Introduction

Significant radionuclide activities were released into the environment as a consequence of the serious damage to the Fukushima Dai-ichi Nuclear Power Plant (FDNPP) following the earthquake and tsunami that occurred on 11th March 2011. This resulted in a large area in the Fukushima region being highly contaminated due to deposition of radioactive debris. In particular, as a large proportion of the land in Fukushima is covered by forest, it is important to understand the levels and behaviors of the deposited radionuclides in forest systems in support of post-accident decontamination procedures. Accordingly, numerous data on Fukushima-derived radionuclides in the forest environment of Fukushima have been reported recently [1–9]. These studies mainly focused on investigations of local distributions, post-accident uptake and translocation of radiocesium in different forest systems. However, the activities and behaviors of other long-lived radionuclides (i.e. 129I and 14C with half-lives of 1.57 × 107 and 5730 years, respectively) in these systems have been sparsely addressed. In particular, 14C is an important radionuclide for regional radiological assessment because it gives a significant fraction of the effective dose to the general public via the atmosphere-agricultural food-ingestion pathway. However, 14C remains one of Fukushima’s most understudied radionuclides [10]. In addition to natural cosmogenic sources, anthropogenic 14C can be produced by the nuclear reactions 14N(n,p)14C, 17O(n,α)14C and 13C(n,γ)14C in nuclear reactors and nuclear weapons testing. A total of 213 PBq 14C from atmospheric nuclear weapons testing and 44 TBq from the Chernobyl accident in 1986 were estimated to have been released into the environment [11, 12]. In our previous study of 14C in annual rings of a 30-years-old Japanese cedar from Iwaki [13], local fossil fuel combustion resulted in a reduction in the Δ14C values compared to the accepted clean-air environment. The source was identified as originating from increasing traffic on two nearby expressways in the 1990’s. Meanwhile, a small but visible 14C pulse observed in the 2011 tree ring in Iwaki was thought to imply release from the Fukushima accident. Within the context of this background, the current work further investigates the radioactivity levels of long-lived radionuclides, 14C and 129I as well as 134,137Cs, in the leaves of Japanese cedar (Cryptomeria japonica) that were collected from the same site as the tree ring samples in [13]. To our best knowledge this is the first report of 14C and 129I analyses in the post-accident Japanese cedar forest system, which is widely distributed in the Fukushima region.

Experimental

The sampling site is located at Iwaki, Fukushima (37°0.53′N and 140°48.57′E), ~50 km southwest of the FDNPP (Fig. 1). Two branches (living and very recently dead) of the cedar were collected in January 2013. It is difficult and sometimes impossible to identify when a cedar branch fell from a tree. However, as the leaves on the fallen branch remained quite green in colour, this strongly suggests that a short time had elapsed since it fell. Thus, it is considered highly likely that the leaves sprouted before the Fukushima accident and the branch fell during the period of 2011 or 2012. Here we treat them as the pre-fallout-expended leaves (sprouted or developed before the 2011 accident) following the term used in [4]. In contrast, the living leaves, collected from the most recent growth, are catalogued as the post-fallout-expended leaves as they sprouted (or developed) in spring-summer 2012.

Fig. 1
figure 1

a Japanese map. The rectangle stands for the studying area. b Enlarged map of the rectangle in (a) showing sampling site marked with a star

Full size image

For 14C determination, several pieces of leaf were taken from the top, middle and base of the branches. They were then chemically treated by the routine acid–base–acid (A–B–A) procedure. The treated samples were combusted at 850 °C to obtain CO2. The purified CO2 samples were reduced to graphite, and the 14C/13C ratios in the prepared graphite samples were measured using the 5 MV accelerator mass spectrometer (AMS) at the Scottish Universities Environmental Research Centre (SUERC). Aliquots of CO2 were measured for δ13C by conventional isotope ratio mass spectrometry (IRMS) using a VG SIRA 11. A detailed description of the chemical pretreatment and graphitisation, the AMS procedures and data reduction are presented elsewhere [13, 14].

Iodine in the leaf samples was extracted by combustion followed by trapping with a 0.4 M NaOH–0.05 M NaHSO3 solution [15]. An aliquot of the trapping solution was taken for 127I determination by inductively coupled plasma mass spectrometry (ICP-MS) at the Technical University of Denmark. The iodine in the remaining trapping solution was further separated using solvent extraction after the addition of 2 mg of 127I carrier and the separated iodine, in the form of iodide, was precipitated as AgI for 129I/127I ratio measurement conducted by AMS at SUERC [16].

