INTRODUCTION
During a large-scale field survey carried out in the Pollino National park in Southern Italy, different Heldreich’s pines (Pinus heldreichii H. Christ) were sampled in order to assess their ages and study their growth history trajectory. Indeed, previous studies carried out in the same area allowed the identification of trees with ages approaching one thousand years (Serre-Bachet Reference Serre-Bachet1985; Biondi Reference Biondi1992). In this paper we present the results of the analyses performed on one of these ancient trees, which was identified as the potentially oldest specimen. The tree was named Italus after the name of the legendary king of Enotri, the population ruling this region during the Bronze Age. The determination of the tree age was not possible in this case by simple ring counting and cross-dating because of the degradation of the innermost part of the trunk (hollow stem). It required the development of a novel approach based on wiggle match (WM) accelerator mass spectrometry radiocarbon (AMS 14C) dating performed on tree-ring sequences extracted from the exposed roots. Through this approach, the relative dating information established between the different samples is used as a priori known information to constrain, in a Bayesian framework, the obtained 14C ages, resulting in a significant refinement of the achievable chronological resolution. The second stage of the study exploits the data from Italus as a proxy record of the radiocarbon concentration in the past terrestrial atmosphere.
In this paper we concentrated on this second last aspect by measuring the 14C concentration in single rings extracted from the tree trunk in the period 984–1003 CE. The aim of the study was the identification of the 993–994 CE rapid increase in the atmospheric radiocarbon concentration as linked with one of the so-called Miyake events. These events, so far identified in 774–775 CE (Miyake et al. Reference Miyake, Nagaya, Masuda and Nakamura2012), 993–994 CE (Miyake et al. Reference Miyake, Masuda and Nakamura2013, Reference Miyake, Masuda, Hakozaki, Nakamura, Tokanai, Kato, Kimura and Mitsutani2014, Reference Miyake, Timothy Jull, Panyushkina, Wacker, Salzer, Baisan, Lange, Cruz, Masuda and Nakamura2017), and 660 BCE (Park et al. Reference Park, Southon, Fahrni, Creasman and Mewaldt2017), are generally explained as due to very large SPEs (solar proton events) or a series of SPEs (Mekhaldi et al. Reference Mekhaldi, Muscheler, Adolphi, Aldahan, Beer, McConnell, Possnert, Sigl, Svensson, Synal, Welten and Woodruff2015). The study of these events has attracted great interest in the scientific community in the last years and research efforts are aimed at clearly identifying the causal mechanism or the mechanisms, their exact timing (also with a sub-annual resolution) and event periodicity and magnitude as summarized by Dee and Pope (Reference Dee and Pope2016).
MATERIALS AND METHODS
Italus is an Heldreich’s pine (Pinus heldreichii H. Christ) with a dead spike top and strip-bark trunk, typical attributes found in old conifers. The diameter at chest height is ca. 1.6 m and the tree is located at an altitude of about 1850 m along the steep rocky slopes of the Pollino Massif in the Pollino National park in Southern Italy (Figure 1).
Location of the studied tree in the Pollino National Park, Southern Italy.
The present study was performed in different steps and was intended to illustrate new aspects of the methods for absolutely dating the tree and then identifying the 993–994 Miyake event in the xylem ring sequence.
Determining the Age of Italus
The most straightforward method for the determination of the age of a living tree is, of course, ring counting on extracted cores. The tree-ring series should be dendrochronologically cross-dated to obtain an accurate age. This was not directly possible in the case of Italus because the innermost part of the stem was highly degraded and the corresponding tree-rings missing. Though the dating strategy has been detailed in previous papers (Piovesan et al. Reference Piovesan, Biondi, Baliva, Presutti Saba, Calcagnile, Quarta, D’Elia, De Vivo, Schettino and Di Filippo2018a, Reference Piovesan, Biondi, Baliva, Calcagnile, Quarta and Di Filippo2018b), we report here, for reason of completeness, the employed strategy which was structured as follows:
1. Extraction of 5-mm-diameter cores by an increment borer at different positions around the stem, determination and cross-dating of the tree-ring width series, cross-dating the different series with the available master chronologies for the same species in the same area; determination of the age of the wood cores.
