⁶Sr values in Bjäre and Halland, compared to, for example, northeast Scania, probably indicate a certain degree of sea-spray effect, as these regions are closer to seawater ⁸⁷Sr/⁸⁶Sr values of approximately 0.70918 (McArthur et al., 2001). Klassen et al. (2020) found that ⁸⁷Sr/⁸⁶Sr values from faunal samples were least radiogenic along the southern Halland coast (minimum 0.7087) and most radiogenic inland (maximum 0.7189). The lowest ⁸⁷Sr/⁸⁶Sr values in that study were from three rodent crania (0.7087, 0.7098, and 0.7099), collected from coastal sites in Halland, leading Klassen et al. (2020) to infer that sea spray must be affecting strontium ratios in western Sweden in general. In our plant data, we do not fully observe the sea-spray effect among the samples from the coastal settlements, especially Ängelbäcksstrand (Wil 20–22) and Skummeslöv (Wil 11–12). The latter is located on younger sedimentary rocks than in most of Bjäre and Halland (Figure 1), so this is somewhat surprising. The effect of sea spray on bioavailable strontium seems to vary locally, especially as geological soil formation is complex, due to the idiosyncratic ways the ice retreated during the Late Glacial (see Klassen et al., 2020, p. 416). At Skummeslöv, located on limestone (Figure 1), the soil type is similar to most of the gneissic and granitic Halland coast (postglacial sand/gravel). This may have an averaging effect on the bioavailable strontium in the whole area. The till from which the Bjäre samples were taken is similar to most of inland western Sweden in this area, so this may be one explanation for why the values are not as low as expected.

To contextualize our modern plant ⁸⁷Sr/⁸⁶Sr values from Halland, we turn to previous baseline samples as reported by Klassen et al. (2020), particularly those deriving from bone or teeth of smaller mammals (Supplementary Table S1). These values show a slightly lower ⁸⁷Sr/⁸⁶Sr baseline of 0.7111 ± 0.0041 (2σ, n = 9) than the plants. Integrating Klassen et al.'s faunal samples from Halland with our plants generates a multiproxy baseline for Halland of 0.7122 ± 0.0055 (2σ, n = 24). Klassen et al. (2020) provided one value from a hare, 0.7101, from Hallands Väderö, an islet outside the tip of Bjäre. This fits into the Bjäre baseline range proposed earlier (0.7100 ± 0.0024).

Like Bjäre, most of northern and middle Scania, down to the Ringsjö lakes and a few kilometers east of the city Hässleholm, is located on a granitic to syenotoid migmatitic gneiss. This area of Scania constitutes the largest part of “Area 1” as delineated by Ladegaard-Pedersen et al. (2021) in their division of Scania in terms of bioavailable strontium through a multiproxy approach, using plants, water, and soil leachates (see Figure 1). The strontium baseline for this area is 0.7184 ± 0.0061 (2σ, n = 16; Ladegaard-Pedersen et al., 2021). For the east of Scania, “Area 2,” covering the southern part of the previously mentioned gneiss bedrock to the eastern Scanian coastline, the calculated baseline was 0.7140 ± 0.0043 (2σ, n = 48; Ladegaard-Pedersen et al., 2021). Using these baselines, both areas are described with more radiogenic ⁸⁷Sr/⁸⁶Sr values than those we observe in Bjäre and Halland. The Bjäre ⁸⁷Sr/⁸⁶Sr values are most similar to southwest Scania, Ladegaard-Pedersen et al. (2021)'s “Area 3 + 4,” with an ⁸⁷Sr/⁸⁶Sr baseline of 0.7110 ± 0.0030 (2σ, n = 39). This is also comparable with many Danish baselines (Frei et al., 2019; Klassen et al., 2020), excluding the Bornholm region.

