Volume 41, Issue 8 p. 3004-3010
Research Letter
Free Access

Excursions in the 14C record at A.D. 774–775 in tree rings from Russia and America

A. J. Timothy Jull

Corresponding Author

A. J. Timothy Jull

NSF Arizona AMS Laboratory, University of Arizona, Tucson, Arizona, USA

Department of Geosciences, University of Arizona, Tucson, Arizona, USA

Institute for Nuclear Research, Debrecen, Hungary

Correspondence to: A. J. T. Jull,

[email protected]

Search for more papers by this author
Irina P. Panyushkina

Irina P. Panyushkina

Laboratory of Tree-Ring Research, University of Arizona, Tucson, Arizona, USA

Search for more papers by this author
Todd E. Lange

Todd E. Lange

NSF Arizona AMS Laboratory, University of Arizona, Tucson, Arizona, USA

Search for more papers by this author
Vladimir V. Kukarskih

Vladimir V. Kukarskih

Laboratory of Dendrochoronology, URAN Institute of Plant and Animal Ecology, Yekaterinburg, Russia

Search for more papers by this author
Vladimir S. Myglan

Vladimir S. Myglan

Department of Humanities, Siberian Federal University, Krasnoyarsk, Russia

Search for more papers by this author
Kelley J. Clark

Kelley J. Clark

Laboratory of Tree-Ring Research, University of Arizona, Tucson, Arizona, USA

Search for more papers by this author
Matthew W. Salzer

Matthew W. Salzer

Laboratory of Tree-Ring Research, University of Arizona, Tucson, Arizona, USA

Search for more papers by this author
George S. Burr

George S. Burr

NSF Arizona AMS Laboratory, University of Arizona, Tucson, Arizona, USA

Department of Geosciences, National Taiwan University, Taipei, Taiwan

Search for more papers by this author
Steven W. Leavitt

Steven W. Leavitt

Laboratory of Tree-Ring Research, University of Arizona, Tucson, Arizona, USA

Search for more papers by this author
First published: 04 April 2014
Citations: 87

Abstract

The calibration of radiocarbon dates by means of a master calibration curve has been invaluable to Earth, environmental and archeological sciences, but the fundamental reason for calibration is that atmospheric radiocarbon content varies because of changes in upper atmosphere production and global carbon cycling. Improved instrumentation has contributed to high-resolution (interannual) radiocarbon activity measurements, which have revealed sudden and anomalous activity shifts previously not observed at the common resolution of 5–10 years of most of the calibration scale. One such spike has been recently reported from tree rings from Japan and then again in Europe at A.D. 774–775, for which we report here our efforts to both replicate its existence and determine its spatial extent using tree rings from larch at high latitude (northern Siberia) and bristlecone pine from lower latitude (the White Mountains of California). Our results confirm an abrupt ~ 15‰ 14C activity increase from A.D. 774 to 776, the size and now the hemispheric extent of which suggest that an extraterrestrial influence on radiocarbon production is most likely responsible.

Key Points

  • Excursion in 774-776 A.D. due to a rapid change in 14C production
  • Event must be global and uniform in scale
  • Phenomenon is reproduced in two new locations, making a total of five

1 Introduction

Cosmic rays interact with the Earth's atmosphere to produce secondary particles and also products. One of the most well known of these products is carbon-14, which is produced in the atmosphere by the action of secondary galactic cosmic ray (GCR) neutrons on nitrogen in the atmosphere due to the reaction 14N(n,p)14C [van der Plicht, 2007; Burr, 2007]. The mean production rate has been the subject of much discussions in the past, but the current consensus values are 1.6–2.0 atoms/cm2/s [Masarik and Reedy, 1995; Mak et al., 1999; Kovaltsov et al., 2012].