2. Extraction of four 5-mm-diameter cores from exposed roots (Figure 2) and as close as possible to tree germination pith, development of the ring-width series, cross-dating of the different series and comparison with available reference chronologies. Table 1 shows the number of rings of each of the four tree-ring series and the relative cross-dating. From each sequence two samples were selected to be submitted to AMS 14C dating. Thanks to the floating root tree-ring chronology it was possible to “wiggle match” the 14C data to the calibration curve by using the dendrochronology information as statistical constraints for the construction of a Bayesian model (wiggle match dating).
The Italus tree with the exposed roots shown in the white circle. Red dots indicate the sampling points where roots’ cores were extracted. (Please see electronic version for color figures.)
Summary of the samples extracted from the four exposed roots, laboratory code, corresponding relative dating and conventional radiocarbon ages.
Selecting the Sequence for High-Resolution Analyses
Twelve single-year samples were identified, selected, and cut from one of the tree-ring sequences extracted from the stem and with ages corresponding to the period between 984–1003 CE thus bracketing the 993–994 CE Miyake event. The samples had masses between 11 and 16 mg and were submitted for AMS 14C dating at CEDAD (Centre for Applied Physics, Dating and Diagnostics), Department of Mathematics and Physics “Ennio de Giorgi”-University of Salento.
The samples, as well as the previous ones extracted from the roots, were processed by following the base-acid-base-acid-bleaching (BABAB) procedure as described in Nemec et al. (Reference Nemec, Wacker, Hajdas and Gäggeler2010). The BABAB procedure is a modified version of the widely used alkali-base-alkali (ABA) protocol with two important modifications. The first one consists of the first alkali step at the beginning of the protocol (4% NaOH at 75°C for 8 hr). The second modification is the addition of a bleaching step after the ABA and consisting in the addition of 5 mL of 10% NaClO2, 5 mL of deionized water, and 4% HCl to the wood suspension followed by 15 min of ultrasonication. The yield of this chemical process, established as the ratio between the extracted purified material and the original sample mass, ranged from 20 to 30%. The amount of purified material available for the following steps ranged between 2 and 4 mg. The purified material was then sealed under vacuum in a quartz tube together with CuO and silver wool and then combusted to CO2 at 900°C. The carbon dioxide was then cryogenically purified and transferred to the graphitization lines where it was reduced at 600°C to graphite by using H2 as reductant on iron powder as the catalyst (D’Elia et al. Reference D’Elia, Calcagnile, Quarta, Rizzo, Sanapo, Laudisa, Toma and Rizzo2004). Extracted graphite was then pressed in the aluminium cathodes of the ion source of the AMS 14C beam line in operation at CEDAD where 14C/12C ratios were measured (Calcagnile et al. Reference Calcagnile, Quarta, D’Elia, Rizzo, Gottdang, Klein and Mous2004, Reference Calcagnile, Quarta and D’Elia2005). Isotopic ratios were then corrected for mass fractionation by using the ∂13C term measured on line with the accelerator and machine and processing blanks. 14C ages and Δ14C values were then calculated according to Stuiver and Polach (Reference Stuiver and Polach1977).
RESULTS AND DISCUSSION
The cross-dating of the different stem tree-ring series also with available reference or master chronology for the same species in the same area, allowed us to date the last preserved ring of the core to 955 CE. Considering the presence of the hole in the middle of the trunk, with a radius of 23.8 cm, this has to be considered as a minimum age of the tree.
