Toward the Estimation of Radiocesium Activity Concentration in Trunks of Coppiced Quercus serrata: Leaf Availability Instead of Felling

Abstract

With the long-term goal of resuming the use of contaminated deciduous broad-leaved forests, which previously provided bed logs for the cultivation of shiitake mushroom (Lentinula edodes), we are developing a method for estimating the radiocesium (¹³⁷Cs) activity concentration in the trunks of coppiced konara oaks (Quercus serrata) by using current-year branch in the dormant stage. In this study, we examined whether leaves could be used instead of current-year branches in the dormant stage. Our assessment revealed that the leaves obtained at the phenologically stable stage (from the end of the flushing stage to prior to the defoliation stage) can be used to estimate the ¹³⁷Cs activity concentration of current-year branches (in the dormant stage) by multiplying the leaf ¹³⁷Cs activity concentration by 0.54 ± 0.03. This result suggests that leaves collected at this stage are suitable for the estimation of ¹³⁷Cs activity concentration in the trunk. Given that the leaves are more easily to sample than the current-year branches, we propose that ¹³⁷Cs activity concentration in the trunk of coppiced konara oak can be estimated more easily with this method using leaves.

13.1 Introduction

Although it has been more than 10 years since the accident at Fukushima Daiichi Nuclear Power Plant (FDNPP), radiocesium (¹³⁷Cs) released from the plant remains in the forest. Previously, trunks of coppiced konara oak (Quercus serrata) growing in the Fukushima area were widely used as bed logs for the cultivation of shiitake mushroom (Lentinula edodes) (Miura 2016). However, the production of bed logs was stopped due to the ¹³⁷Cs contamination of the deciduous broad-leaved forests. The index value of 50 Bq kg⁻¹ has been set for bed logs to prevent the mushrooms from exceeding the food standard limit of 100 Bq kg⁻¹ (Forestry Agency 2012). The possibility exists of resuming the use of contaminated deciduous broad-leaved forests if konara oak trunks with ¹³⁷Cs activity concentration below this threshold can be found. Even in forests with the same degree of contamination, the extent to which konara oaks uptake ¹³⁷Cs from the soil is highly variable (specifically, it may differ 100 times) (Kanasashi et al. 2020). Therefore, felling surveys to measure the ¹³⁷Cs activity concentration in the trunk are not efficient, and developing an alternative method for estimating the ¹³⁷Cs activity concentration in the trunk is necessary.

To achieve this goal, our research group focused on the use of the “current-year branch” in the dormant stage as an indicator of root uptake of ¹³⁷Cs from the soil to tree (Kanasashi et al. 2020). The movement of ¹³⁷Cs in the tree is considered to be stable in the dormant stage. In addition, Kobayashi et al. (2019) confirmed that ¹³⁷Cs activity concentration in the trunk is positively correlated with that of current-year branch during this dormant stage. Based on the above findings, we are developing a method using current-year branch in the dormant stage to estimate the ¹³⁷Cs activity concentration in the trunk without felling.

Here, we assume that the survey of the current-year branches of konara oaks generally takes place during the dormant stage, when tree growth generally stopped. This limited period was one of the issues in developing our alternative method, given that the felling of coppiced konara oaks is generally performed in autumn and winter, and judging the degree of contamination is necessary before felling takes place. Accordingly, we assessed the seasonal stability of ¹³⁷Cs in the current-year branches of coppiced konara oaks (Sakashita et al. 2021). We found that the period for surveying the current-year branches can be approximately doubled and that current-year branches before the felling season can be used for estimating the ¹³⁷Cs activity concentration in the trunk in Miyakoji, Tamura, Fukushima Prefecture. This new finding is expected to be useful in related investigations that require a survey of ¹³⁷Cs activity concentration in the trunk. However, it is not always possible for those in charge of the logging operations to correctly identify the current-year branches. Therefore, instead of current-year branches, another index is required as a proxy measure for the estimation of the ¹³⁷Cs activity concentration in the trunk.

