1. Introduction
The Fukushima Dai-ichi Nuclear Power Plant (FDNPP) accident was caused by the 11 March 2011 Great East Japan Earthquake and tsunami. Due to the hydrogen explosions at the FDNPP, many types of radionuclides were released into the environment [
1]. The Ministry of Health, Labour, and Welfare (MHLW) in Japan set the provisional regulation values for radionuclide in foods on March 17 as an urgent response [
2,
3,
4,
5]. The provisional regulation values were established as radionuclide concentration based on an effective dose of 5 mSv/y for radioactive cesium. For instance, the value for radioactive cesium in drinking water and milk/dairy products was 200 Bq/kg, and that in vegetables, grains, meat, eggs, and fish was 500 Bq/kg. These values of radioactivity concentration were set including the contribution of radioactive strontium to the radiation dose. In addition, the provisional regulation values were also set for isotopes such as radioactive iodine based on a scenario different from that of radioactive cesium. New standard limits (i.e., the current criteria), which were intended for the existing exposure situation, was established on 1 April 2012 based on the discussion in the Pharmaceutical Affairs and Food Sanitation Council [
6]. The current criteria as radionuclide concentration for radioactive cesium (sum of cesium-134 (Cs-134) and cesium-137 (Cs-137)) was calculated based on 1 mSv/y. The current criteria are listed in
Table 1. The value of 1 mSv/y is consistent with a reference level adopted by the Codex Alimentarius Commission (CAC) based on the International Commission on Radiological Protection (ICRP) publication 82 [
7,
8,
9,
10,
11]. Furthermore, because the new standard limits are aimed at long-term application, radionuclides whose half-lives are longer than one year were selected as the regulation target. In particular, Cs-134 and 137, strontium-90, plutonium-238, 239, 240, and 241, and ruthenium-106 were considered to establish new standard limits. Since the measurement of radionuclides other than radioactive cesium require complicated processing, the representative limits for radioactive cesium were established considering the radiation dose from radionuclides other than radioactive cesium [
12,
13,
14,
15,
16,
17]. In other words, the standard limits for radioactive cesium were calculated based on the estimated concentration ratio of each radionuclide in foods [
12,
17] so that the effective dose from all regulated radionuclides would not exceed 1 mSv/y. Therefore, unless the radioactive cesium in foods exceeds the standard limits (e.g., 100 Bq/kg for general food), the radiation dose from all regulated radionuclides such as strontium-90 does not exceed 1 mSv/y. In fact, the dominant radionuclide in the long term was radioactive cesium, and the contribution of other radionuclides to the radiation dose from foodstuffs was estimated as 12% for age 19 and older when the current limits were derived [
17].
Based on the current criteria, many monitoring tests have been mainly conducted by the local government in 17 prefectures located in the eastern part of Japan in accordance with the guideline for the monitoring inspection of radioactive materials in foods [
17,
18]. Foods containing radioactive materials exceeding the limits are recalled and disposed. Furthermore, the distribution of the food is restricted by type on a prefectural basis (or for smaller areas within a prefecture), if the inspections revealed an increase in the areas where radioactive materials exceed the limits. Thus, rigorous measures are being taken to ensure that foods exceeding the standard limit of radioactive materials are not distributed.
Many monitoring tests have been performed since the provisional regulation values were established in 2011. Approximately 300,000 examinations have been conducted since the current criteria were applied in 2012 [
17,
19]. The accumulated data are likely to reach 3,000,000 in the near future. The monitoring results are collected and released at the MHLW website, etc. [
19,
20].
In this study, to evaluate the radiation dose reduction effect by food regulation, the internal exposure dose due to the ingestion of foods after the FDNPP accident was estimated using the accumulated monitoring results. The regulation effect was evaluated by comparing the radiation dose estimated with the results within the standard limits and that estimated with all the monitoring results, which includes the values exceeding the standard limits.