In general, 14C levels are measured as fraction of modern carbon, F [Donahue et al., 1990]. Either 14C/13C or 14C/12C ratios may be used to calculate F. Here we consider only 14C/13C ratios. In this case F is defined as
urn:x-wiley:00948276:media:grl51610:grl51610-math-0001(1)
where (14C/13C)S[−25] is the measured ratio for the sample, blank corrected and adjusted to a δ13C = −25‰; and (14C/13C)1950[−25] is the measured ratio of the standard, blank corrected, adjusted to δ13C = −25‰, and recalculated to 1950 A.D. [see Burr and Jull, 2009].
It is often of interest to understand past levels of 14C. This can be done by converting the 14C measured in a sample of known age in the past to a Δ14C value according to equation 2
urn:x-wiley:00948276:media:grl51610:grl51610-math-0002(2)
where F is defined above in equation 1, λ is the true decay constant of 14C (1.209 × 10−4 yr−1) and t is the known age of the material [Stuiver and Polach, 1977]. At the time the sample formed Δ14C is a measure of the radiocarbon content of the atmosphere, relative to the A.D. 1950 atm, expressed in per mil (‰). This term allows time-dependent changes in atmospheric 14C to be accurately identified.

In a recent series of papers studying annual tree rings, Miyake et al. [2012] reported on the existence of excursions in the radiocarbon record at A.D. 774–775, followed by a less intense event at A.D. 993–994 [Miyake et al., 2013]. The A.D. 774–775 “spike” is observed as a change in Δ14C of ~ 12–15‰ in a 1–2 year period. Apart from the event at A.D. 993–994, there are no other reported excursions of this magnitude in the last several thousand years [Usoskin and Kovaltsov, 2012]. The initial work of Miyake et al. [2012] on the A.D. 774–775 event was based on annual rings from Japanese cedar trees. The first event has been independently confirmed by other investigators on European oak trees [Usoskin et al., 2013], with a change in Δ14C of ~ 15‰. In addition, Güttler et al. [2013a, 2013b] report on a record from the Southern Hemisphere using Kauri wood from New Zealand. This Kauri record shows the same amplitude in Δ14C but with a small offset due to the Southern Hemisphere regional effect. Liu et al. [2014] have recently reported on a similar excursion in Δ14C determined from dated corals in the South China Sea.

Our purpose in this study is to investigate whether this signal is truly global by selecting continental locations from the western United States and northwestern Siberia. These sites were chosen to be as diverse as possible from the previously published locations. A second goal was to elucidate any changes in regional effects that might be manifest in the results. These locations should also sample different regional 14C offsets and would not show effects due to either the ocean (as in the case of Japan) or the Southern Hemisphere [Stuiver et al., 1998; Hogg et al., 2013]. Third, we then discuss the possible implications of these results for past extreme radiation events or other possible cosmic events.

2 Samples and Methods

In this study, we investigate the variations in 14C in annual tree-ring records from different locations in the western United States (California White Mountains bristlecone pine [Pinus longaeva] 37°77′N, 118°44′W, 3539 mean annual sea level (MASL)) and Siberia (Yamal Peninsula larch [Larix sibirica Ldb] 67°31′N, 70°40′E, 350 MASL). All samples were crossdated and absolutely dated with large-sample size tree-ring records using standard dendrochronological methods [Salzer et al., 2009; Hantemirov and Shiyatov, 2002]. Annual tree rings from each site were separated for the period 760 to 800 A.D.

Samples were pretreated to extract cellulose in the accelerator mass spectrometry (AMS) laboratory. These were then combusted to CO2 and converted to graphite using standard procedures [Jull et al., 2008]. The graphite powders produced are pressed into AMS targets and measured using the National Electrostatics Corporation AMS system at the University of Arizona, at a terminal voltage of 2.5 MV. The 14C/13C ratio of the sample is compared to known standards (Oxalic_I and II, National Institute of Standards and Technology standards SRM4990B and 4990C, respectively), and the result corrected to the measured value of δ13C made offline on a stable isotope mass spectrometer, giving a value for fraction of modern carbon (F). Details of the AMS calculations at Arizona are given by Donahue et al. [1990] and Burr et al. [2007]. The 14C results are also converted from F to Δ14C as discussed in equation 1.

3 Results

3.1 Time Variance of the Measurements

Our results are shown in Figure 1 and also summarized in the supporting information (Table S1). Here we plot our results for both bristlecone pine and Siberian larch from known-age tree rings. We observe a marked offset in the Δ14C value between A.D. 774 and 776. There is remarkable agreement between our record and that of the two previous published results obtained by Miyake et al. [2012] and Usoskin et al. [2013]. The data of Güttler et al. [2013a, 2013b] show a similar effect. In Figure 2, we compare our results to these previous records. The amplitude of the effect appears to be almost identical between the different data sets, and the differences between the data are small regional offsets of the order of 3–5‰ [Stuiver et al., 1998].