As detailed elsewhere (Piovesan et al. Reference Piovesan, Biondi, Baliva, Presutti Saba, Calcagnile, Quarta, D’Elia, De Vivo, Schettino and Di Filippo2018a), a first estimation of the age of the tree was then obtained by extrapolating the number of missing rings (assuming a circular shape) from the radius of the central hole and the measured average value of the last 20 or 50 rings: 1.05 and 1.16 mm, respectively. By adding the estimated number of missing rings (205–227) to the cross-dated age of the oldest dated ring (955 CE), the final age of the tree could be estimated to be between 727–749 CE. We underline that this is a rough estimation, which is likely to produce an overestimated number of missing rings considering typical juvenile ring patterns for conifers, but it does provide a first approximation of the tree lifespan.
A much more robust estimation of the longevity of Italus was obtained by analyzing the results of the 14C analyses performed on the roots samples. The analysis of the ring-width patterns obtained from the cores from the roots showed little statistical correlation with the stem growth patterns and prevented the direct cross-dating of the root sequences. However, the four sequences obtained from the four root cores showed good correlation between themselves and allowed the production of a 320-yr-long floating chronology (Wrońska-Wałach et al. Reference Wrońska-Wałach, Sobucki, Buchwał, Gorczyca, Korpak, Wałdykowski and Gärtner2016; Piovesan et al. Reference Piovesan, Biondi, Baliva, Calcagnile, Quarta and Di Filippo2018b). A detailed analysis of the roots cross-dating, including the statistical synchronization between the floating ring-width series from the external roots, is given in the supplemental material of Piovesan et al. (Reference Piovesan, Biondi, Baliva, Presutti Saba, Calcagnile, Quarta, D’Elia, De Vivo, Schettino and Di Filippo2018a). In this way, each of the eight samples taken from the four roots could be relatively dated and the age gap between each sample established (Table 1). The relative dating information and the measured 14C ages where then used to match the obtained data with the IntCal13 atmospheric dataset (Reimer et al. Reference Reimer, Bard, Bayliss, Beck, Blackwell, Bronk Ramsey, Buck, Cheng, Edwards, Friedrich, Grootes, Guilderson, Haflidason, Hajdas, Hatté, Heaton, Hoffmann, Hogg, Hughen, Kaiser, Kromer, Manning, Niu, Reimer, Richards, Scott, Southon, Staff, Turney and van der Plicht2013; Figure 3). The refined analysis was performed by building a Bayesian model in OxCal v.4.3 using the D_Sequence function (Galimberti et al. Reference Galimberti, Bronk Ramsey and Manning2004; Bronk Ramsey Reference Bronk Ramsey2009), which allowed dating of the oldest 14C dated rings (LTL17796A) to 781–804 CE with a probability of 68.2% and 770–816 with a probability of 95.4% (Figure 4). It is worth noting that also this age has to be considered as a minimum age of the tree as it seems to be confirmed by comparing the 14C-determined age with the age extrapolated from the stem (727–749 CE), especially when, as mentioned before, it is likely that this is an overestimated age of the tree.
Visual match between the 14C ages obtained on the tree-ring samples extracted from the stem and the IntCal13 calibration curve (errors bars refer to 1σ confidence level).
Wiggle match Bayesian model obtained for the analyzed samples.
Our results show that Italus is older than Adonis, a tree of the same species growing in Northern Greece which, with a dendrochronologically determined age of 1076 yr, was so far considered the oldest scientifically dated living tree in Europe (Konter et al. Reference Konter, Krusic, Trouet and Esper2017).
