Radioactive and stable cesium isotope distributions and dynamics in Japanese cedar forests

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad

Radioactive and stable cesium isotope distributions and dynamics in Japanese cedar forests Vasyl Yoschenkoa,∗, Tsugiko Takasea, Thomas G. Hintona, Kenji Nanbaa, Yuichi Ondab, Alexei Konopleva, Azusa Gotoa, Aya Yokoyamaa, Koji Keitokua a b

Institute of Environmental Radioactivity of Fukushima University, 1 Kanayagawa, Fukushima, Fukushima Prefecture, 960-1296, Japan Center for Research in Isotopes and Environmental Dynamics, University of Tsukuba, Tsukuba, Ibaraki Prefecture, 305-8572, Japan

A R T I C L E I N F O

A B S T R A C T

Keywords: Fukushima accident Forest ecosystems Radiocesium Stable cesium Radionuclide distribution Radionuclide fluxes

Dynamics of the Fukushima-derived radiocesium and distribution of the natural stable isotope 133Cs in Japanese cedar (Cryptomeria japonica D. Don) forest ecosystems were studied during 2014–2016. For the experimental site in Yamakiya, Fukushima Prefecture, we present the redistribution of radiocesium among ecosystem compartments during the entire observation period, while the results obtained at another two experimental site were used to demonstrate similarity of the main trends in the Japanese forest ecosystems. Our observations at the Yamakiya site revealed significant redistribution of radiocesium between the ecosystem compartments during 2014–2016. During this same period radionuclide inventories in the aboveground tree biomass were relatively stable, however, radiocesium in forest litter decreased from 20 ± 11% of the total deposition in 2014 to 4.6 ± 2.7% in 2016. Radiocesium in the soil profile accumulated in the 5-cm topsoil layers. In 2016, more than 80% of the total radionuclide deposition in the ecosystem resided in the 5-cm topsoil layer. The radiocesium distribution between the aboveground biomass compartments at Yamakiya during 2014–2016 was gradually approaching a quasi-equilibrium distribution with stable cesium. Strong correlations of radioactive and stable cesium isotope concentrations in all compartments of the ecosystem have not been reached yet. However, in some compartments the correlation is already strong. An increase of radiocesium concentrations in young foliage in 2016, compared to 2015, and an increase in 2015–2016 of the 137Cs/133Cs concentration ratio in the biomass compartments with strong correlations indicate an increase in root uptake of radiocesium from the soil profile. Mass balance of the radionuclide inventories, and accounting for radiocesium fluxes in litterfall, throughfall and stemflow, enabled a rough estimate of the annual radiocesium root uptake flux as 2 ± 1% of the total inventory in the ecosystem.

1. Introduction On March 11th, 2011, 14:46 JST, the Great East Japan earthquake of magnitude 9.0, the worldwide fourth largest earthquake recorded in history, occurred off the Tohoku region of Japan. The tsunami that followed severely damaged Units 1–4 of Fukushima Daiichi Nuclear Power Plant (FDNPP). As a result, large amounts of radionuclides were released into the environment (Atomic Energy Society of Japan, 2015). The accident at the FDNPP was the second largest nuclear accident in human history, after the Chernobyl accident (IAEA, 2011). In contrast to the Chernobyl accident, the Fukushima release consisted mainly of volatile radionuclides, and the only long-lived radionuclide released in significant amounts was 137Cs (T1/2 30.1 y). Estimated total releases to the environment are 19–24 PBq (Aoyama et al., 2016). This value is two orders of magnitude higher than the estimated release of the long-lived ∗

90 Sr and 5-7 orders of magnitude higher than the release estimates of the long-lived 129I and isotopes of Pu (Steinhauser et al., 2014). Accordingly, these radionuclides contribute little to the total radioactive deposition onto the Fukushima Prefecture. For example, 129I deposition levels do not exceed some Bq m−2 along the north-western trace of the Fukushima release (Miyake et al., 2012). In general, the area contaminated by the Chernobyl accident is much larger (Steinhauser et al., 2014; Ohta, 2011); however, 137Cs deposition values in the near zones of the two accidents are similar. Impacts to forest systems at Chernobyl and Fukushima were compared in a review paper (Yoschenko et al., 2017b). The Fukushima Prefecture is dominated by forests cover (about 71%; Fukushima Prefecture, 2014). About 343,000 ha of the forests (∼35%) are artificial plantations (MAFF, 2012). The main forestry species is Cryptomeria japonica D. Don (also called Japanese cedar, or

Corresponding author. E-mail address: [email protected] (V. Yoschenko).

http://dx.doi.org/10.1016/j.jenvrad.2017.09.026 Received 1 April 2017; Received in revised form 26 September 2017; Accepted 26 September 2017 0265-931X/ © 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Yoschenko, V., Journal of Environmental Radioactivity (2017), http://dx.doi.org/10.1016/j.jenvrad.2017.09.026

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

Fig. 1. Location of the experimental sites. Background maps: Nuclear Regulation Authority (2013) and http://www.freemap.jp/. Legend: 2013).

Sugi). Japanese cedar contributed 450,000 m3 to the annual roundwood production of 655,000 m3 in 2014 (MAFF, 2014). Extensive decontamination measures are being conducted in agricultural and residential areas of Fukushima's evacuation zones. For some towns and villages the evacuation orders have already been lifted. However, large-scale decontamination of the forests is not planned;

Cs deposition, Bq m−2 (as of March 11,

137

instead, decontamination activities in forests are aimed only to reduce air dose rates and thus are performed in limited areas adjacent to human settlements (Fukushima Prefecture, 2015; JAEA, 2015a,b; IAEA, 2015). In absence of large-scale decontamination, a strategy should be developed for managing the radioactive contaminated Fukushima 2

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

Table 1 Summary of the experimental sites description. Site/Areaa

Coordinates

Elevation/ Slopec,d

137 Cs deposition, MBq m−2

Start of observations

Plantation densitye, ha−1

DBHf, cm

Hg, m

Biomass distribution

Yamakiya/Area 2 Tsushima/Area 3 FDNPP/Area 3

37°35′18.24″N 140°42′38.94″E 37°33′13.10″N 140°47′12.00″E 37°24′55.45″N 141° 1′35.47″E

560–575 m a.s.l./10° 390–400 m a.s.l./10° ∼30 m a.s.l/ 0–35°

0.6–1b 0.7 ± 0.4c 1–3b 1.6 ± 0.5d 30 ± 11d

May 2014

335

35.4 ± 7.9c

24.3 ± 4.5c

August 2015

4430

23.7 ± 11.8d

n/a

Yoschenko et al., 2017a n/a

March 2016

n/a

n/a

n/a

n/a

a

METI (2015). As of June 28, 2012 (MEXT, 2012). As of November 28, 2014 (Yoschenko et al., 2017a). Mean ± STD. d This study. As of September 14, 2016 (Tsushima) and March 24, 2016 (FDNPP). Mean ± STD. e Plantation density of Cryptomeria japonica. f Diameter of the tree at breast height. Mean ± STD. g Height of the tree. Mean ± STD. b c

accurately estimated because the initially intercepted radiocesium was not yet fully removed from the aboveground biomass. The estimates assumed that the radionuclide inventories in the biomass compartments were formed only by root uptake. In this paper, we summarize our observations of radiocesium dynamics in Japanese cedar forests, by comparing its distributions in the aboveground biomass during 2014–2016 with stable cesium distributions. Future studies will include detail characterization of the distributions and chemical forms of radioactive and stable cesium isotopes in the root-inhabited layer of soil, and determination of the root distribution in the soil profile, which will enable us to refine estimates of cesium isotope fluxes in the cedar forests.

