INTRODUCTION
In environmental research, past radiocarbon (14C) levels are of great interest because they allow us to gain information about carbon cycle dynamics and carbon sources in the past. These data sets are very valuable, especially in studies of human impact on carbon in the environment. There are two anthropogenic effects influencing 14C levels in the atmosphere—input of excess radiocarbon produced by nuclear technology (increasing radiocarbon activity) and input of 14C-free fossil carbon dioxide into the atmosphere (decreasing radiocarbon activity).
Atmospheric nuclear weapons tests that took place in 1950s and at the beginning of 1960s caused up to 100% increase of radiocarbon activity in the atmosphere and biosphere of the northern hemisphere (Hua et al. Reference Hua, Barbetti and Rakowski2013). Since the Partial Nuclear Test Ban Treaty in 1963, the 14C activity in the atmosphere and biosphere has been decreasing and in recent years the nuclear industry (mainly nuclear power plants and reprocessing facilities) became the main source of anthropogenic radiocarbon released into the environment. Measurements of radiocarbon content of tree rings have been used to study the effects of nuclear power plants and radioactive waste disposal facilities on surrounding environment (Stenström et al. Reference Stenström, Skog, Thornberg, Erlandsson, Hellborg, Mattsson and Persson1997; Ješkovský et al. Reference Ješkovský, Povinec, Steier, Šivo, Richtáriková and Golser2015; Janovics et al. Reference Janovics, Kelemen, Kern, Kapitány, Veres, Jull and Molnár2016; Ežerinskis et al. Reference Ežerinskis, Šapolaite, Pabedinskas, Juodis, Garbaras, Maceika, Druteikiene, Lukauskas and Remeikis2018).
Since the industrial revolution human activities using fossil fuels have become a significant source of fossil carbon dioxide released into the Earth’s atmosphere. As fossil CO2 does not contain any radiocarbon, it dilutes 14C already present in the atmosphere and therefore decreases its specific activity. This process is called Suess effect (Suess Reference Suess1955). In areas with heavy industrial activity and in heavily urbanized regions with intense traffic, the burning of fossil fuels has had a significant effect on local radiocarbon levels in the environment. Radiocarbon analysis of annual growth rings has been successfully used to determine the intensity of Suess effect, as well as to estimate CO2 emissions (Rakowski et al. Reference Rakowski, Kuc, Nakamura and Pazdur2004; Capano et al. Reference Capano, Marzaioli, Sirignano, Altieri, Lubritto, D’Onofrio and Terrasi2010).
There are only a few radiocarbon laboratories with atmospheric 14CO2 data sets several decades long (e.g., Levin and Kromer Reference Levin and Kromer2004; Kuc et al. Reference Kuc, Rozanski, Zimnoch, Necki, Chmura and Jelen2007; Povinec et al. Reference Povinec, Holý, Chudý, Šivo, Sýkora, Ješkovský and Richtáriková2012) and their use for past 14C studies is limited by the length of the record and the sampling location. If past 14C levels outside of the scope of these data sets are required, proxies in other carbon cycle compartments must be used. The biosphere offers an ideal proxy in the form of annual tree rings.
Since the advent of accelerator mass spectrometry in 1977, this analytical technique has become the method of choice for radiocarbon analysis, especially in the case of archaeological samples, where the small sample size presents considerable advantage over the decay counting methods used previously for 14C measurement. However, in the case of environmental samples such as atmospheric 14CO2 samples, the sample size is usually not a limiting factor and radiometric methods are still being used (Krajcar Bronić et al. Reference Krajcar Bronić, Horvatinčić, Barešić and Obelić2009; Levin et al. Reference Levin, Kromer and Hammer2013; Kontuľ et al. Reference Kontul’, Povinec, Šivo and Richtáriková2018).
The radiocarbon laboratory at Comenius University in Bratislava, Slovakia has a long tradition of atmospheric 14C monitoring—first measurements took place in 1967 (Povinec et al. Reference Povinec, Šáro, Chudý and Šeliga1968) and continuous monthly measurements of atmospheric 14CO2 in Bratislava have started in 1984 and continue to the present day. Two data sets are presented in this paper—monthly atmospheric 14CO2 data from the campus of Comenius University in Bratislava, Slovakia, and radiocarbon activity in annual tree ring samples from a tree at the same sampling site. The aim of this study was to determine the past radiocarbon levels in the biosphere of Bratislava, and to compare them with the atmospheric CO2 measurements. This comparison could show to what extent the observed atmospheric 14CO2 variations in an area heavily influenced by fossil CO2 emissions are recorded in a local tree-ring record.