A gamma spectrometer with a coaxial Ge detector (EG&G ORTEC LoAx-51370/20P) was used for 134Cs and 137Cs measurement at SUERC. Details of the measurement procedure and data reduction have been reported elsewhere [17].

Results and discussion

Table 1 lists the analytical results for δ13C and Δ14C in the leaf samples. The δ13C results are expressed as per mil deviations from the Vienna Pee Dee Belemnite (VPDB) standard. The 14C data are reported as Δ14C i.e. per mil deviations from the primary standard [0.7459 times the activity of NBS oxalic acid II (SRM-4990C)]. Table 2 lists the analytical results for 134,137Cs and 127,129I concentrations in this study and some previous literature values. All measured and cited 134,137Cs and 131I activities have been decay-corrected to 11th March 2011 when the FDNPP was shut down.

Table 1 δ13C and 14C results of Japanese cedar leaves from Iwaki, Fukushima
Full size table
Table 2 Cesium and iodine isotopic compositions of Japanese cedar leaves from the Iwaki area, Fukushima*
Full size table

14C activities

The Δ14C values from three different parts of the defoliated sample ranged from 45.8 to 51.0 ‰ with a weighted mean and one standard deviation of 47.8 ± 1.3 ‰, whereas those of the living sample varied from 32.1 to 34.1 ‰ with a weighted mean value of 33.1 ± 1.0 ‰. Obviously, the samples of living leaves have about 15 % less 14C than the defoliated leaves.

The 14C fixed in a tree by photosynthesis is considered to be a proxy for the 14C concentration in the ambient atmosphere during the growth period of the leaves. Among the chemical fractions within the tree, the alpha cellulose of the tree ring is considered the most reliable for determining the 14C incorporated at the time of growth [18]. Plant leaves are often useful for investigating 14C concentrations in the air during the timespan of leaf growth [19], however, in the case of cedar leaves, 14C values need to be integrated over a lifespan of several years. Sakurai et al. [20] showed that the 14C concentrations in leaves reflect the ambient atmospheric 14C concentration during the year that the leaves were sampled, or the year prior. In their study, the 14C concentrations in 3 pine needles were higher than those in paired bamboo leaves by 0.2–0.4 pMC. This was considered to be an indication of the longer lifespan of pine needles compared to that of bamboo leaves.

Figure 2 compares the Δ14C values in leaf samples with those in the global atmosphere represented by a monthly mean of May–August measurements of atmospheric CO2 and tree rings from Northern Hemisphere Zone 2 [21], where May–August represents the approximate growth period of these trees in the Fukushima area. The leaf values are also compared with values for rings from the same tree. The Δ14C values in the post-fallout-expended leaves are lower than the global atmospheric level in the sampling year of 2012 but higher than that in the contemporary tree ring of year 2012 (26.7 ± 1.9 ‰ in early wood and 23.6 ± 2.1 ‰ in late wood) [13]. However, they are comparable with values for the 2010 and 2011 rings [13]. Thus, the relatively higher Δ14C values in 2012 leaves compared to the contemporary tree ring might suggest that the leaves fixed 14C from previous years (i.e. 2011 and even early years) in addition to the current year of 2012. Similar results have been observed in pine needles and bamboo leaves in Yamagata in 2011–2012 [20].

Fig. 2
figure 2

Temporal variations of Δ14C in Iwaki Japanese cedar leaves (this study) and tree rings [13] in comparison with global atmospheric 14C [21]

Full size image

The Δ14C values in the pre-fallout-expended leaves were higher than those in the contemporary global atmosphere. Normally, this might be attributed generally to discharges or releases from the operating reactors or a nuclear accident, as observed in many nuclear sites around the world [11]. However, it might not be the case in this study for the following reasons. As described above, the fallen leaves were most likely defoliated in 2011 or 2012. However, their Δ14C values are significantly higher than those in the tree rings of 2011 or 2012. Instead, they are comparable with those in the pre-2009 rings, within the error margin of the analytical uncertainty (Fig. 2). This fact is compatible with a lifespan of several years for the evergreen’s leaf. Hence, the higher Δ14C values in the pre-fallout-expended leaves, with respect to the contemporary rings of years 2011 or 2012, would suggest that the 14C in the leaves derived largely from the high 14C values in years prior to 2009, which reflects a potential lifespan of 3–4 years for the Japanese cedar leaf.