forests. For evaluation of the forestry perspectives, a reliable prognosis of the long-term radiocesium redistribution in the forest ecosystems, particularly in the typical artificial forest plantations, is necessary. Numerous studies have quantified key processes governing radiocesium redistribution in Fukushima forests in the early stages after the accident (e.g. Kato et al., 2012a, 2017; Kato and Onda, 2014; Loffredo et al., 2014; Endo et al., 2015 and many others). Recently, Gonze and Calmon (2017) compiled results and derived generic parameters to model the redistribution of radiocesium during the early post-accident stage. In that period, the dominant process was removal of the initially intercepted radiocesium from the tree crowns and trunk surfaces to soil by precipitation and litterfall mechanisms. Similar processes were observed in the early stages after the Chernobyl and Kyshtym accidents (Bunzl et al., 1989; Tikhomirov and Shcheglov, 1994). However, with time, radionuclide activities in the biomass compartments increased due to their root uptake from soil. Eventually, a quasi-equilibrium state was reached, the magnitude depending on the ratio between the two major processes, radionuclide root uptake and its return to soil (Tikhomirov and Shcheglov, 1994). For Fukushima forests, the role of root uptake in redistributing radiocesium in the ecosystem has yet to be clarified. After the plume of release deposited in the forest ecosystem, radionuclides were redistributed among system components by the same processes that redistribute natural stable isotopes of the same elements, or their chemical analogs. In the long-term, when the radionuclides’ initial deposition are fully removed from the aboveground forest biomass, and the radionuclides are distributed in the root inhabited soil layer, the radionuclides reach equilibrium with the natural stable isotope present in the ecosystem (if any). Yoshida et al. (2002, 2004, 2011) showed that 12 years after deposition, the Chernobyl-derived 137Cs reached equilibrium with the natural stable isotope of cesium, 133Cs. Their distributions in the aboveground biomass were similar and their soil-to-plant transfers to the biomass compartments had the same magnitudes. In their study, equilibrium of cesium isotopes in biomass was reached in absence of an equilibrium in the soil profile (i.e., radiocesium concentrations sharply decreased with soil depth, while stable cesium concentrations in the soil profiles were almost constant; Yoshida et al., 2004). The current distributions of stable cesium may thus be useful for predicting the future dynamics of the Fukushima-derived 137Cs in forest ecosystems. The overall aims of our ongoing research are characterization of the radiocesium distribution at the beginning of the late stage after the deposition, quantification of its fluxes and modelling of its long-term redistribution in typical Fukushima forest ecosystems. In our previous paper (Yoschenko et al., 2017a), we made the first estimates of radiocesium fluxes in a typical Japanese cedar forest ecosystem, based on radionuclide distributions measured at the end of 2014. However, in that paper we emphasized that the values of the fluxes could not be

2. Material and methods 2.1. Experimental sites and sampling The study was carried out at two specially equipped sites, Yamakiya and Tsushima, and at a non-equipped site close to the FDNPP (Fig. 1). The Yamakiya site has been described in detail (Yoschenko et al., 2017a). The Tsushima site was equipped similarly to Yamakiya and is an artificial plantation of Cryptomeria japonica in Area 3 (METI, 2015) of the Fukushima zone. The observations near the FDNPP were conducted in a heavily contaminated mixed forest adjacent to the southern border of the power plant. Due to limited access, the FDNPP site could not be equipped for long-term monitoring, and detail characterization of the biomass distribution was not conducted. Description of the sites is presented in Table 1. Air dose rates at the sites at the beginning of observations were 2.6 ± 0.2 μSv h−1 in Yamakiya, 5.2 ± 0.2 μSv h−1 in Tsushima, and varied in the range of n × 10 μSv h−1 near the FDNPP. At the Yamakiya site, the stand density was 335 trees ha−1; however, according to our earlier results (Yoschenko et al., 2017a), even at such a low density, tree biomass in 2014 was the main contributor to both the total biomass (more than 99%) and the total radiocesium activity in the biomass at the site (more than 98%). For this reason, in 2015 and 2016 we did not sample the understory plants at Yamakiya, nor at Tsushima and FDNPP where they were present in much less amounts. The Yamakiya and Tsushima sites were equipped for monitoring radiocesium fluxes. At Yamakiya, the trees were classified according to DBH2, where DBH is the tree diameter at breast height (Yoschenko et al., 2017a). In absence of the height values, H, measured for the whole array of trees and taking into account an almost linear dependence of H on DBH at the site (Yoschenko et al., 2017a), classification was based on DBH2 because the biomass inventories in the Japanese cedar compartments are proportional to H⋅DBH2 (Usoltsev, 2010). Each class consisted of 11 trees. From each class one control tree (with average DBH2 within the class) was selected and all further studies in 3

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

using telescopic pruners and saws, and divided into compartments of young and old foliage, small branches without leaves and big branches. According to our earlier observation (Yoschenko et al., 2017a), radiocesium concentrations in small branches with a diameter of less than 1 cm can be up to an order of magnitude higher than what was measured in big branches, and therefore their contributions to the total activity of the radionuclide in biomass must be calculated separately. Young foliage (foliage of the current year) was identified based on the position of the shoot at the branch, shape and color of leaves. Similar to other researchers (e.g., Coppin et al., 2006) we did not distinguish between the foliage of different years in the old foliage compartment; instead, we determined the average radiocesium concentration in this compartment. For this purpose, after collection of all young foliage from the sampled branch the residue foliage was cut into small pieces and carefully mixed, and then a necessary amount was randomly collected as an old foliage sample. Samples of outer and inner bark were collected at a height of approx. 1.3 m, from areas of about 50 cm2 on the northwestern surface of the trunk (these surfaces were expected to contain less radiocesium activity from the initial deposit). After removal of the bark, core samples of wood were collected using a 12-mmdiameter increment borer (Haglöf, Sweden). Cores were then divided into SW1-3 (last three annual rings of sapwood), SW4+ (rest of sapwood) and HW (heartwood). Similar sampling equipment and strategies were successfully applied by other researchers for characterization of the average TF, SF and LF fluxes in Fukushima forests (Endo et al., 2015; Kato et al., 2017; Coppin et al., 2006). At the FDNPP site, however, the biomass was sampled from two trees only, and the results obtained at this site are used mainly for confirmation of the regularities obtained at other sites.