MATERIALS AND METHODS
The studied tree-ring series comes from a black poplar (Populus nigra) tree in the close proximity to the building of the Faculty of Mathematics, Physics and Biophysics of Comenius University in Bratislava, Slovakia (where the monthly atmospheric sampling takes place). Sampling site location on a map of Slovakia can be seen in Figure 1. A whole section from the tree was available, therefore there was enough material for radiometric method of measurement. The tree section was divided into individual growth rings and the samples were further processed and measured at two radiocarbon laboratories—at Faculty of Mathematics, Physics and Informatics at Comenius University in Bratislava, Slovakia, and at Nuclear Physics Institute of Czech Academy of Sciences in Prague, Czech Republic.
Figure 1 Map of Slovakia showing the location of Bratislava, Bohunice Nuclear Power Plant and Jasná.
At the Prague radiocarbon laboratory, the samples were chemically pretreated in a Soxhlet extraction apparatus with a 2:1 benzene-ethanol mixture. After the pretreatment, the sample was combusted in atmosphere of pure oxygen in a quartz apparatus. The prepared CO2 was further purified using a wet process with the AgNO3 solution and dried. Consecutively, benzene was synthesized from the CO2. The procedure involved synthesis of lithium carbide, its hydrolysis to acetylene, purification and catalytic trimerization of the acetylene to benzene form. After one-month storage for radioactive decay of 222Rn, gravimetric dosing of benzene into low background 3 mL Teflon vials was performed with addition of 1.5% (weight) butyl PBD scintillator. The benzene retrieved from the synthesis was measured using a low-background liquid scintillation spectrometer Quantulus 1220 (more details can be found in Svetlik et al. Reference Svetlik, Fejgl, Tomaskova, Turek and Michalek2012). Oxalic acid NIST SRM 4990C was used as a standard for the evaluation of the measurement results.
At the Bratislava radiocarbon laboratory, the chemical pretreatment of tree-ring samples consisted of acid-base-acid treatment with HCl and NaOH followed by NaClO2 bleaching. The bleached wood samples were then dried and subsequently combusted in a stream of pure oxygen. The resulting CO2 was then purified in a glass vacuum line and then used to prepare methane in a reactor with heated ruthenium catalyst with the addition of hydrogen. The samples were stored in glass containers for one month prior to their measurement in a large-volume gas proportional counter (more details can be found in Povinec Reference Povinec1972, Reference Povinec1978). Oxalic acid NIST SRM 4990B was used as a standard for the evaluation of the measurement results.
At both laboratories a small portion of the sample CO2 gas was used for determination of δ13C by isotope-ratio mass spectrometry. The activity of 14C acquired by the measurement is reported in terms of Δ14C following the Stuiver-Polach convention (Stuiver and Polach Reference Stuiver and Polach1977).
The atmospheric 14CO2 data set from Bratislava used for comparison with tree-ring data is a result of radiocarbon analysis of continuous monthly atmospheric CO2 samples absorbed in NaOH solution (Povinec et al. Reference Povinec, Holý, Chudý, Šivo, Sýkora, Ješkovský and Richtáriková2012).