In summary, our present results fully support that the evergreen leaves may contain 14C fixed from previous years, as observed previously [20, 22]. As a result, care should be taken when 14C in the leaf of an evergreen tree is considered as a proxy for atmospheric 14C with an annual time resolution.

134,137Cs activity concentrations and 134Cs/137Cs activity ratio

134Cs and 137Cs concentrations in the pre-fallout-expended leaves are 3.15 ± 0.27 kBq kg−1 and 3.18 ± 0.05 kBq kg−1, respectively, resulting in a 134Cs/137Cs activity ratio of 0.99 ± 0.08 (Table 2). The post-fallout-expended leaf sample has 134Cs and 137Cs concentrations of 1.29 ± 0.03 and 1.32 ± 0.04 kBq kg−1, respectively, corresponding to a 134Cs/137Cs activity ratio of 0.98 ± 0.04. Radiocesium activities in the pre-fallout-expended leaves are >2 times higher than those in the post-fallout-expended leaves. The 137Cs concentration in our pre-fallout-expended leaf sample is comparable with those in living and fallen leaf samples collected in Iwaki on 18th and 29th March 2011 [23, 24]. Kanasashi et al. [9] made comprehensive measurements of 137Cs in Japanese cedar leaves from pre-2010, 2011 and 2012. Among these, 137Cs analyses in the Iwaki area produced activities of 0.3–4.6 kBq kg−1 in leaves that formed initially in pre-2010, 0.3–2.7 kBq kg−1 in those formed initially in 2011 and 0.1–1.5 kBq kg−1 in 2012. Clearly, the 137Cs concentration in our pre-fallout-expended leaf sample is comparable with their pre-2010 samples, while our post-fallout-expended leaf sample is similar to their 2012 samples.

The 134Cs/137Cs activity ratio is known to be a reliable source identifier. The average 134Cs/137Cs ratio at the time of the Fukushima accident (11th March 2011) was found to be approximately 1 [25]. Thus, the identical 134Cs/137Cs activity ratios in this study are consistent with the initial value observed in worldwide environmental samples, indicating a total dominance of Fukushima accident releases of radiocesium in the samples.

When leaching experiments were conducted on contaminated cedar leaves using deionized water, surfactant and acetone [7], it was found that the reduction in activity for deionized water leaching varied from 0 to 40 %, indicating that up to 40 % of the radiocesium was loosely held on the leaf surfaces. The surfactant and acetone leaching only resulted in further reductions of <10 %. These experiments demonstrated that the remaining radiocesium in the contaminated leaves (both fresh and fallen) was strongly fixed in the leaf tissues, unless the leaf tissues were decomposed. Thus, it can be concluded that radiocesium in the pre-fallout-extended leaves had originated from the direct deposition of Fukushima-derived components. This can be supported by similar 137Cs activities in leaves between this study and those collected in a similar region during the early stage of the accident [23, 24].

The radiocesium activity in 20 woody plant species, including Japanese cedar from Abiko, ~200 km southwest of the FDNPP, was investigated in August 2011 [4]. Among the evergreen coniferous species, the average activity in pre-fallout-expended leaves was 2.4 times higher than that in the post-fallout-expended leaves. Furthermore, a distinct variation in the activity among the evergreen coniferous species could be observed for the post-fallout-expanded leaves but not for the pre-fallout-expanded leaves [4]. Kanasashi et al. [9] reported similar 137Cs concentrations in cedar pollen and flowers from 2012 and concluded that this was due to uptake and translocation of a significant proportion of the intercepted activity. The average 137Cs concentration in leaves laid down from 2012 is ~30 % lower than those laid down prior to the accident but still on the tree. Hence, in comparison with these results, the similar trend observed in this study suggests that the radiocesium in the post-fallout-extended leaves was most likely elevated by the translocation of Fukushima-derived radiocesium from other parts of plant.