2014 were performed on the group of 9 trees. At each model tree, a collar-type stemflow (SF) collector (Thimonier, 1998) was installed. Also, 9 collectors of throughfall (TF) and litterfall (LF) were installed at random positions within the area where the control trees were located. TF collectors consisted of 10 L polyethylene tanks and 21-cm-diameter funnels with evaporation suppressors and polyethylene mesh filters to reduce contamination of the samples with fallen leaves, and to prevent clogging of the funnel. Taking into account the low plantation density, the TF collectors were randomly relocated after each sampling, i.e. in periods from 2 weeks to 2 months. 1 m × 1 m LF traps were constructed from metal coated plant stake supports and garden mesh material. In the end of 2014, the control tree from the mean class was cut to measure the actual total weights of each tree compartment and to measure radiocesium concentrations. The measured compartment weights in this tree (Yoschenko et al., 2017a) were used to estimate the weights in the trees of other classes using their dependence on H⋅DBH2 (Usoltsev, 2010). The total inventory of biomass in each compartment per unit area (1 m2) at Yamakiya was calculated as the product of average weight of compartment in the model trees and plantation density expressed in m−2 (Yoschenko et al., 2017a). Annual increments of biomass in each compartment were estimated as the differences between the corresponding biomass inventories for two subsequent years, based on the current values of H and DBH and their predictions for the next year derived from the literature (Usoltsev, 2010). Further observations in 2015 and 2016 continued on the remaining 8 control trees, and 8 samplings each of SF, TF and LF. At Tsushima, the trees were divided according to DBH2 into 5 classes, each consisting of 14 trees, and observations were carried out on 5 control trees; 5 TF collectors and 5 litter traps similar to those used at Yamakiya were installed at random positions within the site. In contrast to Yamakiya, the plantation density at Tsushima site is rather high (4420 trees ha−1; Table 1) and TF collector positions were fixed. Observations at Yamakiya (Yoschenko et al., 2017a) showed that stemflow contribution to the radiocesium removal from the aboveground biomass is negligible, compared to litterfall and throughfall. Similar results were reported by Kato et al. (2017). For this reason, we did not install SF collectors at Tsushima. The SF sampling, however, continued at Yamakiya. At Yamakiya, during 2014, we observed variations in radiocesium concentrations among individual TF and SF samples collected in different sampling campaigns. However, the cumulative radiocesium fluxes with throughfall and stemflow during 2014 increased linearly with cumulative precipitation (R2 = 0.93 and R2 = 0.95, respectively; Yoschenko et al., 2017a). Because of this regularity, the number of measurements of the low radioactive TF and SF samples in 2015 and 2016 were optimized by merging annual samples of throughfall and stemflow. For this purpose, from each individual sample of TF or SF that was collected during the year, we took a volume proportional to the ratio of the volume of water in the sampler at the moment when the sample was collected to the total volume collected during the year by all TF or SF samplers. To characterize radiocesium deposition to the forest litter and soil and to measure its vertical distribution in the soil profile, litter samples were collected from an area of 0.1 m2 each at 5–9 locations, chosen randomly within the experimental sites. Forest litter horizons, L, F and H, were distinguished visually in field conditions according to the degree of decomposition of material, and were sampled separately. In many cases, especially if forest litter was too wet, we were not able to distinguish reliably between the horizons and merged the samples. Soil core samples were collected using DIK-110C liner sampler with asampling area of 19.6 cm2 and sampling depth of 30 cm (Daiki Rika Kogyo Co., Ltd.). The samples were collected before the period of intensive litterfall. The biomass of various tree compartments from control trees at Yamakiya and Tsushima was sampled in late autumn, at the beginning of the dormancy period. The crown biomass samples were collected

2.2. Sample preparation and measurements of cesium isotope concentrations After measurements of weights at the natural moisture content (f.w.), samples of soil, litter and vegetation were dried at 105 °C until constant weight (d.w.). All measurement results are presented for d.w. Litter and biomass samples were ground to 1–2 mm, and all samples were thoroughly homogenized before analyses. Radiocesium activities in the samples were measured by means of γspectrometry using HPGe detectors GC-4020 (Canberra) and Lynx Digital Signal Analyzer. Soil, litter and biomass were measured in 100mL U-8 containers (AS ONE, Tokyo, Japan), and TF and SF in 2-L Marinelli containers. Efficiencies of the spectrometer at 661.6 keV were 1.04% and 1.74% for measurements in 2-L Marinelli and 100-mL U-8 containers, respectively. Calibration error did not exceed 2%. Statistical measurement uncertainties for solid and water samples did not exceed 5% and 12%, respectively. 137 Cs/134Cs activity ratios in all samples, including soil samples from various depths, were characteristic of the Fukushima release at the time of measurement. Therefore, other sources (global fallout, Chernobyl fallout etc.) contributed little to radiocesium contamination of the studied forest ecosystems. Taking into account the short half-life of 134Cs (2.1 years), future levels of radiocesium contamination in Fukushima forests will be mainly determined by 137Cs, thus in the next sections we present only 137Cs activities. For measurements of stable 133Cs concentrations in the biomass samples we applied standard procedures. Dried and ground samples were filtered through a 2-mm mesh sieve and from each sample a 1 g subsample was taken. Subsamples were ashed by heating in a muffle oven at 450 °C for 12 h and put into 100-mL PTFE vessels. The subsamples in the vessels were dissolved with 5 mL of concentrated HNO3, and the solutions were then filtered through 0.45 μm filters (Advantec). The filtered solutions were adjusted to 5% HNO3 by adding Milli-Q water to the 50 mL test tubes. Contents of stable 133Cs in the solutions were measured by ICP-MS using Perkin Elmer ELAN. The calibration 4

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

curve (R2 = 0.999) was built from a set of standard samples using concentrations of 0, 1, 10, 50, 100, 500 and 1000 ppt. The relative standard deviation of the stable cesium concentration measurements was less than 3%.

9 ± 5% in 2014, 11 ± 6% in 2015 and 9 ± 3% in 2016. The changes were not significant (p values between the subsequent years were equal to 0.31 and 0.34, respectively), which means that at Yamakiya radiocesium redistributed mainly between the litter and the 5-cm topsoil layer. Percentages in the deeper soil layer at Tsushima the radiocesium fractions changes beneath the depth of 5 cm were larger; they decreased from 17 ± 13% in 2015 to 12 ± 7% in 2016 (p = 0.25). However, in 2015 we found what appeared to be an outlier in one soil core, with the radiocesium fraction in this layer exceeding 30%. Without the abovementioned core with the high activity of radiocesium in the layer 5–10 cm, the total fraction of radiocesium beneath the depth of 5 cm at the Tsushima site in 2015 would be the same as in 2016, 11 ± 6% (p = 0.46). It is unlikely that cross-contamination of the 5–10 cm sample in that core could lead to such a big increase of the radiocesium activity. One possible explanation is the presence of preferential downward radiocesium migration in that soil coil. For example, studies in Chernobyl (Shestopalov et al., 2003) revealed the heterogeneity of the radionuclides 2-d distributions in the vertical cross-sections of soil due to preferential downward migration. In general, the radiocesium distributions and dynamics were similar in the soil profiles at all our experimental sites. The fractions of radiocesium in litter decreased to 4–8% of the total radiocesium activity in the soil profiles, with the major accumulations occurring in the 5-cm topsoil layers. Significant amounts of radiocesium did not migrate downward from the 5-cm topsoil layer during 2014–2016 at Yamakiya and 2015–2016 at the Tsushima site. Current radiocesium fractions in litter are much lower than those observed in the first years after the deposition (up to 90%; Koarashi et al., 2012). However, similar to that period (e.g. Kato et al., 2012b; Koarashi et al., 2012; Ohno et al., 2012; Matsunaga et al., 2013; Teramage et al., 2014), radiocesium still remains localized in the mineral portion of the 5-cm topsoil layer. Our results are in good agreement with the data by Coppin et al. (2006). They reported that in 2013 the litter and soil within a mature Japanese cedar forest contained 17% and 77% of the total radiocesium activity in the soil, respectively, while the fresh litterfall contributed 5%, and 2% was associated with roots. Dynamics of radiocesium in the fresh litterfall and in the litter layers at Yamakiya are presented in Fig. 3. The concentrations decreased during 2014–2016 with a half-time period of approximately 1.1 years in the litter and approximately 2 years in the fresh litterfall. At Tsushima, during one year, radiocesium in the L layer of litter decreased approximately twofold, from 52 ± 20 kBq kg−1 in 2015, to 27 ± 13 kBq kg−1 in 2016. The decrease in the F + H layer was slower, from