Details are in the caption following the image
Record of Δ14C (‰) between A.D. 760 and A.D. 800 in tree rings from bristlecone pine (White Mountains, California, USA) and larch (Yamal Peninsula, Siberia, Russian Federation), compared to the record of Miyake et al. [2012] from Japanese cedar.
Details are in the caption following the image
Record of Δ14C (‰) between A.D. 760 and A.D. 800 in tree rings from White Mountains bristlecone pine and Yamal Peninsula larch, compared to the record in German oak [Usoskin et al., 2013].

Upon closer inspection, some additional details are apparent. First, the rise in Δ14C is uniform between the different records but with some small differences. In the Miyake et al. [2012] record, the increase appears to begin in 774 and continue to A.D. 776. A similar effect is seen in the bristlecone pine data. However, the European oak data [Usoskin et al., 2013] and the Siberian larch results indicate that by A.D. 774, the tree rings were already elevated above the previous year. A similar effect appears to be present in the record of Güttler et al. [2013a]. This suggests that the incorporation of the 14C signal from the event is not completely uniform. An analysis of our bristlecone pine and Siberian larch indicates that they are statistically indistinguishable based on the student's t test, t = 1.489 with 40 degrees of freedom. In all cases, the peak of the event in the 14C record is in A.D. 776.

4 Discussion

4.1 Size of the 14C Effects

The atmosphere today contains ~ 750 Gt of carbon [Schimel et al., 1996], equivalent to ~ 4.4 × 1028 14C atoms at a prebomb 14C/12C value of 1.177 × 10−12. In A.D. 774, we expect that due to the lower concentration of CO2, this value was closer to 600 Gt or ~ 3.5 × 1028 14C atoms (~ 814 kg 14C). With an annual average production rate of ~ 1.8 ± 0.2 14C/cm2/s [Kovaltsov et al., 2012], this would produce ~ 3 × 1026 14C/yr (6.7 kg/yr). Because the amount of 14C in the atmosphere is clearly much higher than the annual average production rate, a large amount of change in production is needed to explain the increase in the A.D. 774–776 sequence studied here. This value becomes even larger when exchange of atmospheric carbon with the surface ocean and biosphere is considered. Hence, it is clear from these values that a twofold increase in production rate would only increase the 14C in the atmosphere a few percent. Since the atmosphere is in equilibrium with the oceans and the biosphere, exchanging ~ 90 Gt carbon per year with the ocean, and experiencing a net loss of ~ 60 Gt/yr to the biosphere [Schimel et al., 1996], researchers have resorted to box model calculations to explain these effects [e.g., Damon et al., 1995; Usoskin and Kovaltsov, 2012]. Because these models are all based on the same values for the carbon cycle (e.g., for reference levels), it is not necessary to repeat these calculations again. Previous authors have estimated the production rate of 14C (atoms/cm2/yr) required to explain these observations at between 1.3 × 108 and 1.7 × 108 atoms/cm2/yr [Usoskin et al., 2013; Pavlov et al., 2013; Miyake, 2014]. An earlier higher estimate from the original Miyake et al. [2012] paper has been reduced by a factor of 4 [Miyake, 2014].

4.2 Constraints on the Intensity of a Possible Solar Flare Event

The initial work on this event by Miyake et al. [2012] and Usoskin et al. [2013] has generated a large amount of modeling as well as speculation as to the cause of this 14C excursion. Almost all studies propose that there is a rapid change to the production rate of 14C, which must be several times the normal galactic cosmic ray (GCR) production rate. Most authors speculate on a solar cause due to enhanced solar proton fluxes from solar cosmic ray or coronal mass ejection events. Lingenfelter and Ramaty [1970] were the first to note that a large solar proton event could produce more 14C than the average annual GCR rate. Hambaryan and Neuhäuser (2013) and Pavlov et al. [2013] favor a gamma ray burst (GRB) as a source. Other authors have proposed solar proton events [Usoskin and Kovaltsov, 2012; Thomas et al., 2013; Usoskin et al., 2013], a cometary impact into the Sun generating intense solar flares [Eichler and Mordecai, 2013] and a comet impact into the Earth's atmosphere [Liu et al., 2014].