4. Discussion
Almost a decade has passed since the FDNPP accident. Many radionuclide monitoring tests of foods have been conducted continuously. Since these monitoring tests incur some cost, it is important to verify the effect of food regulation in reducing the radiation dose. In this study, the effects of regulations, such as establishing the standard limits, restricting foods that violate the standards, were estimated using the accumulated monitoring results.
In FY2012, many types of samples, except milk or infant foods, exceeded the standard limits as shown in
Table 3. In FY2016, the radioactivity concentration and the samples exceeding the standard limits were considerably less than in FY2012. This is because of the decay of radionuclides due to its half-life, weathering effect and measures to reduce the radionuclides in food such as feed management, decontamination of soil and wood and potassium fertilization [
35,
36,
37]. On the other hand, high radioactivity concentration was observed in samples of wild animal meat and agricultural products in FY2016. The samples with as high radioactivity in the categories of wild animal meat include the meat of the wild boar, bear, deer, and wild birds, and those in agricultural products include wild vegetable and mushroom. Since it was difficult to manage feeding or cultivation related to these categories, it is believed that these items contained a high concentration of radioactivity even after several years. In the meanwhile, because there are some wild vegetables cultivated under controlled conditions, restrict distribution are being cancelled for those wild vegetables [
38].
In FY2012, the radiation exposure dose with regulation was smaller than that without regulation. Especially, in the high percentiles of radiation dose, the radiation dose with regulation was extremely lower than that without regulation. Therefore, it is thought that the public who would have received a relatively high radiation dose were protected by the adoption of food regulations by the authorities. When provisional regulation values were applied in 2011, the radiation exposure dose was estimated to be 0.139 mSv/y (median) at most from the monitoring results within the provisional regulation values in a similar manner [
31]. In our study, the median of internal exposure dose in FY2012 with regulation was estimated to be 0.0430 mSv/y (throughout Japan). The reduction in the radiation dose was likely caused by the decrease in the amount of radionuclides in foods and the adoption of stringent criteria. In FY2016, at high percentiles, the radiation dose with regulation was slightly smaller than that without regulation, while the radiation dose with or without regulation was similar. The impact of radionuclides in foods was considerably small several years after the accident. The radiation exposure dose with regulation in FY2012 and that with and without regulation in FY2016 were significantly small compared to the reference level of 1 mSv/y, which was adopted as the basis for the current criteria. Hence, the measures of foods regulation in Japan after the accident were effective, and the food safety was ensured with regard to radionuclides.
The common samples with high concentrations of radionuclides (95th and 99th percentiles) in the monitoring results (
Table 4) were mainly other animal meats, other poultries, mushrooms and other fishes. On the other hand, with regard to the intake of radionuclides (
Table 5), common food classification with high radionuclide intake (95th and 99th percentiles) were other beverage, rice, mushroom, teas, other fruits, other fishes, other vegetables, and other animal meats. Other animal meats and other poultries were included as foods with high concentrations of radionuclides (
Table 4). However, with regard to the intake of radionuclides (
Table 5), other animal meats had low ranking and other poultries were not included as high radionuclide intake. Since the food intake of other animal meats and other poultries were small, it was assumed that the intake of radionuclides from these foods was small. Consequently, it is seen again that the internal radiation exposure dose is determined by not only the radioactive concentration in the food, but also the food intake. Furthermore, although the agricultural product in the monitoring results (
Table 3) includes wild vegetable, which often has a high concentration of radionuclides, because the intake volume of wild vegetables was not available, they were classified among other green and yellow vegetables or other vegetables in our study. Therefore, in our dose estimation, the radiation exposure dose arising from consuming wild vegetables might have been underestimated or overestimated. Since wild vegetables are valuable foodstuffs for local residents in Japan [
39], in future work, we would like to consider the intake of wild vegetables. Similarly, in terms of food intake, because an average intake was used in this study, biases of individual dietary habits could not be reflected. Furthermore, food intake for processed food was taken into account only for main cooking ingredient. Radionuclides in foodstuff other than the main cooking ingredient in the processed foods were not reflected in the radiation exposure dose. From this viewpoint, the internal exposure dose might have been underestimated to some extent.