4 Archaeological applications

4.1 The Bjäre excavations in 2021

In advance of the construction of new water pipes along the southern coast of Bjäre, contract archaeology investigations were conducted following standard Swedish heritage legislation. Several localities were investigated as part of this project (Svensson and Liahaugen, forthcom). Archaeologically, the Bjäre Peninsula is understudied, so this project resulted in new information concerning almost all archaeological periods on a large diversity of site types. The preservation of organic material is generally poor in this part of Scania, and in addition, the osteological material, most suitable for chemical analysis, was often heavily burned and fragmented. The features with preserved bone material predominantly consisted of pits from the pre-Roman Iron Age and Late Iron Age–medieval periods. Materials from three investigated sites were the subjects of the current analysis.

4.2 Sampled specimens

Seven bones and teeth from humans and other animals were sampled for ⁸⁷Sr/⁸⁶Sr. Three samples are from three separate humans, retrieved from cremation pits of different kinds: B1 and B7, both skull fragments, derived from contexts dated to the 10^th century ce, and B2, a human long bone fragment, recovered from a cremation pit dated to Late Neolithic/Early Bronze Age. The other four samples were derived from cattle or sheep/goats. One adult cattle tooth (P3–) and one burned bone from a sheep/goat were recovered from the same Viking Age pit as one of the human samples (B2). One sheep/goat burned long bone and one adult cattle tooth (M1–) were recovered from pits dated to the pre-Roman Iron Age. The cattle sample was retrieved from a cooking pit. The samples, archaeological information, and ⁸⁷Sr/⁸⁶Sr results are presented in Table 2. The laboratory procedures for the ⁸⁷Sr/⁸⁶Sr analysis follow the outlined methodologies in the Material and Methods section. Four samples were derived from burned bone. Bone remodels through life, meaning that the ⁸⁷Sr/⁸⁶Sr is indicative of the lived environment on average (e.g., Wolska, 2020). Two samples are from teeth, both from cattle, from which slaughter ages have also been estimated.

4.3 The human remains

Three samples from human remains yielded ⁸⁷Sr/⁸⁶Sr ratios. B1, representing an individual living during the Late Neolithic/Early Bronze Age, yielded the ratio 0.7146. B2, representing a human living in the 10^th century ce, produced the ratio 0.7137. Both can be fitted within all baseline values from Halland (0.7087–0.7177; Table 1, Supplementary Table S1). Using the sample mean (x) ± 2σ, gives an even wider range (0.7122 ± 0.0055; Figure 2). If we complement the commonly used x ± 2σ, with the confidence interval (CI) of the mean, the most likely range for bioavailable strontium can be distinguished. In the case of Halland, the CI, at a confidence level of 95%, would be 0.7111–0.7133 (α = 0.05, 1σ = 0.0027, n = 24). B1 and B2 have higher ⁸⁷Sr/⁸⁶Sr values than the 95% CI but are well within the Halland baseline, as proposed earlier. Although we cannot exclude coastal Halland as lived environs for these individuals, as these regions fit within the proposed Halland ⁸⁷Sr/⁸⁶Sr baseline, it is perhaps more probable that they lived most of their lives further inland, in Halland or north/east Scania, as suggested by the 95% CI (Figure 2). More reference values may change this interpretative framework, but currently, only one sample, B7 from 10^th century ce, reflects a likely local coastal ⁸⁷Sr/⁸⁶Sr ratio, 0.7109. This person may also have lived somewhere along the Scanian west coast or western Denmark (excluding Bornholm), but a local provenience on Bjäre is perhaps more reasonable.

Figure 2

2021, p. 23) and from Halland (Supplementary Table S1; Klassen et al., 2020, p. 414).

4.4 Bovidae as mobility indicators

Cattle and sheep/goats were among the other archaeological specimens examined in this study. Starting with the cattle, their known death ages let us contextualize the ⁸⁷Sr/⁸⁶Sr values temporally. B3, the third premolar in the mandible (P3–), belonged to an animal that died after 3 years of age. This tooth, however, is formed earlier in life, between 6 and 18 months, following Brown et al. (1960, p. 21). We use the longer enamel maturation period of approximately 6 months compiled by Kohn (2004). This means that the ⁸⁷Sr/⁸⁶Sr value is an average of the geological background for this animal between 0.5 to approximately 2 years of age. The ⁸⁷Sr/⁸⁶Sr value 0.7115 (Figure 2) is close to the Bjäre Peninsula ratios, meaning that this animal may have lived most of its life in this area.