Taking into account these recent developments, our research group focused on the potential of using leaves as data sources. Sakashita et al. (2021) reported not only the seasonal stability of current-year branch ¹³⁷Cs activity concentration but also the relation between ¹³⁷Cs activity concentration in the leaves and the corresponding values in the current-year branches. This relation is expected to be stable from the end of the flushing stage to prior to the defoliation stage. From this result, we newly hypothesized that the ¹³⁷Cs activity concentration in the leaves at this phenological stage was as stable as that of the current-year branches; thus, the leaves could be used to estimate the ¹³⁷Cs activity concentration in the trunk. If this hypothesis is true, estimating the ¹³⁷Cs activity concentration in the trunk more easily will be possible, because the leaves can be sampled more easily than current-year branches. In addition, the damage to the coppiced oaks from leaf sampling is probably small compared to the damage from sampling current-year branches. Because of the recent interest in estimating the ¹³⁷Cs activity concentration in the trunk using current-year branches in the dormant stage, we decided to verify whether ¹³⁷Cs activity concentration of current-year branches in the dormant stage can, in fact, be estimated from sampling the leaves.

Our study took place in Miyakoji, Tamura, Fukushima Prefecture. To test our hypothesis, we determined the period in which the ¹³⁷Cs activity concentration in the leaves of konara oaks in the study area is stable. Second, we examined the relation between the ¹³⁷Cs activity concentration in the current-year branches in the dormant stage and the corresponding values in the leaves obtained at the phenologically stable stage. Finally, we verified whether the ¹³⁷Cs activity concentration in current-year branches in the dormant stage can be actually estimated from the leaves.

13.2 Materials and Methods

13.2.1 Data

We used the ¹³⁷Cs activity concentration data (decay corrected to June 1, 2018) of the current-year branches and leaves (Fig. 13.1a, b) in coppiced konara oak (Q. serrata) reported by Sakashita et al. (2021). Here 1-year branches after the dormant stage were defined as the “current-year branches,” and “dormant stage” was defined as from November to April of the following year based on the analysis of the current-year branch ¹³⁷Cs activity concentration (Sakashita et al. 2021). The data from 2018 to 2020 (sampling intervals: from 10 to 98 days; intensive sampling: from April to July in 2020) were obtained from six study sites in Miyakoji, Tamura, Fukushima Prefecture (Fig. 13.1c). Konara oaks at these sites regenerated after the FDNPP accident and had been coppiced (2011–2015). The data from four study sites (Sites 1–4) were used to evaluate the seasonal stability of leaf ¹³⁷Cs activity concentration and the relation between the current-year branch in the dormant stage and leaf ¹³⁷Cs activity concentration. Based on these relations, data from the remaining two study sites (Sites 5–6) were used to verify whether the current-year branch ¹³⁷Cs activity concentration in the dormant stage could actually be estimated from the ¹³⁷Cs activity concentration in leaves.

Fig. 13.1

(a) Sample photo before dividing. (b) Sample photo after dividing into three parts (leaves, current-year branches, and previous-year branches). (c) Map showing six study sites in Miyakoji, Tamura, Fukushima Prefecture. Darker shades of gray indicate higher ¹³⁷Cs deposition densities (decay corrected to December 28, 2012) reported by Mext (Ministry of Education, Culture, Sports, Science and Technology, Japan) (2013). Local place names are given in parentheses. [This map was modified from Sakashita et al. (2021)]

13.2.2 Analysis

To evaluate seasonal stability of ¹³⁷Cs in leaves at Sites 1–4, the values of ¹³⁷C activity concentration in leaves were normalized to zero mean and one standard deviation. The means and standard deviations were calculated using the data from May 28 to September 15, 2019, which was the only period in which samplings were performed at the same time at all four study sites (Sites 1–4). Then, we assessed the seasonal stability of ¹³⁷Cs activity concentration in leaves by merging 3-year observation data on a horizontal axis of 365 days per year. This horizontal axis begins on May 1, because the leaves of konara oaks in Miyakoji generally start to open (flushing) beginning in May. As an additional note, we assumed in this analysis that no interannual variation was found in the ¹³⁷Cs activity concentration of the leaves.

At four study sites (Sites 1–4), we assessed the relation between ¹³⁷Cs activity concentration in the current-year branch (collected in the dormant stage) and corresponding values in the leaves; we applied robust regression, in which the effects of outliers can be reduced (MathWorks 2021). The regression coefficient (±standard error) was used as the coefficient for estimating the ¹³⁷Cs activity concentration in the current-year branches from the leaves. Using the data from Sites 5–6, we also verified whether ¹³⁷Cs activity concentration in the current-year branches can actually be estimated from the ¹³⁷Cs activity concentration of the leaves (collected in July, August, and October) based on this robust regression.