The estimated radiation exposure dose (without regulation) with the monitoring results in FY2012 had an extremely high dose (>10 mSv). As shown in
Table 5, this high value is mainly attributed to the high radionuclide intake via consumption of other beverages. Specifically, these other beverages were powdered beverages made of plant leaves. The powdered beverage is basically intended to be consumed as a dilute solution. However, as mentioned above, powdered beverage is consumed by itself and is consumed in a wide variety of ways, and we did not adjust the concentration according to the manner of consumption. Therefore, the high radiation dose was perhaps overestimated. Furthermore, because detailed information of the tested samples were not provided in the released monitoring results, there is the possibility that the ‘powdered beverage’ was actually dried and cut leaf (brewed tea is consumed). We tried recalculating the radiation dose under the assumption that the extreme high radioactivity concentration powdered beverage was consumed with one-tenth concentration by dilution. The radiation doses of the 99th and 99.9th percentile without regulation in FY2012 throughout Japan were 1.10 and 1.26 mSv/y, respectively. Furthermore, under the assumption that that powdered beverage was consumed with one-fiftieth concentration by dilution, the radiation doses of the 99th and 99.9th percentile without regulation in FY2012 throughout Japan were 0.285 and 0.518 mSv/y, respectively. Thus, the actual radiation dose of the high percentile may have been smaller. On the other hand, even if the dilution is considered, the high percentile radiation dose without regulation is smaller than that with regulation. Therefore, we regard the food regulation after the FDNPP accident as effective.
Our study was based on the assumption that foodstuffs exceeding the standard limits were not distributed because of the rigorous food management system in Japan. However, some foodstuffs exceeding the standard limit may have been distributed in rare cases. Therefore, our research might have underestimated the radiation exposure dose. However, research using actual distribution of foods (‘a market basket study’) regarding the radionuclides in foods was conducted, and the study showed that the radiation exposure dose due to radioactive cesium was considerably smaller than the reference level of 1 mSv/y [
40]. For instance, the radiation dose in 2012 was estimated to be 0.0009~0.0057 mSv/y in market basket study [
41]. These results show that the methodology adopted in this study is reasonable. In the meanwhile, the estimated exposure dose in our study was generally large than that by estimated in the market basket study. This is attributed to large LOD in the monitoring tests. Since the monitoring test needs to be conducted conveniently, a larger LOD, which depends on the measurement time, sample volume, etc., is adopted in the monitoring tests than in the market basket study. This is one of the limitations of our research; hence, the internal exposure dose in our study is regarded as overestimated. However, the purpose of our study was to evaluate the effect of food regulation, and because the radiation dose with and without regulation were estimated under identical conditions, we think that the large LOD is not a major problem. Additionally, in the point of view of the overestimation, because monitoring test mainly targets foods which might include high concentration radioactivity, estimated internal exposure dose may be larger than actual radiation dose. Furthermore, radiation dose was estimated based on the assumption that same foods is consumed over a year. The radiation dose for high percentile might have been overestimated, while total diet study is generally conducted in a similar manner.
While the current standard limits were established taking into account radionuclides other than radioactive cesium, such as strontium-90, the internal exposure dose due to radionuclides other than radioactive cesium was not considered in dose estimation in this study. The influence of radionuclides other than radioactive cesium should be included to correctly estimate the radiation dose. However, the dominant radionuclides were Cs-134 and 137, and the concentration of strontium-90 in foods after FDNPP accident was estimated to be within that before the accident [
42]. Therefore, we think that the other radionuclides did not affect the estimated dose much.
Despite the limitations of this study, the effect of reduction of the radiation dose by food regulation was verified using the monitoring test results. Since the monitoring test is a meaningful measure of food safety, we would like to continue the investigation on the sequential change of the effect of monitoring tests.