Monitoring of radioactive cesium in wild boars captured inside the difficult-to-return zone in Fukushima Prefecture over a 5-year period

Following the Fukushima Daiichi Nuclear Power Plant accident in 2011, tissue samples from wild boar (Sus scrofa) outside the evacuation zone (difficult-to-return zone, DRZ) tended to show high activity concentrations of cesium-137 (137Cs). Understanding the 137Cs dynamics of wild boar populations inside the DRZ is necessary because they affect 137Cs dynamics and wild boar management in areas outside the DRZ. Since few detailed, long-term studies have been conducted inside the DRZ, we measured 137Cs activity concentrations in 221 wild boar muscle samples obtained from wild boar caught inside the DRZ and surrounding areas over a 5-year period. Our results showed that the 137Cs activity concentration in wild boar from inside the DRZ were higher than those in wild boar outside this zone. No significant difference was observed between muscle and soil 137Cs levels, but significant correlations were observed between muscle 137Cs activity concentrations and body length and body weight in the low-activity-concentration season, but not between all seasons and the high-activity-concentration seasons. It is considered that the size effects observed during the low-activity-concentration season may be due to factors related to metabolism and changes in food habit. This is the first long-term survey of 137Cs in wild boar inside the DRZ.

www.nature.com/scientificreports/ affect radiocaesium intake in wildlife [e.g., 5,7 ]. After the FDNPP accident, numerous studies reported that 137 Cs was present in many wildlife species, and extensive studies have been conducted on 137 Cs dynamics in organisms distributed in and around Fukushima (e.g., insects 8 ; amphibians 9 ; fishes [10][11][12] ; birds 13 ; and mammals [14][15][16][17][18][19][20][21]. Since the FDNPP accident in 2011, the government of Fukushima Prefecture has conducted radioactive monitoring surveys of game meat, such as meat from wild boars (Sus scrofa) and Asian black bears (Ursus thibetanus) 22 . The results have shown that radiocaesium contamination differs between these species, and that wild boars have higher radiocaesium levels 16,22 . However, these studies did not examine the DRZ, which has high levels of radionuclide contamination. A previous study showed that there was a positive correlation between the 137 Cs activity concentration in wild boar muscle and 137 Cs deposition on soil at the sites where the wild boars were captured 16 . 137 Cs deposition on soil in the DRZ was high after the accident and most of the area has not yet been decontaminated. Therefore, the wild boars in the DRZ likely have high levels of 137 Cs contamination. In addition, 137 Cs activity concentration levels of wild boars remain high and constant over the several decades after the Chernobyl nuclear power plant accident (hereafter, Chernobyl accident) 15,23 . According to the longterm monitoring, radiocaesium activity concentration of wild boars observed little to no decline or even a slight increase in activities in some case 15 . In case of the FDNPP accident, the long-term contamination in wild boars is of great concern.
After the FDNPP accident, restrictions were placed on the shipment of foodstuffs throughout Fukushima Prefecture and on the consumption of wild boar meat in parts of the prefecture 22 . Removal of these restrictions is difficult because several wild boar meat samples have been found to contain radionuclide activity concentrations that exceed acceptable levels for consumption (i.e., samples that have total radionuclide ( 134 Cs and 137 Cs) levels that exceed the 100 Bq/kg limit prescribed by Japanese food standards).
Wild boars are an important game species in the region, and they cause serious agricultural damage. In recent years, agricultural damage by wild boars has increased and become more widespread due to the marked increase in the wild boar population, which in turn is attributed to the decrease in human activity in rural areas, the increase in deserted arable land, and the decrease in hunting pressure as hunters age and decrease in number 24,25 . In response to this increase in wild boar numbers, prefectures and municipalities have implemented control measures to capture wild boars and reduce the damage that they cause.