3. Results and discussion 3.1. Radiocesium dynamics in soil and litter 137 Cs activities in soil and litter samples for 2015 and 2016 indicated total depositions to these compartments at Yamakiya of 0.73 ± 0.37 MBq m−2 and 0.78 ± 0.14 MBq m−2, respectively (recalculated to Nov 28, 2014). The values agree with an earlier estimate of 0.72 ± 0.40 MBq m−2, based on date from 2014 (Yoschenko et al., 2017a), and confirms the reliability of the sampling methods. At Tsushima, estimates of the total 137Cs deposition to soil and litter for the two subsequent years, 2015 and 2016, were 1.4 ± 0.8 MBq m−2 and 1.6 ± 0.5 MBq m−2, respectively (recalculated to Nov 20, 2015). At the FDNPP site values of the total 137Cs deposition to soil and litter varied in March 2016 widely, from approximately 21 to 49 MBq m−2; the latter value was found near one of the Japanese cedar trees chosen for biomass sampling. The general dynamics of radiocesium in the soil profiles at Yamakiya and Tsushima were similar (Fig. 2). At Yamakiya, radiocesium activity fractions in litter decreased from 23 ± 9% of the total deposition to soil and litter in 2014, to 8 ± 2% in 2015, and to 4 ± 2% in 2016, while at Tsushima this fraction decreased from 20 ± 11% in 2015, to 8 ± 4% in 2016. In all cases, the differences between the fractions in litter in two subsequent years were significant at t-test p < 0.05. Further observations are necessary to clarify if the radiocesium fractions in litter at the two sites decrease at the same rate (half-time period). As radiocesium levels decreased annually in litter a concomitant increase occurred in topsoil. Radiocesium fractions in the 5-cm topsoil layer at Yamakiya increased from 68 ± 11% in 2014, to 82 ± 7% in 2015 (p < 0.05), and to 86 ± 4% in 2016 (p = 0.13). At Tsushima, the increase of radiocesium in topsoil was 63 ± 14% in 2015, to 80 ± 7% in 2016 (p < 0.05). In the lower soil layers (5–10 cm), the radiocesium inventory remained stable during three years of observations at Yamakiya, equal to

Fig. 3. The radiocesium concentration dynamics in the litter compartments and in the litterfall at the Yamakiya site in 2014–2016. Average values of the radiocesium concentrations and their STD are presented along with their approximations with the exponential dependences. 2014: n = 9 for all compartments; 2015: n = 8 for the litterfall and n = 6 for the litter layers; 2016: n = 8 for the litterfall and n = 5 for the litter layers. The concentration values are decay corrected to the litter sampling date in 2014.

Fig. 2. The radiocesium distributions in soil profiles at the experimental sites in Yamakiya and Tsushima. Average values of the radiocesium fractions in litter and in soil layers and their STD are presented.

5

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

230 ± 30 kBq kg−1 in 2015, to 150 ± 80 kBq kg−1 in 2016, while the concentration in the fresh litterfall increased slightly (p < 0.05) from 18 ± 3 kBq kg−1 to 23 ± 3 kBq kg−1. Note that the above halftime values characterize the current period only; their utilization for assessments of the radiocesium concentrations in the litter layers and in the fresh litterfall in the past or in the future may not be warranted. Indeed, radiocesium concentrations in the litter layers depend on its leaching rates, which may be stable for each litter compartment, and on inputs (i.e. the radiocesium concentrations in the fresh litterfall or in the upper litter horizons) that changed in time since 2011. Assuming that the major part of the fresh litter is deposited to the ground in the late autumn-winter period (Yoschenko et al., 2017a) (i.e. after sampling of soil and litter) the radiocesium decrease in the L layer may be associated with its leaching from the fallen foliage components and branches. At both field sites, radiocesium concentrations in this layer decreased twofold during one year. Interestingly, in 2016 radiocesium concentrations in the litterfall were equal to its concentrations in the L layer of litter at both field sites. Taking into account the slower rate of radiocesium decrease in the litterfall, compared to litter (Yamakiya, Fig. 3), one may expect that in the future the concentrations in fresh litterfall will exceed those in the L layer. Assuming a half-time period of 1 year, full removal of the initially intercepted deposition from the aboveground forest biomass will be achieved when the fresh litterfall to L layer concentration ratio reaches 2.

previous year) and to young foliage. In outer bark, the radiocesium concentration in 2015 decreased almost twofold compared to 2014. The radionuclides from this compartment could be leached to stemflow or could diffuse to inner bark and sapwood and translocated further to heartwood as reported by Ogawa et al. (2016). Chernobyl studies (Shcheglov et al., 2001; Goor and Thiry, 2004; Yoschenko et al., 2006; Thiry et al., 2009) showed that radiocesium, similar to its chemical analog potassium, was translocated from the senescing biomass compartments to the young growing organs of the trees, such as young foliage. Yoshihara et al. (2016) also reported translocation of potassium to young foliage of Japanese cedars and a similar trend occurred for radiocesium in 2011–2015. Thus the observed decrease of the radiocesium concentration in young foliage in 2015 (Fig. 4) may be related to radiocesium decrease in old foliage. The internal translocations of radiocesium between the tree compartments at the Yamakiya site will be discussed in section 3.3. In 2016, we observed increase radiocesium concentrations in the last three sapwood rings and in young foliage at Yamakiya, compared to the concentrations measured in 2015, resulting in young foliage becoming the most contaminated compartment of the trees. Since the concentrations in these compartments are formed by the above-mentioned internal translocations and by root uptake, their increase may be explained by an increase in root uptake. Indeed, the radiocesium concentrations in 2016 did not change in outer bark, other wood compartments, or in old foliage. Over the entire period of observations at Yamakiya, radiocesium concentrations did not significantly change in older sapwood and in heartwood (Fig. 4). The concentrations in heartwood were higher than in sapwood except in the three last annual rings. It should be noted that according to our earlier results (Yoschenko et al., 2017a), radiocesium concentrations are higher only in the last annual ring (current year), while concentrations in the next two rings are close to that of sapwood. In the present study, sapwood was sampled with the increment borer. The samples were small and radiocesium activities in single annual rings could be too low for reliable measurements. For this reason, we measured radiocesium activities as a composite of 3-rings. Therefore, assuming that each sample consisted of three rings of equal mass, the actual concentrations of radiocesium in the last sapwood ring could be higher (perhaps by 1.5 times) than those presented in Fig. 4, and might exceed the concentrations in heartwood. Similar results were reported by Ogawa et al. (2016). They showed that the intensive translocations of radiocesium between the wood compartments in Japanese cedar were completed in 2011–2013, and that these translocations led to accumulation of radiocesium in heartwood. In other tree species heartwood serves as a source for the nutrients recycling to sapwood (including potassium) and as a storage of toxic chemical substances (Bamber, 1987; Taylor et al., 2002). However, Ohta et al. (2014) showed that in contrasts to other species, Japanese cedar can accumulate potassium in heartwood in higher concentrations than sapwood. The high concentration of potassium is considered a possible reason for the phenomenon of blackening in Japanese cedar heartwood (Kubo and Ataka, 1998; Ishiguri et al., 2003). Ogawa et al. (2016) stated that, in addition to diffusion, at the early phase after the deposition there were other mechanisms of radial translocation of radiocesium through sapwood to heartwood. Radiocesium similar to another alkali element–rubidium–which was used in the experiments of Okada et al. (2012), was transported to heartwood via the ray system. This conclusion was supported by analysis of the annual changes of the radial distributions of radiocesium depending on height, close to the tree crown the sapwood thickness decreases and thus radiocesium reaches heartwood earlier than at the lower heights (Ogawa et al., 2016). Similar pattern in the radiocesium height distribution was shown in our earlier study (Yoschenko et al., 2017a). In contrasts to Yamakiya, the highest radiocesium concentrations at Tsushima in December 2015 and at the FDNPP site in March 2016 were found in outer bark (Fig. 5). However, other trends in radiocesium