13.3 Results and Discussion

13.3.1 Seasonal Stability of Leaf ¹³⁷Cs Activity Concentration

At four study sites (Sites 1–4), the normalized ¹³⁷Cs activity concentration (normalized to zero mean and one standard deviation) in leaves indicated that the concentration is highest in early May and then decreases rapidly until early July (Fig. 13.2). This period from May to June (or early July) corresponds to the flushing stage of konara oak in the study area; a similar trend can be seen in the seasonal variation of ¹³⁷Cs activity concentration in current-year branches (Sakashita et al. 2021). Given that the seasonal movement of ¹³⁷Cs in the tree is considered to be most active in this stage, using the leaves in this flushing stage is probably not feasible for estimating the current-year branch ¹³⁷Cs activity concentration (in the dormant stage).

Fig. 13.2

Seasonal variation of normalized leaf ¹³⁷Cs activity concentration at four study sites (Sites 1–4). ¹³⁷Cs activity concentrations are normalized to zero mean and one standard deviation from May 28 to September 15, 2019, and decay was corrected to June 1, 2018. Lines inside the boxes indicate the medians, and the upper and lower boundaries of the boxes represent the 25th and 75th percentiles, respectively. Whiskers extend to show the data distribution within 1.5 × interquartile range. Open circles indicate the outliers, and gray shadings indicate the 95% confidence intervals (CI) of the medians

From late July, after the flushing stage, we found that the normalized ¹³⁷Cs activity concentration in leaves was stable until the beginning of October; the median 95% confidence intervals (CI) overlapped during this period (Fig. 13.2). In addition, plentiful leaves are found in the canopy during this period (Fig. 13.3a–d); thus, obtaining enough leaf samples to measure the ¹³⁷Cs activity concentration is easy. Accordingly, the leaves sampled during this period are considered suitable for estimating the ¹³⁷Cs activity concentration of the current-year branches.

Fig. 13.3

Photos of konara oaks (Q. serrata) at Miyakoji (Tamura, Fukushima Prefecture) on July 25, 2018 (a), August 8, 2019 (b), September 20, 2016 (c), October 3, 2018 (d), November 15, 2018 (e), and December 20, 2019 (f)

The normalized ¹³⁷Cs activity concentrations in leaves from November to February were significantly lower (at 5% significant level) than the corresponding values from late July to early October (Fig. 13.2). This is thought to be due to the translocations of cesium from the leaves to other tree organs before the dormant stage (e.g., Yoshihara et al. 2019; Kenzo et al. 2020). The ¹³⁷Cs activity concentration was also stable from November to February; however, in the study area (Miyakoji, Tamura), most of the leaves in the canopy generally fall after November, after which it is not feasible to obtain a large enough sample of leaves (Fig. 13.3e, f).

13.3.2 ¹³⁷Cs Activity Concentration in Current-Year Branches, Estimated Using Sampled Leaves

At Sites 1–4, where we previously evaluated the seasonal stability of ¹³⁷Cs activity concentration in the leaves of konara oak, we found that the average leaf ¹³⁷Cs activity concentration from late July to October was significantly correlated with that of current-year branches (Fig. 13.4; R² = 0.93, p < 0.001). The regression coefficient ± standard error was 0.54 ± 0.03, and the intercept of the regression line was 8.5; 95% CI of the intercept ranged from −57 to 74. In order to simplify the estimation method, the intercept was set to zero since no significant difference from zero was found.

Fig. 13.4

Relation between average of ¹³⁷Cs activity concentration in leaves from late July to October and average of current-year branch ¹³⁷Cs activity concentration in the dormant stage at Sites 1–4 (cyan circle: Site 1; red circle: Site 2; yellow circle: Site 3; gray circle: Site 4). Error bars represent the standard deviations. No error bar is found on the horizontal axis of Site 4, because the leaves were collected only once from July to October. Black solid line indicates the robust regression line, and black dotted lines show the 95% CI of the regression line

At Sites 5–6, we examined whether ¹³⁷Cs activity concentration in the current-year branches can be estimated solely by multiplying the above regression coefficient (0.54 ± 0.03) by the leaf ¹³⁷Cs activity concentration during late July–October. Our comparisons between estimated and measured ¹³⁷Cs activity concentration in the current-year branches indicated that 99% CI of the robust regression lines overlapped one-to-one correspondence lines on July 25, August 25, and October 3, 2018 (Fig. 13.5). Although the verification data set is relatively small, the leaves obtained on these dates could actually be used to estimate ¹³⁷Cs activity concentration in the current-year branches (from the dormant stage).