B6 is the first mandibular molar from a mandible belonging to an animal slaughtered aged 34–43 months. It was retrieved from a pre-Roman Iron Age cooking pit. For cattle, the first molar's crown is 1/3 formed in utero and completed at 2–3 months. Using the enamel maturation period of large bovids, the ⁸⁷Sr/⁸⁶Sr value of the first molar approximately reflects the prenatal period and the first 8–9 months of life. The ⁸⁷Sr/⁸⁶Sr value of 0.7157, is indicative of it not being bred on Bjäre but rather from zones with higher ⁸⁷Sr/⁸⁶Sr values, most probably inland toward the east, such as inland Halland or northeastern Scania. The two sheep/goat bones (B4, B5) also showed relatively high ⁸⁷Sr/⁸⁶Sr values, 0.7127 and 0.7142, indicating a lived-in environment outside of Bjäre (Figure 2). These results support the a priori assumption of this article that domesticated mammals can move distances and that they are therefore likely not suitable for constructing strontium isotope baselines.

5 Conclusion

This short communication presents new strontium isotope data from the Bjäre Peninsula and Halland in southern Sweden. Based on plant data, we suggest 0.7100 ± 0.0024 (2σ, n = 8) as a specific ⁸⁷Sr/⁸⁶Sr baseline for Bjäre. The results indicate that the Bjäre Peninsula is affected by marine sea-spray effects, as the values are lower than values from inland localities with similar geological backgrounds. There is a greater ⁸⁷Sr/⁸⁶Sr variation in Halland, with generally higher ⁸⁷Sr/⁸⁶Sr values. Together with previously published faunal samples, we suggest a revised strontium isotope baseline for Halland to 0.7122 ± 0.0055 (2σ, n = 24). We applied these baselines to seven archaeological samples from human, cattle, and sheep/goat remains, dating to the Late Neolithic/Early Bronze Age, pre-Roman Iron Age, and the late Viking Age, retrieved from recent excavations in the southern part of Bjäre. The results from both human and animal samples are interesting for the discussion of mobility and interaction. Several samples yielded ⁸⁷Sr/⁸⁶Sr values indicative of provenience outside of Bjäre, which highlights the potential of studying mobility, communication, and interaction in this area through isotopic analysis. This research is incremental and can be added to the recent surge of studies, increasing the understanding of bioavailable ⁸⁷Sr/⁸⁶Sr in southern Scandinavia. Importantly, more studies refining and modifying our suggested ⁸⁷Sr/⁸⁶Sr baseline will surely follow in the future.

References

• 1

  BlankM.SjögrenK.-G.KnipperC.FreiK. M.StoråJ. (2018). Isotope values of the bioavailable strontium in inland southwestern Sweden—A baseline for mobility studies. PLoS ONE13:e0204649. 10.1371/journal.pone.0204649

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2

  BoethiusA.KjällquistM.Kielman-SchmittM.AhlströmT.LarssonL. (2022). Diachronic forager mobility: untangling the Stone Age movement patterns at the sites Norje Sunnansund, Skateholm and Västerbjers through strontium isotope ratio analysis by laser ablation. Archaeol. Anthropol. Sci.14:176. 10.1007/s12520-022-01640-0

  • CrossRef
  • Google Scholar

• 3

  BrownW. B.ChristoffersonP. V.MasslerM.WeissM. B. (1960). Postnatal tooth development in cattle. Am. J. Veter. Res.21, 7–34.

