Introduction
Agricultural crops were directly contaminated with radioactive fallout due to the TEPCO’s Fukushima Daiichi Nuclear Power Plant (FDNPP) accident that occurred on March 11, 2011. In affected areas, such as Fukushima Prefecture and its surrounding prefectures, activity concentrations of 131I (T1/2=8.02 days) and radiocesium (134Cs: T1/2= 2.065 years+137Cs: T1/2=30.2 years) in some leafy vegetable samples exceeded the provisional limits at that time, 2,000 Bq·kg−1 and 500 Bq·kg−1, respectively (from April 1, 2012, the standard limit for total radiocesium was set to 100 Bq·kg−1). When the radioactivity concentration(s) in a crop sample in a certain municipality exceeded the limit, the municipality was not allowed to sell their crop products. Then, once a week, radioactivity monitoring needed to be carried out to confirm that the activity concentration data obtained in the following continuous 3 weeks were lower than the limit values, which were the required conditions to sell and to eat the products again.
When a new nuclear accident happens, we can use the monitoring data obtained in 2011 to estimate the initial radioactivity levels in leafy vegetables and how fast the radioactivity in the crops would decrease with time. For that purpose, it is better to parameterize the related factors, that is, mass interception fractions (FBs) and weathering half-lives (Tws). Following deposition of radionuclides on vegetables, the initial radionuclide activity concentrations of the vegetables can be estimated by FB, which is the ratio between the initial radioactivity concentration of a radionuclide retained by the edible part of the plant (Bq·kg−1 fresh weight [FW]) and the amount of deposited radionuclide in that area (Bq·m−2). Tw is the time taken for the quantity of radionuclides on and in plants to be reduced by half due to weathering (rain and window), tissue ageing, mass increase, and other processes. Thus, following activity concentrations in the vegetable can be estimated using Tw.
These parameter data were available in the International Atomic Energy Agency (IAEA) Technical Report Series No. 472 (TRS 472) [1]; the values were originally from laboratory experiments and observation results after the Chernobyl accident. Recently published technical document of the IAEA No. 1927 (TECDOC 1927) [2] summarized the FB and Tw observed after the FDNPP accident, but no detailed analysis was reported yet. Therefore, in this study, we have re-collated these values and checked the differences among crop species.
Materials and Methods
Because the radioactivity monitoring of fresh vegetables was not carried out immediately after the FDNPP accident started, it was difficult to obtain the initial activity concentrations in those vegetables. According to the TECDOC 1927 [2], we can back-calculate the initial activity concentrations from continuous monitoring data, which were available from the food monitoring results in March to April 2011 reported by the Ministry of Health, Labour and Welfare (MHLW), Japan [3]. The image of the calculation is shown in Fig. 1.
The exponential decline in radioactivity concentration in the edible part of the crop over time is explained as follows.
where At is activity concentration (Bq·kg−1 FW) of 131I, 137Cs or total radioactive Cs (134Cs+137Cs) at time t (day), A0 is the expected initial activity concentration (Bq·kg−1 FW), and λeff is the radionuclide loss rate constant. Using the concentration data, a best fit exponential line for each leafy vegetable sample set collected in a certain municipality was calculated using KaleidaGraph software version 4.5 (Synergy Software, Reading, PA, USA).
Using A0 and total deposition amounts in the area As (Bq·m−2), mass interception fraction FB is calculated as follows.
For radiocesium, As was calculated using the second soil monitoring results among nine sampling campaigns carried out in 2011–2016 by in situ Ge gamma-ray spectrometry, which was conducted for 134Cs and 137Cs (decay corrected to March 1, 2012). Details of the soil data are found in our report [4]. For 131I, we used activity concentrations of measured 131I in soil and estimated 131I from 129I in soil carried out by the Ministry of Education, Culture, Sports, Science and Technology and the Nuclear Regulation Authority, Japan [5]. All the data for 131I, 134Cs, and 137Cs were corrected to March 20, 2011 because the heaviest radioactivity deposition was observed in large areas in eastern Japan on March 20–22 [6]. Corresponding soil data were based on data in each municipality in Fukushima Prefecture and the geometric mean value was calculated to obtain As values. For other areas, if there were no corresponding data for 131I and radiocesium in a municipality, then the data in each prefecture’s deposition monitoring site were applied.