3.2. Radioactive and stable cesium isotope concentrations in aboveground biomass The 137Cs dynamics in the forest ecosystem compartments at Yamakiya are presented in Fig. 4. The radiocesium distributions significantly changed during the observation period. In 2014, the highest radiocesium activity concentrations were observed in the compartments containing the residues of the initial deposition (i.e., small branches and old foliage). In 2015, the concentrations decreased twofold in small branches and threefold in old foliage, due to radiocesium leaching with precipitation and removal of the old foliage with litterfall. Part of the radiocesium inventory from these compartments is assumed to have translocated to big branches (since the average concentrations in this compartment in 2015 increased as compared to the

Fig. 4. The radiocesium concentration dynamics in the tree compartments at the Yamakiya site in 2014–2016. Average values of the radiocesium concentrations and their STD (vertical bars) are presented along with the indication of the significant differences: • - t-test p < 0.1 between current and previous years,: - t-test p < 0.05 between current and previous years, ∴- t-test p < 0.01 between current and previous years, •• - one-way ANOVA p < 0.1 for three years and ••• - one-way ANOVA p < 0.5 for three years. n = 9 in 2014 and n = 8 in 2015 and 2016. The concentration values are decay corrected to the sampling date in 2014.

6

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

Fig. 5. Comparison of the radiocesium distributions in Japanese cedar at the experimental sites in 2015–2016. Average values of the radiocesium concentrations and their STD (vertical bars) are presented (n = 8 in Yamakiya, n = 5 in Tsushima and n = 2 at the FDNPP).

distribution (e.g., higher concentrations in heartwood compared to sapwood, and higher in young foliage, compared to old foliage) were similar at all sites. 3.3. Radioactive and stable cesium isotope distributions in the ecosystem The future dynamics of radiocesium can be predicted from the current distributions of stable cesium. When the residues of the initial deposition of radiocesium are removed from the aboveground biomass and the radionuclide is distributed in the root-inhabited layer of soil, radiocesium is recycled in the forest ecosystem by root uptake and return fluxes similarly to natural stable cesium. In the long-term, the distributions of the two cesium isotopes should become similar. In the present study, we measured stable cesium concentrations in the aboveground biomass samples collected at Yamakiya in 2014 (the same samples where the radiocesium concentrations presented at Fig. 5 were measured). We assumed that stable cesium is in a state of quasiequilibrium and that its concentrations in the tree compartments did not change significantly in the two year period. Comparisons of the average stable and radioactive cesium isotope concentrations in the tree compartments are presented int Figs. 6–8. For both isotopes, the concentrations are presented in μg kg−1. The concentrations of the two isotopes do not correlate if all compartments are considered (Figs. 6a, 7a and 8a). However, correlations between the isotope concentrations became strong if the compartments with residues of initial radiocesium deposition are not considered (i.e., fresh litterfall, old foliage, outer bark, and small and big branches; Figs. 6b, 7b and 8b). In 2016, the correlation remains strong (R2 = 0.83) even if the old foliage and outer bark compartments are included. Stabilization of the radiocesium concentrations in old foliage and in outer bark in 2015–2016 can be seen in Fig. 4. It means that radiocesium concentrations in old foliage and outer bark gradually approach their equilibrium levels. Five years post-accident, the tree crowns contain very little foliage that was present on the trees in March 2011. Part of that foliage was deposited to the ground surface with litterfall, and another part lost leaves and transformed into small branches. Portions of the initially deposited radiocesium have been washed from the outer bark by stemflow, and other portions have been translocated to other compartments. As equilibrium approaches, the radiocesium fraction available for translocation from outer bark should decrease. The 137Cs/133Cs concentration ratio in the compartments with

Fig. 6. Concentrations of radiocesium vs concentrations of stable cesium in the Japanese cedar forest compartments at the Yamakiya site in 2014. Average values for each compartment and STD are presented (n = 9). Correlations were built for all compartments (a) and excluding the compartments (open circles) containing the residues of the initial deposition of radiocesium (b). FL – fresh litterfall, SB – small branches, OF – old foliage, OB – outer bark, BB – big branches, YF – young foliage, IB – inner bark, HW – heartwood, SW1-3 – 3 last annual rings of sapwood, SW4+ - rest of sapwood.

strong correlations (i.e., last sapwood, rest of sapwood, heartwood, inner bark and young foliage; Figs. 6b, 7b and 8b) decreased from 8⋅10−5 in 2014 to 4⋅10−5 in 2015, and then increased to 6⋅10−5 in 2016. This ratio is formed by competition of the two isotopes, i.e. by the ratio of their amounts available for uptake and translocations. In 2014, the residues of the initially deposited radiocesium in the old foliage compartment were much higher than in 2015 (Fig. 4), and radiocesium from old foliage could be translocated to young foliage (along with potassium necessary to support growth of the new biomass) in larger amounts than in 2015. This explains the overall decrease of the 137 Cs/133Cs concentration ratio between 2014 and 2015. The subsequent increase of the ratio in 2016 might be explained by an increase in the available amount of 137Cs through root uptake. Concentrations of stable cesium in young foliage in our study are close to that reported for Japanese cedar by Nishikiori et al. (2015); however, their study was carried out at a location with much lower radiocesium deposition and the 137Cs/133Cs was lower than in our study. The radiocesium concentrations during the observation period did not significantly change in the wood compartments, and variations of the concentration ratio were caused by changes of the radiocesium concentrations in inner bark and young foliage. As mentioned above, 7

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

Fig. 7. Concentrations of radiocesium vs concentrations of stable cesium in the Japanese cedar forest compartments at the Yamakiya site in 2015. Average values for each compartment and STD are presented (n = 8 for radiocesium and n = 9 for stable cesium). Correlations were built for all compartments (a) and excluding the compartments (open circles) containing the residues of the initial deposition of radiocesium (b). See Fig. 6 for compartments denotation.