RESULTS AND DISCUSSION
The results of radiocarbon analysis of tree-ring samples from Bratislava are shown in Figure 2 (a table with the results is included in the supplementary materials). This figure also shows the monthly atmospheric data measured in Bratislava (Povinec et al. Reference Povinec, Holý, Chudý, Šivo, Sýkora, Ješkovský and Richtáriková2012) and monthly clean air background values represented by Schauinsland 14CO2 data (Hammer and Levin Reference Hammer and Levin2017). In general, the comparison with clean air background from Schauinsland shows that both the effect of anthropogenic 14C releases and fossil CO2 emissions can be seen in this data. Monthly values above the clean air background indicate residual bomb radiocarbon from the stratosphere and 14C releases from Bohunice nuclear power plant about 50 km northeast from Bratislava (Povinec et al. Reference Povinec, Chudý, Šivo, Šimon, Holý and Richtáriková2009). Atmospheric data in Figure 2 confirm that the influence of fossil CO2 is the dominant anthropogenic effect influencing atmosphere in Bratislava. This is apparent especially in the 1980s and 1990s, when several extremely low Δ14C values (as low as –210‰) have been observed. They were probably caused by certain meteorological conditions changing the prevailing air mass transport modes and preventing efficient ventilation of air masses from the Bratislava area, thus concentrating fossil CO2 and depleting radiocarbon activity in the local atmosphere to this extent (Povinec et al. Reference Povinec, Holý, Chudý, Šivo, Sýkora, Ješkovský and Richtáriková2012). In these conditions, local fossil CO2 sources such as oil refinery on the outskirts of Bratislava (only 8 km southeast from sampling location) and other industrial sources in northwest could have significant impact on radiocarbon levels in the affected area. In the period after 1994 the industrial emissions and pollution in Bratislava decreased due to lower activity of heavy industries as some of them ceased their operation.
Figure 2 14C activity in tree rings from Bratislava compared with monthly atmospheric 14CO2 data from Bratislava (Povinec et al. Reference Povinec, Holý, Chudý, Šivo, Sýkora, Ješkovský and Richtáriková2012) and Schauinsland sampling stations (Hammer and Levin Reference Hammer and Levin2017).
Tree-ring radiocarbon levels are also considerably lower than clean air background in 1980s and 1990s, however, they do not reach the extremely low atmospheric values, nor do they attain the increased values caused by excess 14C observed in the atmosphere. The majority of the observed deep atmospheric minima occurred outside of spring and summer months, therefore outside of the growing season, and these minima should not be recorded in the tree-ring samples. However, several of the extremely low atmospheric values (–120‰ in July 1984 and –210‰ in July 1993) fall into the expected growing season (April–September) and it appears that in this particular case these low 14C activities were not efficiently recorded in the studied tree rings. The changes of radiocarbon activity in tree-rings from Bratislava are relatively smooth, there are no sudden changes present in the data.
For a better comparison (taking into account the expected vegetation period), we can compare the tree-ring results with mean atmospheric values. From the monthly atmospheric data, we calculated average values for the period from April to September, which represent the mean 14C activity during the probable growing period of this tree. The standard deviation of atmospheric mean values is large because of significant variation of monthly Δ14C values in the months for which the average was calculated (Figure 2). This comparison is shown in Figure 3 and it demonstrates that in time period prior to 1994, the atmospheric levels are in general lower than tree ring levels. They are not systematically shifted, there is a large scatter of these atmospheric mean values. This suggests that the studied tree probably did not efficiently incorporate and record the erratic and presumably rapid changes of radiocarbon activity in the atmosphere. Changes of local climatological parameters could have affected photosynthetic assimilation and growth of this tree (Weiwei et al. Reference Weiwei, Xinxiao, Guodong, Hanzhi and Ziqiang2018), but we do not have any information suggesting extreme environmental stress for this tree in the past. After 1994, both the atmospheric and tree-ring values get much closer to clean air levels, and there is much better agreement between these two data sets from Bratislava.
Figure 3 Comparison of Bratislava tree-ring data with mean atmospheric values from Bratislava (Povinec et al. Reference Povinec, Holý, Chudý, Šivo, Sýkora, Ješkovský and Richtáriková2012) and Schauinsland (Hammer and Levin Reference Hammer and Levin2017).
δ13C values measured in tree-ring samples ranged from –26.9 to –24.3‰ (Figure 4). There are no statistically significant trends or discontinuities in the observed δ13C variation and δ13C levels in Bratislava tree rings are in agreement with tree rings from a comparable urban area (Krakow; Rakowski et al. Reference Rakowski, Nadeau, Nakamura, Pazdur, Pawełczyk and Piotrowska2013).
Figure 4 δ13C in the analyzed tree-ring samples.