To calculate Tw, effective half-life (Teff, in day) needs to be calculated first using λeff as follows.
The Teff is expressed as follows.
where Tp is the physical half-life of the radionuclide. Using these equations, Tw is then calculated as follows.
We checked the correlation factor between time (day) and the logarithm of the activity concentrations by Student t-test to accept the data set for FB and Tw calculations or not. Firstly based on p-values, we considered that if p<0.05 and the coefficient of determination (R2) value was higher than 0.6, then the data set was used for Teff calculation. Next, if R2 was higher than 0.9, then we included the data set to calculate Teff even if p-values were more than 0.05.
Results and Discussion
After examination of the food monitoring results by MHLW [3], leafy vegetable data obtained in Fukushima Prefecture and surrounding five prefectures (Chiba, Gunma, Ibaraki, Saitama, and Tochigi) were found. Numbers of data sets used for decline fittings by Equation (1) were 15 for radiocesium (eight datasets were from Fukushima Prefecture), and 42 for 131I (19 datasets were from Fukushima Prefecture). Crops were celery, crown daisy, Japanese honeywort, kakina, komatsuna, kukitachina, mizuna, parsley, shinobufuyuna, shiso, and spinach. For comparison, we included heads of cabbage and flowering shoots of Brassica species (e.g., broccoli). Additional data from our experiments for giant butterbur [6] were also included in the analysis. Summary of the data analysis carried out using JMP 9.0 (SAS Institute Inc., Cary, NC, USA) is shown in Supplementary Table S1.
1. Mass Interception Fractions of 131I and Radiocesium
By applying Equation (1), we successfully obtained A0 values for crops mentioned above with good fitting results. However, because some areas were difficult to find corresponding total deposition data, As, thus, the numbers of FB values were smaller compared to those listed above. In addition to the FB calculation using openly available data sets, because we measured edible wild plants on March 28, 2011 and deposition samples [7], the data were also included for analysis.
FB values should be affected by the plant shape (e.g., cabbage, spinach) and the cultivation conditions, i.e., outside and indoor (greenhouse and tunnel) cultivations, thus FB data were summarized in the leafy vegetables and other shape green vegetables (head and flowering shoots), and also classified into outdoor and indoor cultivations, if possible. The FB data are summarized in Table 1; because we lacked knowledge about the distribution characteristics of the data for all groups, arithmetic mean (AM), standard deviation (SD), geometric mean (GM), and geometric standard deviation (GSD) are provided with ranges of the FB data. Using the sample groups having the largest n for 131I and radiocesium were examined for their data distribution. Based on the Shapiro-Wilk normality test result, the FB of leafy vegetables cultivated under outdoor conditions did not follow a normal distribution for each radionuclide. When the logarithms of the FB data were calculated, the data agreed to a log-normal distribution for each radionuclide. We assumed that log-normal distributions were suitable for other groups, thus, data comparisons were carried out using the logarithm of the FB data.
From the results for FB of Cs, leafy vegetables cultivated under outdoor conditions tended to show higher values (GM= 0.13 m2·kg−1 FW) than that for other green vegetables (GM= 0.07 m2·kg−1 FW) but the numbers of data were too small for statistical analysis. For FB of 131I, the logarithm of the data for leafy vegetables cultivated under outdoor conditions (GM= 0.057 m2·kg−1 FW) was significantly higher than those from indoor cultivation (GM=0.015 m2·kg−1 FW) by Wilcoxon rank sum test (p=0.018). However, when we used the data only for spinach, no difference was found between these different cultivation conditions. When the data of leafy vegetables were compared to those of other green vegetables both grown under outdoor conditions, a significant difference was observed by the same test (p<0.001). From these results, we assumed that crop type (shape) and/or species would affect the FB value rather than cultivation conditions.