  • Google Scholar

• 4

  FreiK. M.BergerbrantS.SjögrenK. G.JørkovM. L.LynnerupN.HarvigL.et al. (2019). Mapping human mobility during the third and second millennia BC in present-day Denmark. PLoS ONE14:e0219850. 10.1371/journal.pone.0219850

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5

  HoltE.EvansJ. A.MadgwickR. M. (2021). Strontium (87Sr/86Sr) mapping: a critical review of methods and approaches. Earth-Sci. Rev.216:103593. 10.1016/j.earscirev.2021.103593

  • CrossRef
  • Google Scholar

• 6

  KlassenL.PriceD. T.SjögrenK.-G.WincentzL.PhilippsenB. (2020). “Strontium and lead isotope studies of faunal and human remains from Kainsbakke and Kirial Bro” in The Pitted Ware Culture on Djursland: Supra-regional significance and contacts in the Middle Neolithic of southern Scandinavia, ed. L. Klassen (Aarhus: Aarhus University Press), 407–446.

  • Google Scholar

• 7

  KohnM. J. (2004). Comment: Tooth enamel mineralization in ungulates: implications for recovering a primary isotopic time-series, by B. H. Passey and T. E. Cerling (2002). Geoch Cosmoch Acta68, 403–405. 10.1016/S0016-7037(03)00443-5

  • CrossRef
  • Google Scholar

• 8

  KristiansenK. (2022). Archaeology and the Genetic Revolution in European Prehistory. Cambridge: Cambridge University Press. 10.1017/9781009228701

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9

  Ladegaard-PedersenP.SabatiniS.FreiR.KristiansenK.FreiK. M. (2021). Testing late bronze age mobility in southern Sweden in the light of a new multi-proxy strontium isotope baseline of Scania. PLoS ONE. 16:e0250279. 10.1371/journal.pone.0250279

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10

  MacheridisS. (2024). “Osteologisk analys,” in En väg till Bjäres historia: Arkeologisk undersökning 2021, eds. T. Björk, A. Nilsson (Kristianstad: Sydsvensk Arkeologi), 124–132.

  • Google Scholar

• 11

  McArthurJ. M.HowarthR. J.BaileyT. R. (2001). Strontium isotope stratigraphy: LOWESS Version 3: best fit to the marine sr-isotope curve for 0–509 ma and accompanying look-up table for deriving numerical age. J. Geol.109, 155–170. 10.1086/319243

  • CrossRef
  • Google Scholar

• 12

  MeiriM.StockhammerP. W.MaromN.Bar-OzG.Sapir-HenL.MorgensternP.et al. (2017). Eastern mediterranean mobility in the bronze and early iron ages: inferences from ancient DNA of pigs and cattle. Sci. Rep.7:701. 10.1038/s41598-017-00701-y

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 13

  SlovakN. M.PaytanA. (2012). “Applications of sr isotopes in archaeology in handbook of environmental isotope geochemistry,” in Advances in Isotope Geochemistry, eds. M. Baskaran (Springer, Berlin, Heidelberg). 10.1007/978-3-642-10637-8_35

  • CrossRef
  • Google Scholar

• 14

  SvenssonA.LiahaugenS. (forthcom.). Ett utsnitt av Bjärehalvöns historia i långtidsperspektiv. Undersökning inför dricksvattenledning mellan Grevie kyrkby och Rammsjö. Sydsvensk Arkeologi Rapport 2024:XX. Kristianstad: Sydsvensk Arkeologi.

  • Google Scholar

• 15

  WilhelmsonH. (2017). Perspectives from a human-centred archaeology: Iron Age people and society on Öland. Doctoral Thesis. Acta Archaeologica Lundensia, Series altera in 8o, no 68 | Studies in Osteology 2. Lund: Lund University.

  • Google Scholar

• 16

  WolskaB. (2020). Applying isotope analyses of cremated human bones in archaeological research – a review. Analecta Archaeol. Ressov.15, 7–16. 10.15584/anarres.2020.15.1

  • CrossRef
  • Google Scholar