However, direct post-accident atmospheric fallout of radiocesium onto the post-fallout-expended leaves cannot be ruled out. The long-term 134Cs and 137Cs activities in aerosol samples within Fukushima City have been monitored [26] and it has been observed that the levels of 134Cs and 137Cs in 2012–2013 were significantly higher than the background levels prior to the accident. For example, the average 137Cs activities are 2 × 10−3 Bq m−3 in 2011, 5.2 × 10−4 Bq m−3 in 2012, and 2.7 × 10−4 Bq m−3 in 2013 [26]. Monitoring was not conducted during the early stages of the accident (between 11th March 2011 and 17th May 2011), therefore, the actual 137Cs activity should be significantly higher than 2 × 10−3 Bq m−3 reported for 2011. As a result, the difference in atmospheric 137Cs between 2011 and 2012 would be more than one order of magnitude, significantly larger than that of contemporary leaf samples. Therefore, the contribution of post-accident atmospheric fallout to the post-fallout-expended leaves should be insignificant.

127,129I concentrations and 129I/127I atomic ratios

The pre-fallout-expended leaf sample has 127I and 129I concentrations of 3.1 μg g−1 and 3.6 × 109 atom g−1, respectively, corresponding to a 129I/127I atomic ratio of 2.5 × 10−7 (Table 2). On the other hand, 127I and 129I concentrations in the post-fallout-expended leaf sample were measured to be 0.4 μg g−1 and 1.6 × 109 atom g−1, respectively, resulting in a 129I/127I atomic ratio of 8.4 × 10−7. The large variation in the 127I concentrations in the two samples is comparable with the range (0.22–2.0 μg g−1) observed in Japanese pine needles [27]. The 129I concentrations are significantly higher than that observed in the background area (<7 × 107 atom g−1) at Ito, Japan [27]. The resulting 129I/127I ratios (2–8 × 10−7) are 1–2 orders of magnitude higher than local soils before the Fukushima accident [28], and other background areas within Japan [27, 29]. Hence, these results show a clearly enhanced 129I level in the Iwaki forest environment after the Fukushima accident.

129I and 127I are considered to have different sources in the environment of the Fukushima area. There is no doubt that 129I in the atmosphere and soils originated from the Fukushima accident releases, whereas 127I has multiple sources. The 127I in soil is accumulated from atmospheric deposition and weathering of rocks, of which deposition from the atmosphere might be a major contribution to land near the sea. However, the origins of the atmospheric 127I might originate from releases from both the land and sea. Because the post-fallout-expended leaves were collected from the same tree as the pre-fallout-expended leaves, the source from the soil is considerable the same. Therefore, the large difference in 127I concentration between these two samples would tend to indicate significant variation (averaged over years of lifespan as shown by 14C results) in the atmospheric contribution. As the sampling site is only 15 km from the sea, variations in the contribution of 127I from the sea, dispersed and deposited on land by onshore winds, would seem the most likely cause.

Like radiocesium, the high 129I activity in the pre-fallout-expended leaves might be attributed to direct foliar uptake from the atmosphere during the Fukushima accident. Indeed, the value is comparable with activities observed in the environment in the Iwaki region by other studies. The 131I activity in plant samples collected on 18th March 2011 from two sites (44.4 and 63.7 km from FDNPP) have previously been reported as 516 and 349 kBq kg−1, respectively [23]. As there is no significant isotopic fractionation of the releases from the FDNPP, including 134Cs/137Cs, 133I/131I, 129I/131I [25, 30], the 129I activity can be calculated from the Fukushima-derived 129I/131I ratio and the measured 131I concentration. Taking a 129I/133I atomic ratio of 16 [16, 30, 31], the corresponding 129I activities can be estimated to be 8.3 × 109 atom g−1 and 5.6 × 109 atom g−1, respectively. Similarly, relatively high 131I activities of 781 and 2603 kBq kg−1 have been found in fresh and fallen leaves collected on 29th March 2011 [24]. Using the 129I/133I ratio of 16 discussed above, we calculate corresponding 129I activities of 1.2 × 1010 and 4.2 × 1010 atom g−1, respectively. The 129I activity observed in the pre-fallout-expended leaves in this work does not largely differ from the reported values if the different species of plant is taken into account. Therefore, like radiocesium, the Fukushima-derived 129I has been deposited on the surfaces of plant leaves and absorbed into the plant tissue through the stomata.