These discussions focus primarily on possible explanations for an external source of cosmic radiation which could be sufficient to produce this change in the 14C production rate. According to Miyake et al. [2012], the production rate in this year must have been enhanced from the mean value by between 3.6 and 10 depending on the period of the event and the assumptions about storage in the stratosphere, however, this has since been reduced by a factor of 4 [Miyake, 2014]. Others have estimated the production rate of 14C (atoms/cm2/yr) required to explain these observations at between 1.3 × 108 and 1.7 × 108 atoms/cm2/yr [Usoskin et al., 2013; Pavlov et al., 2013; Miyake, 2014]. Of course, all these estimates depend on the assumptions and models used.

Several authors have proposed different solar particle fluxes necessary to produce the effects discussed. Since the flux of solar particles is dependent on the energy spectrum of the event, there can be some variability in these estimates. We therefore standardize these reports for a flux E > 30 MeV [Reedy, 1996]. Miyake [2014] has revised their original estimate for the proposed solar proton event total fluence (E > 30 MeV) necessary to produce the A.D. 774–775 14C effect to 5 × 1010 p/cm2. Usoskin et al. [2013] estimated a fluence of 4.5 × 1010 p/cm2 and Cliver et al. [2014] have estimated 8 × 1010 p/cm2.

Shea and Smart [1990] summarized all major solar proton events between 1955 and 1986. Subsequently, Reedy [1996] summarized solar flare events over the period A.D. 1954–1996 and found a few events of 1 × 1010 p/cm2 (for E > 30 MeV) but no larger events.

The lunar record provides a good baseline of the long-term fluence of solar cosmic rays, since it provides an integral record. Jull et al. [1998] derived a long-term flux of 1.3 × 109 p/cm2/yr, based on spallogenic 14C in lunar rock surfaces. A recent reinvestigation by Kovaltsov and Usoskin [2014] confirms this estimate. Hence, any very large events in the past must be small in number, since they would also create radioactivity in lunar samples. Reedy [1996] constrains very large SPEs to a probability of <10−4/year based on the tree-ring record. Lingenfelter and Hudson [1980] similarly constrain the number of SPEs over the last 7000 years to not more than one with flux > 3 × 1011 p/cm2.

As noted by Miyake et al. [2012], their original calculated fluence of >1011 p/cm2 could produce biologically damaging levels of radiation, however, this has since been revised [Miyake, 2014]. This suggests that their original fluence calculation was too high [Thomas et al., 2013], and also one of the reasons that other possible origins of the 14C event were considered. Hamabaryan and Neuhäuser [2013] proposed a gamma ray burst as an alternative source to produce the 14C excess. Pavlov et al. [2013] also proposed a GRB, with more detailed considerations. Although Miyake et al. [2012] noted a possible correlation with 10Be in ice cores, Pavlov et al. [2013] noted that such a correlation would not be required with a GRB, since most production of neutrons from a γ-ray burst would be at energy below the threshold for spallation production of 10Be from oxygen.

4.3 A Cometary Event?

In a recent paper, Liu et al. [2014] proposed that the 14C increase at A.D. 774–775 was caused by a cometary impact into the Earth's atmosphere. In their work, they observed a similar 15‰ excursion in corals about the same time. With biweekly sampling resolution in the coral, Liu et al. [2014] noted a brief 45‰ increase a few months after the initial rise in 14C, followed by a subsequent increase of 15‰. They also cited Chinese historical records from A.D. 773 that described a major atmospheric disturbance at the time, including a significant dust event [e.g., Napier, 2001]. Liu et al. [2014] estimate that a comet of ~ 1.50 × 1011 kg could produce enough 14C to account for the observed changes in atmospheric radiocarbon. This would be an object of over 1 km diameter, if we assume a density of 0.6 g/cm3. Overholt and Melott [2013] calculated that a much more massive comet (>1014 kg) would be needed to produce approximately this amount (see Figure 1a of their paper). However, their model assumes that comets receive a high cosmic ray dose during their transit of the outer solar system. If one assumes an impacting asteroid or other inner solar system object, then the amount of 14C would be much lower. Usoskin and Kovaltsov [2014] also come to similar conclusions as to the size of a cometary impact needed to produce such effects. Recent studies on the Chelyabinsk meteoroid airburst indicate that it was an object of ~ 20 m diameter, with an energy release of ~ 590 kt trinitrotoluene (TNT) equivalent and a mass of ~ 1.3 × 107 kg [Popova et al., 2013]. We note that the Tunguska airburst event is calculated to be about 20 times larger than Chelyabinsk with an energy release of 5–15 Mt TNT [Borovička et al., 2013]. An object 104 times larger mass than Chelyabinsk would probably have close to 104 times the kinetic energy resulting in a postulated energy release of some 5900 Mt TNT (2.5 × 1016 J). This would be a truly massive event, if proven correct.