A unique problem for the management of wild boar in Fukushima Prefecture is that the motivation among hunters to hunt wild boar has decreased due to the restrictions imposed on the utilization of radionuclidecontaminated wild boar meat 26 . In addition, there is concern that wild boars in human settlements will transport high activity concentrations of radionuclides beyond the boundaries of the DRZ 26 . Of particular concern in the management of wild boar in the DRZ is that a decrease in human activities, such as agricultural activities, in the zone will result in both range expansion and population increases of wild boars in the region 26 . Consequently, clarifying the 137 Cs activity concentration dynamics in wild boar inside the DRZ is considered necessary because this factor appears to be closely related to 137 Cs dynamics and wild boar management in areas outside the DRZ.
Few studies on 137 Cs activity concentrations in wild boars inside the DRZ and the surrounding areas have been conducted; these studies include the results of 1-year trends in Tomioka town 19 , and a comparison of methods used to examine aggregated transfer factors (T ag : radionuclide activity concentration in muscle/radionuclide deposition in soil, m 2 /kg) 27 . However, no detailed reports based on long-term or extensive sampling have been conducted inside the DRZ.
In this study, we conducted a monitoring survey of 137 Cs activity concentrations in muscle samples collected from wild boar inside the DRZ and surrounding areas (Fig. 1) over a 5-year period.

Results
Comparison of 137 Cs activity concentrations in wild boar muscles collected from wild boars captured inside and outside the DRZ. The 137 Cs activity concentrations for all muscle samples from wild boars captured inside the DRZ exceeded detection limits, with levels greater than 100,000 Bq/ kg (FM) obtained for three samples collected in 2016 (Fig. 2, Table 1). The detection range was from 42 to 132,210 Bq/kg (FM), indicating that there was a large variation among individuals (Fig. 2, Tables 1, 2). Significant correlations were found between captured date and the 137 Cs activity concentration in in all four regions (i.e., inside the DRZ and three regions outside the DRZ (Hamadori, Nakadori, and Aizu)) ( Fig. 2). The 137 Cs activity concentrations obtained from wild boar muscle samples collected inside the DRZ tended to be higher than those collected outside the DRZ (Fig. 2). Effective half-life (T eff ) in each region were: 5.1 year (DRZ), 3.0 years (Hamadori), 4.3 years (Nakadori) and 5.5 years (Aizu). In addition, 137 Cs activity concentrations in meat samples varied significantly by sampling month (Fig. 3, Table 2, Kruskal-Wallis test, df = 11, χ 2 = 35.18, P < 0.001). Specifically, significant differences were observed between February in the high-activity-concentration season and August and September in the low-activity-concentration season (Fig. 3, Table 2, Steel-Dwass' test, P < 0.01).
Relationships among 137 Cs activity concentration in muscle and 137 Cs deposition on soil, body length and body weight. From the results of the regression analysis in all seasons and the high-activityconcentration season, no significant differences were observed among the 137 Cs activity concentration in muscle and 137 Cs deposition on soil, body length and body weight (Figs. 4,5). Also, in the low-activity-concentration season, no significant differences were observed between the 137 Cs activity concentration in muscle and 137 Cs deposition on soil, but significant correlations were observed among the 137 Cs activity concentration in muscle and body length and body weight (Fig. 6). www.nature.com/scientificreports/

Discussion
The 137 Cs activity concentrations in wild boar muscle samples from inside the DRZ were higher than those from outside the DRZ. In addition, a large variation was observed among individuals, with differences of several orders of magnitude recorded (Fig. 2). In our study, the range of T eff was 3.0-5.5 years. T eff of wild boars affected by the Chernobyl accident were reported: 7.8 years at Bavaria 15,28 , 1.7 years at southern Germany under 3 years monitoring 15 , 11.7 years at "alienation zone'' and 92 years at ''periodic control zone'' in the Chernobyl exclusion zone 15 . Our results of T eff were less than physical half-lives and not so different from the past reports in the case of Chernobyl [e.g., case of Bavaria, southern Germany]. But our results are based on only 6 years of monitoring the 137 Cs activity concentration. Since a portion of the cases reported little to no decline or even a slight increase in radiocaesium activity concentrations of wild boars after the Chernobyl accident 15 , we need to continue to focus on long-term fluctuations regarding the FDNPP accident. In our study, the 137 Cs activity concentration  www.nature.com/scientificreports/ in wild boar muscles tended to be low in August and September in the summer season and high in February in the winter season inside the DRZ, indicating that our results corroborate the findings of a previous study 16 . In Tomioka town, which includes areas inside the DRZ, the 137 Cs activity concentration in wild boar muscle was previously reported to vary significantly between different months 19 . No significant differences were observed between the 137 Cs activity concentrations in muscle and 137 Cs deposition on soil inside the DRZ in any of the time periods examined (i.e., all seasons, high-activity-concentration and low-activity-concentration seasons) (Figs. 4, 5, 6). 