The 14C concentrations measured for the single-ring samples extracted from the stem and expressed as Δ14C are shown in Figure 5 as a function of the growth year determined by dendrochronological methods. Our results show a sharp increase in the 14C concentration in 993 CE corresponding to 11.3 ± 3.8‰ in the Δ14C scale. This change-value compares well with those published by others for this Miyake-event, for instance for Denmark (10.5 ± 3.4‰) and Japan (11.3 ± 3.5‰) (Miyake et al. Reference Miyake, Masuda, Hakozaki, Nakamura, Tokanai, Kato, Kimura and Mitsutani2014; Fogtmann et al. Reference Fogtmann-Schulz, Østbø, Nielsen, Olsen, Karoff and Knudsen2017). Also the shape of the observed increase is consistent with what is generally observed by others, namely a slight decline before 993 CE, a sharp increase, and a slow decline over the following years probably because of the exchanges with the terrestrial carbon pools. Our results then confirm that the 993–994 CE event was a planet-scale phenomenon, not depending on the hemisphere and whose magnitude does not depend on latitude, at least within the experimental uncertainties associated with the measurements (Büntgen et al. Reference Büntgen, Wacker, Galván, Arnold, Arseneault, Baillie, Beer, Bernabei, Bleicher and Boswijk2018). This observation further supports the hypothesis of a solar origin of the measured increase as point out by Güttler et al. (Reference Güttler, Adolphi, Beerc, Bleicher, Boswijk, Christl, Hogg, Palmer, Vockenhuber, Wacker and Wunder2015) for the 774–775 CE event.
Measured radiocarbon concentrations, decay corrected and expressed as Δ14C as a function of the dendrochronologically determined age for the single year samples extracted from the stem (errors bars refer to 1σ confidence level).
We also note that our results seem to place the event in 993 CE although we do not have sub-annual information because no distinction was made between early- and late-wood. We also note that Hayakawa et al. (Reference Hayakawa, Tamazawa, Uchiyama, Ebihara, Miyahara, Kosaka, Iwahashi and Isobe2017) have found evidence, in ancient written sources, of Aurora observations in the Korean Peninsula, Saxon cities (nowadays Germany) and Ireland clustering in the period between the late 992 and early 993 CE (Hayakawa et al. Reference Hayakawa, Tamazawa, Uchiyama, Ebihara, Miyahara, Kosaka, Iwahashi and Isobe2017). On the basis of these observations, Hayakawa et al. conclude that, if the recorded auroras were associated with the 993–994 Miyake event, the 14C spike should have been in 993 CE. Though further investigations are surely needed, our data obtained on Italus seems to support this hypothesis.
CONCLUSIONS
The combined application of 14C AMS dating and dendrochronology allowed us to absolutely date with a high level of confidence Italus, a Heldreich’s pine growing in the Pollino National Park in Southern Italy. The obtained results allow us to determine a minimum age of the tree ranging between 1201 and 1247 yr in 2017 (with a probability of 95.4%). The comparison of the obtained age with data in the literature data shows that this is currently the oldest scientifically dated tree living in Europe.
Italus can then be considered a proxy record of the climatic and environmental conditions in an area located in the middle of the Mediterranean Basin. As a first application we were able to identify the 993–994 CE increase in the 14C atmospheric concentration (Miyake event). It is worth noting that this high-resolution 14C date is also a direct confirmation of the cross-dated age obtained for the stem through dendrochronological methods. Our results confirm the shape and the amplitude of the increase as observed by others in different locations in both hemispheres. Concerning the timing of the event, our current data seem to indicate that the event occurred in 993 CE though further investigations are needed and already planned.
Concerning future projects, efforts will be devoted to identifying the 993–994 CE Miyake event in the root-derived tree-ring sequences which will also result in a significant improvement of the achieved chronological resolution. Considering the average width of the tree-rings and so the mass of the available material, this will require the use of the gas-accepting ion source in operation at CEDAD (Braione et al. Reference Braione, Maruccio, Quarta, D’Elia and Calcagnile2015; Maruccio et al. Reference Maruccio, Quarta, Braione and Calcagnile2017). Future sampling campaigns will be also carried out in order to identify other roots extending further back in time and, hopefully, including also the other 774–775 CE Miyake event.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit https://doi.org/10.1017/RDC.2019.37.
ACKNOWLEDGMENTS
This research project was funded by Pollino National Park (MATTM Biodiversity 2014). F. Biondi was supported, in part, by a grant from the Office of Research and Innovation at the University of Nevada, Reno (USA).