A more important problem with any impact event is that it would have occurred in one hemisphere. The results shown here do not show any difference between the Northern and Southern Hemispheres at all [Güttler et al., 2013a], in marked contrast to records of the injection of nuclear 14C into the atmosphere during the period of aboveground atomic weapons testing [Hua et al., 2013].

4.4 Does the Tree-Ring Record Show Other Potential Solar Cosmic Events?

Miyake et al. [2013] report on a second event at A.D. 993–994. However, they also assert that they found no evidence for other events over the last 3000 years. There has been much discussion in the literature, not only about solar events that might be observable in the 14C record or ice cores but also other phenomena.

Usoskin et al. [2006] estimated that the A.D. 1956 solar proton event had a fluence of 109 p/cm2, but it is not easy to observe any discernable effect in Δ14C from 1956 due to the beginning of effects due to nuclear testing [Hua et al., 2013]. Similarly, any effect of the A.D. 1946 event of 6 × 109 p/cm2 [Smart et al., 2006] is not clear based on the data of Damon et al. [1973a], which shows a difference of 7 ± 5‰ relative to A.D. 1946, but only 1.6 ± 4.2‰ relative to A.D. 1945. Smart et al. [2006] proposed that the Carrington event, an observed solar flare event that also had many documented effects on the Earth, including direct observations [Carrington, 1860], aurorae, and geomagnetic phenomena [Desnains and Charault, 1859]. Although this event is estimated to have an F30 of 1.9 × 1010 cm−2 [Smart et al., 2006], there is no evidence for this event in the annual 14C record of Stuiver et al., 1998 nor of Miyake et al. [2013]. Wolff et al. [2012] reported that the F30 estimate from Smart et al. [2006] was based on ice core nitrate data, which does not show good evidence for solar flare events. Hence, this value may have been overestimated. In Figure 3, we plot the values of Δ14C for the period A.D. 1850–1900, which shows no effect at A.D. 1859–1860. It is clear, therefore, that either this event was not as intense as proposed, or the 14C system is affected only by certain solar events or perhaps other causes. Interestingly, Fan et al. [1985] and Damon et al. [1973b] appeared to identify anomalous Δ14C excursions in A.D. 1943 wood from Arizona and the Mackenzie Delta, Canada, with excursions of ~ 15‰, presumably associated with A.D. 1942 solar proton events. However, this event was not confirmed in other records [Stuiver et al., 1998], as is the case for the A.D. 774–775 event presented here.

Details are in the caption following the image
Record of Δ14C (‰) between A.D. 1850 and A.D. 1900 from Stuiver et al. [1998].

5 Conclusions

We have confirmed the A.D. 774–775 event in the 14C record at two additional locations, in the western United States and Russia. The amplitude of the event is very similar to previously reported results from Japan, Germany, and New Zealand. This emphasizes the global nature of this phenomenon and according to existing models, only a production-rate change could cause this type of event. The fact that the 14C signal is observed in five very different locations with exactly the same amplitude is remarkable in itself. The exact cause of the event is unclear, although a number of mechanisms have been proposed, all of which require an extraterrestrial origin. It appears then that the A.D. 774–775 event is the first unambiguous case of extraterrestrial enhancement of atmospheric 14C in the tree-ring record. More detailed work on annual samples from dendrochronologically dated wood may identify further examples of these excursions both spatially and temporally.

Acknowledgments

We are most grateful for the technical assistance of R. Cruz and D. Biddulph in the AMS measurements. We also acknowledge M.A. Shea, L. Wacker, J.W. Beck, and A.N. Peristykh for their helpful discussions. This work was initiated while A.J.T.J. was a guest faculty member in Hungary and he is grateful for a financial support from the Hungarian Academy of Sciences. VVK and VSM acknowledge support from Russian Science Foundation (project no. 14-17-04400).

The Editor thanks two anonymous reviewers for their assistance in evaluating this paper.