Fig. 8. Concentrations of radiocesium vs concentrations of stable cesium in the Japanese cedar forest compartments at the Yamakiya site in 2016. Average values for each compartment and STD are presented (n = 8 for radiocesium and n = 9 for stable cesium). Correlations were built for all compartments (a) and excluding the compartments (open circles) containing the residues of the initial deposition of radiocesium (b). See Fig. 6 for compartments denotation.

the increase of radiocesium concentrations in young foliage in 2016 might reflect an increase in root uptake. The future dynamics of the concentration ratio can be better predicted after the distribution of the root system is characterized in the soil profile (work in progress). From Fig. 8a, we expect a further decrease of radiocesium concentrations in litterfall, old foliage, outer bark and branches, and an increase of the concentration ratio heartwood to sapwood. During the observation period, the major fraction of radiocesium was localized in soil (Fig. 9). The radiocesium inventory in this compartment increased from 0.58 ± 0.34 MBq m−2 in 2014 to 0.68 ± 0.34 MBq m−2 in 2015 and 0.74 ± 0.14 MBq m−2 in 2016. During the same time, the radiocesium inventory in litter sharply decreased from 150 ± 47 kBq m−2 in 2014 to 53 ± 25 kBq m−2 in 2015 and to 37 ± 14 kBq m−2 in 2016. In the aboveground tree biomass, the radiocesium inventory decreased from 47 ± 21 kBq m−2 in 2014 to 33 ± 21 kBq m−2 in 2015, and remained at the same level in 2016 (32 ± 18 kBq m−2). The radiocesium distribution in 2014 was rather close to that reported for the mature Japanese cedar forest in 2013 by Coppin et al. (2006). In their study, the average radiocesium fractions were 66% in soil (including roots) compared to 74% in our study, 19% in litter compared to 20% (in both cases, the fresh litterfall was included), and 14% compared to 6% in the aboveground tree biomass. The radiocesium fractions in the aboveground tree biomass in

Fig. 9. Radiocesium deposition distributions between the ecosystem compartments at the Yamakiya site. Decay corrected to November 2014.

8

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

Fig. 10. Dynamics of the radiocesium inventories in the aboveground biomass in Japanese cedar forests. Solid line presents the dynamics predicted by Kato et al. (2017). Open circle and vertical bars present the mean and STD reported for 2013 by Coppin et al. (2006). Shadowed circles and vertical bars present the means and STD at the Yamakiya site (present study).

our study and in Coppin et al. (2006) are in a good agreement (Fig. 10) with the results by Kato et al. (2017). They described the decrease of radiocesium in a tree canopy inventory of mature Japanese cedar using a two-component exponential dependence, with half-time losses of 87 and 550 days (the fractions of total deposition associated with fast and slow loss components were 0.29 and 0.4, respectively). However, in contrasts to the modelled dynamics, in 2016 we observed a stabilization of the radiocesium inventory in the aboveground biomass. Note that the Kato et al. (2017) study was carried out in 2011–2012, when the dominant process was removal of radiocesium from the aboveground tree biomass, and root uptake did not play an important role in the radiocesium redistribution in the ecosystem. Distributions of the tree compartments biomass and both cesium isotopes at Yamakiya are presented in Fig. 11. The tree trunk compartments (i.e. hartwood, sapwood 4+, sapwood 1–3, inner bark, and outer bark) contributed almost 70% to the total mass, and about 50% of

Fig. 12. Dynamics of the radiocesium inventories in the forest ecosystem compartments, litterfall, throughfall and stemflow at the Yamakiya site. The inventories are calculated based on the average concentrations of radiocesium in each compartment (Fig. 4).

the total stable cesium inventory. The total radiocesium fraction in the tree trunk gradually increased since 2014 reaching about 30% in 2016. For foliage, the total radiocesium fraction sharply decreased in 2015, mainly due to its loss from old foliage, and increased in 2016. By the end of the observation period, the radioactive and stable cesium isotope fractions in old foliage were equal, while in young foliage the radiocesium fraction still remained lower than that of the stable isotope. In general, in 2014 the radioactive and stable cesium isotope distributions at Yamakiya differed significantly. The radioactive cesium distribution in our study in 2014 was closer to that reported for mature Japanese cedar in 2013 by Coppin et al. (2006), with the low fraction in the wood compartments and the largest fraction in the tree trunk. Since 2015, however, clear trends indicate that radiocesium distribution is approaching a quasi-equilibrium (stable cesium) distribution. As mentioned, the total radiocesium inventory in the aboveground tree biomass at Yamakiya decreased between 2014 and 2015 and did not change in 2016 (Figs. 10 and 11). During the whole observation period, the total 137Cs inventory in the tree trunks (wood and bark) remained rather stable, about 10 kBq m−2 (Fig. 12). The radiocesium inventories, however, drastically changed in the components of the tree crowns (young and old foliage, and small and big branches). Between 2014 and 2015, radiocesium inventories sharply decreased in young foliage and small branches, with an especially pronounced decrease in old foliage. Obviously, from the latter compartment, radiocesium could be removed to litterfall and leached with precipitations. However, the total inventory of the radionuclide in young and old foliage, small branches and in litterfall, thoughfall and stemflow in 2015 was lower than its total inventory in foliage and small branches in 2014, which may indicate an important role of internal translocation from small to big branches (Fig. 12). The total radiocesium inventory in the tree crowns, litterfall, thoughfall and stemflow in 2015 was roughly equal to its total inventory in the tree crowns in 2014. Along with the stability of the inventory in the tree trunks, it means that between 2014 and 2015 the contribution of root uptake was very low, while the observed changes in the radiocesium inventories in the tree crown compartments were the result of internal translocations of the radionuclide. Based on the average inventories in all aboveground biomass compartments, the root uptake between 2014 and 2015 could reach approximately 0.3% of

Fig. 11. Distributions of biomass and stable and radioactive cesium isotopes between the forest ecosystem compartments at the Yamakiya site. Vertical axis: % of total in the aboveground biomass.

9

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

radiocesium root uptake flux to be 2 ± 1% of its total inventory in the ecosystem.

the total radiocesium inventory, which corresponded to the lower estimate of root uptake that was derived in our previous work (Yoschenko et al., 2017a) based on the methodology by Goor and Thiry (2004). Comparison of the radiocesium inventories in the aboveground biomass in 2015 and 2016, accounting for radionuclide removal by litterfall, throughfall and precipitation, shows that on average about 2% of the total radiocesium inventory in the ecosystem could be recycled from soil to biomass. Thus, increase of the radiocesium inventory in young foliage in 2016 may be explained by increase of intensity of its uptake from soil through the root system. The distributions of radiocesium in the cedar forest are based on average concentrations of radiocesium in the ecosystem compartments. Taking into account the large variations of the radionuclide concentrations in each compartment, we emphasize that the uncertainty of assessing root uptake flux can be very high. For example, between 2015 and 2016 we estimate root uptake as 2 ± 1% of the total radiocesium inventory in the ecosystem. On the other hand, heterogeneity of radiocesium concentrations within the same compartments of forest stands was reported in several Chernobyl and Fukushima studies (Goor and Thiry, 2004; Yoschenko et al., 2006; Thiry et al., 2009; Coppin et al., 2006), and this heterogeneity impacts accuracy of the flux estimates. A more detail consideration of flux estimates, along with modelling the future redistribution of radiocesium in the Japanese cedar forest ecosystems can be done after quantification of the biomass distributions at other sites, and should be the subject of a separate publication.