For leafy vegetables grown under outdoor conditions, the FB data for radiocesium were significantly higher than those for 131I by Wilcoxon rank sum test (p=0.023). According to IAEA TRS 472 [5], FB values for grass observed after the Chernobyl accident were 1.1 m2·kg−1 dry weight for Cs and 0.7 m2·kg−1 dry weight for I, thus FB of Cs was slightly higher. Our present results showed a similar tendency; however, more specifically, we compared the FB values for spinach. The result showed no difference between 131I and radiocesium by Wilcoxon rank sum test (p>0.05). From these results, although the chemical forms of 131I and radiocesium were different, FB values were not much different. Yet, further studies are needed to clarify the chemical form effect on FB values by plants.
Values in IAEA TRS 472 were collated on a dry weight basis; thus, it was difficult to compare the data directly with our data. Fresh mass to dry mass conversion factors are needed for this comparison; according to the TRS 472, dry matter contents for grass species (e.g., Timothy grass, Standard crested grass, Meadow fescue) were 20%–39%. When the median value, ca. 30% was used to calculate FW basis FB for grass in IAEA TRS 472, the calculated values were 0.33 m2·kg−1 FW for Cs, and for 0.21 m2·kg−1 FW for I. The GM of FB values for radiocesium and 131I in the present study was about one-third of the TRS 472 values, respectively, probably because only edible parts were considered in this study.
2. Weathering Half-lives of 131I and Radiocesium
Tw data were calculated using Equations (3)–(5), and the summaries of the results are shown in Table 2. Similar to the FB data analysis, using the sample group having the largest n was examined for its data distribution, and we found that the data distributed log-normally. Thus for Tw data, we also assumed that all data would distribute log-normally thus logarithm of the Tw data were used for analysis.
Tw values of 131I in leafy vegetables obtained under outdoor cultivation conditions ranged from 3.0–23 days (GM=8.4 days), and that for radiocesium ranged 3.1–8.9 days (GM= 6.2 days). There were no significant differences between Tw values of these radionuclides by t-test. In our recent publication, we carefully examined the fate of 131I and radiocesium on plants, and we concluded that these similar Tw values between 131I and radiocesium may be affected by the crop mass increase (dilution effect) rather than decrease by weathering (e.g., wash-out with rain) [6, 8].
For Tw values of Cs, leafy vegetables cultivated under outdoor and indoor conditions were different by Wilcoxon rank sum test (p<0.01). Because the GM of Tw of vegetables cultivated under indoor conditions was longer, it was expected that washing effect by rain caused this difference; however, the crop types were also different in these datasets, thus the analysis results may also be affected by the crop types/species as we found for FB of Cs.
Further comparison was carried using Tw values for 131I; however, no difference was observed between Tws for outdoor (GM=8.4 days) and those for indoor (GM=9.9 days) cultivation conditions by Wilcoxon rank sum test. More specific data comparison was carried out using spinach Tw data of 131I. Because the number of data was small for indoor cultivation (n=4), the result was not reliable enough, but we did not find a difference between the conditions by Wilcoxon rank sum test. Moreover, we compared the Tw of 131I data between leafy vegetables and other green vegetables grown under outdoor conditions, but we did not find a significant difference between these vegetable types as well.
From these results, it was difficult to explain why Tw of Cs between crops grown under outdoor and indoor conditions was different, and vice versa, it was not clear why Tw of 131I between crops grown under outdoor and indoor conditions and different crop types was not different. Further studies are needed to clarify the weathering effect for these radionuclides.
Conclusion
In the present study, we reported FB and Tw values of 131I and radiocesium using openly available monitoring data obtained after the FDNPP accident. The purpose of this study was to provide environmental transfer parameter values to estimate activity concentrations of these radionuclides in agricultural crops when direct contamination occurred. The study results showed that there was no difference between 131I and radiocesium for both parameter values. The results showed that the FB values may be affected by crop types (shapes) and/or species but the difference was not clear for Tws. By using these values presented in this study, we would be able to carry out a rough estimation of the activity concentrations of 131I and radiocesium in the edible part of leafy crops when these radionuclides were released into the atmosphere. However, because the effect of mass dilution by crop growth may lead to overestimation or underestimation, sufficient attention should be paid to apply these factors.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Supplement
Print 