Investigations of iodine in the soil–plant system have indicated that plants can take up iodine from soils [32]. However, the soil-to-plant concentration factor for iodine has been shown to be low due to the strong sorption of iodine by soil components such as organic matter and iron and aluminum oxides, resulting in little translocation from the roots to the aerial parts of the plant. In addition, little iodine in soils has been found to be easily leachable and no correlation between soil and plant iodine has been demonstrated [33]. Therefore, uptake of 129I from the contaminated soils is unlikely to be a major pathway to elevate 129I in the post-fallout-expended leaves, as the Fukushima-derived 129I most likely remained at a relatively high activity inside the tree.

The Fukushima accident caused a significant increase in the 129I concentration in the atmosphere, which has been followed by an exponential decline [16]. The 129I concentration in 2012 was found to be about 2 orders of magnitude lower than those measured in the early stages of the accident. A similar pattern is also found for the atmospheric activities of 134,137Cs in Fukushima City [26]. Therefore, if the 129I in the post-fallout-expended leaves was primarily from the ambient atmosphere during the growth period of the leaves, the activity in the post-fallout-expended leaves would be expected to be lower than that in pre-fallout-expended leaves by similar orders of magnitude as observed in precipitation. Clearly, it is not the case in Iwaki cedar leaf samples in which the 129I activity in the post-fallout-expended leaves differed from the pre-fallout-expended leaves by a factor of two.

Thus, similar to radiocesium, the lower 129I activity in the post-fallout-expended leaves compared to that in the pre-fallout expended leaves is most likely attributed to internal translocation of 129I. However, because there are no available 129I data measured on the plant a more detailed assessment will be required to confirm this.

129I/137Cs activity ratio and fractionation between radiocesium and radioiodine

The calculated 129I/137Cs activity ratios are 1.57 × 10−6 and 1.66 × 10−6 in the pre- and post-fallout-expended leaves, respectively (Table 2). The difference in the activity ratio is small for leaves with different growing periods and consistent with those observed in surface soil from a location of a similar distance to the FDNPP as the leaf samples. For instance, the surface soil samples collected soon after the accident and within a 60 km radius of the FDNPP show 129I/137Cs activity ratios ranging within 0.3–3 × 10−6 [34, 35]. The 131I/137Cs activity ratios in soil and plant samples collected on 18th March 2011 from two sites in Iwaki have been reported [23]. In the first site (44.4 km from FDNPP), the 131I/137Cs ratios in soil and plant were 75 and 99, respectively. In the second site (63.7 km from FDNPP), the 131I/137Cs ratios in soil and plant were 69 and 61, respectively. The corresponding 129I/137Cs activity ratios are 1.55 × 10−6 and 1.68 × 10−6, respectively, for soil samples, and 1.36 × 10−6 and 2.22 × 10−6, respectively, for plant samples (Table 2), determined by taking a 129I/133I atomic ratio of 16 [16, 30, 31]. Therefore, although there are limited numbers of analyses, the similar 129I/137Cs activity ratios observed in the post-fallout-expended and pre-fallout-expended leaves implies that the internal translocation might not have caused significant fractionation between radiocesium and radioiodine. However, the observed 129I/137Cs activity ratios are nearly four times higher than the initial value of the Fukushima-derived component (~4 × 10−7) [30, 36]. This suggests that the elemental fractionation between radiocesium and radioiodine mainly occurred during the transportation of the radioactive plumes, because radiocesium would mainly be in the particle-associated form whereas radioiodine can be in both the aerosol and gaseous form. Even within the aerosol, the attached iodine might become gaseous form. On the other hand, some gaseous radioiodine might become attached to aerosol particles at a later stage.

Conclusions

We report 12,13,14C, 134,137Cs and 127,129I concentrations in pre-fallout-expended and post-fallout-expended Japanese cedar tree leaves collected in Iwaki, Japan. 14C concentrations in the leaves reflect the ambient atmospheric 14C levels during the year the leaves were sampled or defoliated as well as year(s) prior to this. Higher 129I and 134,137Cs concentrations in the pre-fallout-expended leaves suggest direct contamination following the Fukushima accident, whereas the relatively low activities in the post-fallout-expended leaves was most likely caused by translocation from other parts within the tree. In comparison with the initial Fukushima-derived 129I/137Cs ratio, the observed high ratios suggest significant fractionation between radiocesium and radioiodine during processes such as transportation and deposition of radioactive plumes. However, the process of translocation within the tree seem to show insignificant fractionation between radiocesium and radioiodine.