We have also compared the obtained Bratislava tree-ring data with our regional background tree-ring data set from Jasná sampling location (Figure 5). This site is located in the Low Tatras mountain region in central Slovakia, and it is not influenced by any significant local fossil CO2 emission sources (Kontuľ et al. Reference Kontuľ, Svetlik, Povinec, Brabcová and Molnár2020). This data set is in very good agreement with clean air reference data from Schauinsland. The tree-ring radiocarbon levels in Bratislava are about 250‰ lower than Jasná clean-air tree rings in 1970. This difference slowly decreases over time to virtually zero difference in 1995–1998, indicating a significantly higher input of fossil CO2 into the atmosphere of Bratislava in 1970s and 1980s compared to 1990s and 2000s. Bratislava tree-ring data and background data (Jasná and Schauinsland) can be used to determine the mixing ratio of fossil CO2 in the total CO2 in the sampled atmospheric air (expressed as percentage of fossil-derived CO2 in the total CO2 concentration) based on the method described by Levin et al. (Reference Levin, Kromer, Schmidt and Sartorius2003). The mean mixing ratio of fossil-derived CO2 was highest in 1970s (11.5%) and gradually decreased to 5.0% in 1980s and 1.3% in 1990s.
Figure 5 Comparison of 14C activity measured in Bratislava tree rings with 14C levels in tree rings from regional clean air reference sampling site in Jasná (Kontuľ et al. Reference Kontuľ, Svetlik, Povinec, Brabcová and Molnár2020).
In 1994–1995 there is a small change in the decreasing trend of radiocarbon activity of tree-rings, the activity slightly increased and reached clean air values. A similar change has also been observed in a previously measured tree-ring record from a rural area ca. 30 km northwest from Bratislava (Kontuľ et al. Reference Kontul’, Ješkovský, Kaizer, Šivo, Richtáriková, Povinec, Čech, Steier and Golser2017). This could be caused by a weaker Suess effect, an increase of anthropogenic 14C releases or a combination of these two effects. Anthropogenic radiocarbon from nuclear power plants can indeed be a contributing factor, because given specific meteorological conditions there is a correlation between atmospheric radiocarbon activity at Žlkovce sampling station (close to the Bohunice nuclear power plant) and Bratislava sampling site (Povinec et al. Reference Povinec, Šivo, Šimon, Holý, Chudý, Richtáriková and Morávek2008). But frequency of air mass transport from the direction of closest nuclear power plant is relatively low (several percent) and therefore weaker Suess effect is probably the dominant factor. This is also supported by CO2 emission data for Slovakia showing a decreasing trend from year 1990 onward (MESR 2017).
CONCLUSIONS
Radiocarbon analysis of tree rings from Bratislava was used to extend the radiocarbon record for this sampling site further into the past and compare 14C levels in the biosphere with the available monthly atmospheric data. Both the atmospheric and tree-ring radiocarbon levels are considerably lower than clean air background data. These extremely low values clearly indicate significant influence of fossil CO2 released into the atmosphere. Data shows that this effect is not constant over the studied time period (1970–2004). The difference between clean air reference values and measured radiocarbon content of tree rings was slowly decreasing since the beginning of this tree-ring record. In the period before 1994 the difference is much larger than in the years following 1994, where the difference between Bratislava tree rings and clean air data is minimal. This is in all likelihood caused mainly by decreasing CO2 emissions in the region during 1990s. Anthropogenic 14C produced and released from Bohunice nuclear power plant in western Slovakia can also be a contributing factor in increasing 14C activity to clean air levels, but its influence is much smaller than the effect of reduced fossil CO2 emissions.
Comparison of tree-ring data with atmospheric 14CO2 data shows that the studied tree did not efficiently record the high maxima or deep minima that occurred during the expected growing season. This is probably caused by the rapid nature of these 14C-influencing occurrences—they are intense enough to influence monthly atmospheric data, but they do not have observable effect on a tree that integrates carbon from the atmosphere at a slower rate and during a several-times-longer time period.
Tree rings from the last part of the record (1994–2004) show a good agreement with the mean atmospheric values calculated for the expected growing season. The results indicate that the circumstances influencing radiocarbon levels in the atmosphere and biosphere changed in this period. Not only is the radiocarbon activity in the atmosphere much more stable, but the 14C activity in both atmosphere and biosphere reaches almost clean air levels.
ACKNOWLEDGMENTS
The authors acknowledge support provided by the EU project on Advancing University Capacity and Competence in Research, Development and Innovation (ACCORD) ITMS2014 No. 313021X329, by the VEGA grant No. V-2-0625/00, and by International Atomic Energy Agency’s Technical Cooperation Program (Projects No. SLR1001 and RER7014).
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/RDC.2022.95