Fig. 13.5

Relation of estimated ¹³⁷Cs activity concentration in current-year branches and the measured values during the dormant stage at Sites 5 and 6 (black circles: Site 5; black asterisks: Site 6). Here, the ¹³⁷Cs activity concentration of current-year branches (from the dormant stage) was estimated using leaves sampled in 2018, on July 25 (a), August 25 (b), and October 3 (c). Black dotted lines indicate one-to-one correspondence. Red solid lines are robust regression lines, and red dotted lines show 99% CI of the regression lines

Considering the seasonal stability of leaf ¹³⁷Cs activity concentration values in the above three periods (May–early July [flushing stage], late July–October [stable stage], and November–February [defoliation and dormant stage]) and the ease of sampling, only one period—from late July to the beginning of October—is thought to be suitable to estimate the ¹³⁷Cs activity concentrations of current-year branches in the dormant stage in Miyakoji. This period generally corresponds to the time after the flushing stage and before defoliation. Therefore, if these findings are applied to the areas with local climates different from that in Miyakoji, we suggested that the leaves should be obtained from the time when the leaves almost stop growing to the time before the leaves become senescent (before the leaves begin to change colors).

13.4 Conclusions

In this study, we investigated the seasonal stability of ¹³⁷Cs activity concentration in the leaves of coppiced konara oaks, as well as leaf availability, to estimate the ¹³⁷Cs activity concentration of current-year branches taken in the dormant stage. The seasonal variation in the ¹³⁷Cs activity concentration in leaves and the method for estimating the ¹³⁷Cs activity concentration of current-year branches are summarized in Table 13.1. Our results indicated three distinct stages in seasonal variation of ¹³⁷Cs activity concentration in leaves: (1) flushing stage, (2) stable stage, and (3) defoliation and dormant stage. In the flushing stage (from May to early July), ¹³⁷Cs activity concentration decreased rapidly and was not stable, suggesting that it is not suitable for estimating ¹³⁷Cs activity concentration in current-year branches. In the stable stage (from late July to October), ¹³⁷Cs activity concentration values in leaves were stable. Also, we confirmed that ¹³⁷Cs activity concentration of current-year branches (in the dormant stage) can be estimated by multiplying ¹³⁷Cs activity concentration in leaves obtained in the stable stage by the coefficient 0.54 ± 0.03. We showed that estimating ¹³⁷Cs activity concentration in current-year branches is possible by using the corresponding values in sampled leaves as a proxy measure. In the defoliation and dormant stage (from November to February), although ¹³⁷Cs activity concentration in leaves at this stage was significantly lower than during stable stage, the activity concentration was also stable during this stage. However, we concluded that it is not practical to use leaves in this season to estimate current-year branch ¹³⁷Cs activity concentration, given that few leaves are remaining in the canopy at this stage.

Table 13.1 Summary of seasonal variations in leaf ¹³⁷Cs activity concentration in coppiced konara oaks and availability of leaves for estimating current-year branch ¹³⁷Cs activity concentration in the dormant stage. Seasonal variations of current-year branch ¹³⁷Cs activity concentration and the potential for estimating these values for the dormant stage are also described

In summary, we report that, in the case of konara oaks in Miyakoji, Tamura, Fukushima Prefecture, the¹³⁷Cs activity concentration of current-year branches (in dormant stage) can be estimated by sampling leaves in the phenologically stable stage (from late July to October). This finding suggests that material from leaves can also be used as a method for estimating ¹³⁷Cs activity concentration of the trunk without felling the tree. Because leaves are more easily to sample than the current-year branches, this finding exhibits important and practical implications in the search for the konara oaks that can be used as bed logs for shiitake mushroom cultivation in the contaminated area of Fukushima.