Acknowledgements The study was supported by Japan Society for the Promotion of Science (JSPS, grants # 15H00968, # 15K00563 and # 15H04621) and by Fukushima University. References Aoyama, M., Kajino, M., Tanaka, T.Y., Sekiyama, T.T., Tsumune, D., Tsubono, T., Hamajima, Y., Inomata, Y., Gamo, T., 2016. 134Cs and 137Cs in the north pacific ocean derived from the March 2011 TEPCO Fukushima dai-ichi nuclear power plant accident, Japan. Part two: estimation of 134Cs and 137Cs inventories in the north pacific ocean. J. Oceanogr. 72, 67–76. Atomic Energy Society of Japan (Ed.), 2015. The Fukushima Daiichi Nuclear Accident. Final Report of the AESJ Investigation Committee. Springer, Tokyo Heidelberg New York Dordrecht London. http://dx.doi.org/10.1007/978-4-431-55160-7. Bamber, R.K., 1987. Sapwood and Heartwood. Forestry Commission of New South Wales, Tech. Publ. No 2. Bunzl, K., Schimmack, W., Kreutzer, K., Schierl, R., 1989. Interception and retention of Chernobyl-derived 134Cs, 137Cs and 106Ru in a spruce stand. Sci. Tot. Environ. 78, 77–87. http://dx.doi.org/10.1016/0048-9697(89)90023-5. Coppin, F., Hurtevent, P., Loffredo, N., Simonucci, C., Julien, A., Gonze, M.A., Nanba, K., Onda, Y., Thiry, Y., 2006. Radiocaesium partitioning in Japanese cedar forests following the “early” phase of Fukushima fallout redistribution. Sci. Rep. 6, 37618. http://dx.doi.org/10.1038/srep37618. Endo, I., Ohte, N., Iseda, K., Tanoi, K., Hirose, A., Kobayashi, N.I., Murakami, M., Tokuchi, N., Ohashi, M., 2015. Estimation of radioactive 137-cesium transportation by litterfall, stemflow and throughfall in the forests of Fukushima. J. Env. Rad. 149, 176–185. Fukushima Prefecture, 2014. Agriculture, forestry and fishery industries of Fukushima prefecture. http://www.pref.fukushima.lg.jp/site/english/agri-foresty-fish.html (Accessed 29 October 2015). Fukushima Prefecture, 2015. Steps for Revitalization in Fukushima. Gonze, M.-A., Calmon, P., 2017. Meta-analysis of radiocesium contamination data in Japanese forest trees over the period 2011–2013. Sci. Tot. Environ. 601–602, 301–316. https://doi.org/10.1016/j.scitotenv.2017.05.175. Goor, F., Thiry, Y., 2004. Processes, dynamics and modelling of radiocaesium cycling in a chronosequence of Chernobyl-contaminated Scots pine (Pinus sylvestris L.) plantations. Sci. Tot. Environ. 325, 163–180. IAEA, 2011. Fukushima nuclear accident update log. Updates of 12 april 2011. https:// www.iaea.org/newscenter/news/fukushima-nuclear-accident-update-log-15 (Accessed 12 January 2017). IAEA, 2015. The Fukushima Daiichi Accident. Technical Volume 5. Post-accident Recovery. International Atomic Energy Agency, Vienna. Ishiguri, F., Maruyama, S., Takahashi, K., Andoh, M., Yokota, S., Abe, Z., Yoshizawa, N., 2003. Prevention of Sugi (Cryptomeria Japonica D. Don) from turning black by smoke heating. Wood Fiber Sci. 35 (2), 209–216. JAEA, 2015a. JAEA-review 2014-051. Remediation of Contaminated Areas in the Aftermath of the Accident at the Fukushima Daiichi Nuclear Power Station: Overview, Analysis and Lessons Learned. Part 1: a Report on the “Decontamination Pilot Project”. Japan Atomic Energy Agency. JAEA, 2015b. JAEA-review 2014-051. Remediation of Contaminated Areas in the Aftermath of the Accident at the Fukushima Daiichi Nuclear Power Station: Overview, Analysis and Lessons Learned. Part 2: Recent Developments, Supporting R & D and International Discussions. Japan Atomic Energy Agency. Kato, H., Onda, Y., Gomi, T., 2012a. Interception of the Fukushima reactor accidentderived 137Cs, 134Cs and 131I by coniferous forest canopies. Geophys. Res. Lett. 39, L20403. http://dx.doi.org/10.1029/2012GL052928. Kato, H., Onda, Y., Teramage, M., 2012b. Depth distribution of 137Cs, 134Cs, and 131I in soil profile after Fukushima dai-ichi nuclear power plant accident. J. Env. Rad. 111, 59–64. Kato, H., Onda, Y., 2014. Temporal changes in the transfer of accidentally released 137Cs from tree crowns to the forest floor after the Fukushima Daiichi Nuclear Power Plant accident. Prog. Nucl. Sci. Technol. 4, 18–22. Kato, H., Onda, Y., Hisadome, K., Loffredo, N., Kawamori, A., 2017. Temporal changes in radiocesium deposition in various forest stands following the Fukushima Dai-ichi Nuclear Power Plant accident. J. Env. Rad. 166 (3), 449–457. https://doi.org/10. 1016/j.jenvrad.2015.04.016. Koarashi, J., Atarashi-Andoh, M., Matsunaga, T., Sato, T., Nagao, S., Nagai, H., 2012. Factors affecting vertical distribution of Fukushima accident-derived radiocesium in soil under different land-use conditions. Sci. Tot. Environ. 431, 392–401. Kubo, T., Ataka, S., 1998. Blackening of sugi (Cryptomeria japonica D. Don) heartwood in relation to metal content and moisture content. J. Wood Sci. 44, 137–141. http://dx. doi.org/10.1007/BF00526259. Loffredo, N., Onda, Y., Kawamori, A., Kato, H., 2014. Modeling of leachable 137Cs in throughfall and stemflow for Japanese forest canopies after Fukushima Daiichi Nuclear Power Plant accident. Sci. Tot. Environ. 493, 701–707. MAFF, 2012. Forest rate/artificial forest ratio by prefecture. http://www.rinya.maff.go. jp/j/keikaku/genkyou/h24/pdf/shinrin_j_h24.pdf (Accessed 12 January 2017).

4. Conclusion Our observations in the Japanese cedar forest at the Yamakiya site revealed significant redistribution of radiocesium between the ecosystem compartments during 2014–2016. The radionuclide inventory in the aboveground tree biomass was relatively stable, its inventory in forest litter, however, drastically decreased (i.e., from 20 ± 11% of the total deposition in 2014 to 4.6 ± 2.7% in 2016) due to leaching of radiocesium from litter to soil and due to a decrease in radiocesium concentration within the fresh litterfall. In the soil profile, radiocesium accumulated mainly in the 5-cm topsoil layers that contained, in 2016, more than 80% of the total radionuclide deposition in the ecosystem. Similar distributions of radiocesium and regularities of its redistribution were obtained at the experimental site in Tsushima and near the FDNPP. The radiocesium distribution between the aboveground biomass compartments at Yamakiya during 2014–2016 was gradually approaching a quasi-equilibrium distribution with stable cesium. A strong correlation of radioactive and stable cesium isotope concentrations in all ecosystem compartments has not been reached yet because of the residues of the initially deposited radiocesium in the fresh litterfall, small branches (that could be the components of foliage in 2011), big branches and outer bark, while the radiocesium concentration in old foliage has decreased almost to quasi-equilibrium level. However, in some compartments the correlation between radiocesium and stable cesium was strong. Increase of the radiocesium concentration in young foliage, in 2016 as compared to 2015, indicates an increase in root uptake from the soil profile. The same conclusion can be made based on the increase during 2015–2016 of the 137Cs/133Cs concentration ratio in the biomass compartments that were strongly correlated. The future dynamics of the concentration ratio will depend on changes of the radionuclide bioavailability for uptake by plants from the soil profile, due to redistribution of its exchangeable forms between the soil layers. Comparisons of the distributions of radiocesium and stable cesium fractions between the biomass compartments indicate that the radionuclide is gradually approaching a quasi-equilibrium state with stable cesium in the aboveground biomass. A mass balance of the radionuclide inventories, accounting for radiocesium fluxes among litterfall, throughfall and stemflow, permits a rough estimate of the annual 10

Journal of Environmental Radioactivity xxx (xxxx) xxx–xxx

V. Yoschenko et al.

Fukushima nuclear accidents: a review of the environmental impacts. Sci. Tot. Environ. 470–471, 800–817. Taylor, A.M., Gartner, B.L., Morrell, J.J., 2002. Heartwood formation and natural durability – a review. Wood Fiber Sci. 34 (4), 587–611. Teramage, M.T., Onda, Y., Patin, J., Kato, H., Gomi, T., Nam, S., 2014. Vertical distribution of radiocesium in coniferous forest soil after the Fukushima nuclear power plant accident. J. Env. Rad. 137, 37–45. Thimonier, A., 1998. Measurement of atmospheric deposition under forest canopies: some recommendations for equipment and sampling design. Environ. Monit. Assess. 52, 353–387. Thiry, Y., Colle, C., Yoschenko, V., Levchuk, S., Van Hees, M., Hurtevent, P., Kashparov, V., 2009. Impact of Scots pine (Pinus sylvestris L.) plantings on long term 137Cs and 90 Sr recycling from a waste burial site in the Chernobyl Red Forest. J. Env. Rad. 100, 1062–1068. Tikhomirov, F.A., Shcheglov, A.I., 1994. Main investigation results on the forest radioecology in the Kyshtym and Chernobyl accident zones. Sci. Tot. Environ. 157, 45–57. Usoltsev, V.A., 2010. Eurasian Forest Biomass and Primary Production Data. Ural Branch of Russian Academy of Sciences, Yekaterinburg, Russia. Yoschenko, V.I., Kashparov, V.A., Protsak, V.P., Lundin, S.M., Levchuk, S.E., Kadygrib, A.M., Zvarich, S.I., Khomutinin, YuV., Maloshtan, I.M., Lanshin, V.P., Kovtun, M.V., Tschiersch, J., 2006. Resuspension and redistribution of radionuclides during grassland and forest fires in the Chernobyl exclusion zone: part I. Fire experiments. J. Env. Rad. 86 (2), 143–163. Yoschenko, V., Takase, T., Konoplev, A., Nanba, K., Onda, Y., Kivva, S., Zheleznyak, M., Sato, N., Keitoku, K., 2017a. Radiocesium distribution and fluxes in the typical Cryptomeria japonica forest at the late stage after the accident at Fukushima Dai-Ichi Nuclear Power Plant. J. Env. Rad. 166, 45–55. Yoschenko, V., Ohkubo, T., Kashparov, V., 2017b. Radioactive contaminated forests in Fukushima and Chernobyl. J. For. Res. http://dx.doi.org/10.1080/13416979.2017. 1356681. Yoshida, S., Muramatsu, Y., Steiner, M., Belli, M., Pasquale, A., Rafferty, B., Rühm, W., Rantawaara, A., Linkov, I., Dvornik, A., 2002. Stable elements – as a key to predict radionuclide transport in the forest ecosystems. Radioprot. – Colloq. 37, 391–396. Yoshida, S., Muramatsu, Y., Dvornik, A., Zhuchenko, T., Linkov, I., 2004. Equilibrium of radiocesium with stable cesium within the biological cycle of contaminated forest ecosystems. J. Env. Rad. 75, 301–313. Yoshida, S., Watanabe, M., Suzuki, A., 2011. Distribution of radiocesium and stable elements within a pine tree. Radiat. Prot. Dosim. 146, 326–329. Yoshihara, Y., Matsumura, H., Hashida, S-n., Nakaya, K., 2016. Radiocesium contamination in living and dead foliar parts of Japanese cedar during 2011-2015. J. Env. Rad. 164, 291–299.

MAFF, 2014. Annual report on forest and forestry for FY2013. http://www.rinya.maff.go. jp/j/kikaku/hakusyo/25hakusyo/pdf/h25summary.pdf (Accessed 12 January 2017). Matsunaga, T., Koarashi, J., Atarashi-Andoh, M., Nagao, S., Sato, T., Nagai, H., 2013. Comparison of the vertical distributions of Fukushima nuclear accident radiocesium in soil before and after the first rainy season, with physicochemical and mineralogical interpretations. Sci. Tot. Environ. 447, 301–314. METI, 2015. Ministry of Economy, Trade and Industry. Assistance of Residents Affected by the Nuclear Incidents. Area to Which Evacuation Orders Have Been Issued. Miyake, Y., Matsuzaki, H., Fujiwara, T., Saito, T., Yamagata, T., Honda, M., Muramatsu, Y., 2012. Isotopic ratio of radioactive iodine (129I/131I) released from Fukushima Daiichi NPP accident. Geochem. J. 46, 327–333. Nuclear Regulation Authority, 2013. Results of the Airborne Monitoring in the evacuation directed zones. http://radioactivity.nsr.go.jp/en/contents/7000/6936/24/270_ 0513.pdf (Accessed 2017 January 12). Nishikiori, T., Watanabe, M., Koshikawa, M.K., Takamatsu, T., Ishii, Y., Ito, S., Takenaka, A., Watanabe, K., Hayashi, S., 2015. Uptake and translocation of radiocesium in cedar leaves following the Fukushima nuclear accident. Sci. Tot. Environ. 502, 611–616. Ogawa, H., Hirano, Y., Igei, S., Yokota, K., Arai, S., Ito, H., Kumata, A., Yoshida, H., 2016. Changes in the distribution of radiocesium in the wood of Japanese cedar trees from 2011 to 2013. J. Env. Rad. 161, 51–57. Ohno, T., Muramatsu, Y., Miura, Y., Oda, K., Inagawa, N., Ogawa, H., Yamazaki, A., Toyama, C., Sato, M., 2012. Depth profiles of radioactive cesium and iodine released from the Fukushima Daiichi nuclear power plant in different agricultural fields and forests. Geochem. J. 46, 287–295. Ohta, H., 2011. Environmental remediation of contaminated area by the fukushimadaiichi NPP accident. https://www.iaea.org/OurWork/ST/NE/NEFW/WTSNetworks/IDN/idnfiles/IDN_AnnFor2011/Cleanup_activities-OHTA.pdf (Accessed 12 January 2017). Ohta, H., Mahara, Y., Ogawa, H., Kumata, A., Fujinaga, R., Nishi, A., Igarashi, T., 2014. Elucidation of the uptake route of radionuclides in deciduous and coniferous trees using radio cesium and radio potassium in annual tree rings. In: Proceedings of the 15th Workshop of Environmental Radioactivity. KEK, Tsukuba, Japan. March 6-8, 2014, pp. 228–232. Okada, N., Hirakawa, Y., Katayama, Y., 2012. Radial movement of sapwood-injected rubidium into haertwood of Japanese cederin the growing period. J. Wood Sci. 58, 1–8. Shcheglov, A.I., Tsvetnova, O.B., Klyashtorin, A.L., 2001. Biogeochemical Migration of Technogenic Radionuclides in Forest Ecosystems. Nauka, Moscow. Shestopalov, V., Kashparov, V., Ivanov, Y., 2003. Radionuclide migration into the geological environment and biota after the Chernobyl accident. Environ. Sci. Pollut. (1), 39–47. Steinhauser, G., Brandl, A., Johnson, T.E., 2014. Comparison of the Chernobyl and

11