137 Cs deposition on soil tended to be higher in the DRZ than outside the DRZ; however, decontamination efforts have subsequently been initiated in areas close to and inside the DRZ (e.g., Specified Reconstruction and Revitalization Base). Therefore, the distribution of 137 Cs on soil is highly heterogeneous in the region. This heterogeneity might account for the absence of any relationships between muscle 137 Cs activity concentrations and 137 Cs deposition on soil in this area. However, a previous study also reported that muscle 137 Cs activity concentrations in wild boars was highly variable, even when wild boars were captured in areas with similar 137 Cs soil deposition levels 16 . In addition, it is considered that the evaluation of the radiocaesium contamination level in animal species based on soil deposition (i.e., T ag values) is inappropriate for animal species with large home ranges such as wild boars 27 . Similar findings were reported in moose after the Chernobyl accident, where there was a weak correlation between 137 Cs soil deposition and the 137 Cs activity concentration in muscle 29 . In that study, the authors proposed that the annual fluctuations in Table 2. 137 Cs activity concentration in wild boars muscle of each month in difficult to return zone.  www.nature.com/scientificreports/ the T ag values observed in moose may have been attributed to various ecological factors, such as differences in food selection or habitat use 29 . Seasonal variation of food resources and plants organs (e.g., herbs and roots) have been observed in wild boar diet 30,31 and it seems that food preference/selection occurs at an individual level. In addition, 137 Cs activity concentration in food resources of wild boars varies widely depending on 137 Cs contamination levels in the environment (e.g., 137 Cs deposition on soil), plant diversity and the unique distribution of 137 Cs contamination by plant species (e.g., species-specific levels of 137 Cs contamination found in roots, leaves, berries, etc.). Along with the consumption of food resources the ingestion of soil affects the increase of 137 Cs level in wild boar muscle is unclear 7 . In this study, we considered that the variations observed in the 137 Cs activity concentrations in the muscle of wild boars may be more strongly affected by ecological factors, such as food habits and migration than to 137 Cs soil deposition at capture site. In some fish species, a size effect in which there is a tendency for radiocaesium activity concentration in the body to accumulate as a function of increased body size and/or weight has been reported 11 . The factors affecting this size effect were considered to be related to ontogenetic changes in food habits and/or the physiological ability to retain radiocaesium during growth 11,32 . In our study, a significant positive relationship was observed between muscle 137 Cs activity concentration and both body length and weight of wild boars during the   www.nature.com/scientificreports/ low-activity-concentration season, but not during all seasons and the high-activity-concentration season. Cui et al. 19 found no significant correlation between radiocaesium activity concentration and the weight of the wild boars captured over the entire year in 2019-2020 in Tomioka town. A study on the mobility and home range size of wild boar reported that females with piglets have a smaller home range in the summer season than in the fall and parturitional season (i.e., spring) 33 . In France, the home range of wild boars in the hunting season are larger than in the summer season 34 . It is considered that the size effects observed during the low-activity-concentration season in our study may be due to factors related to metabolism and changes in food habits, because the movements and habitat shifts in wild boars are not as great as they are during the other seasons. However, since the activity, movement and home range of wild boars are influenced by numerous factors, such as food availability, population density and hunting pressure (e.g. 35,36 ), it is important to clarify the relationships between muscle 137 Cs activity concentration and the seasonal mobility of wild boars in areas including the DRZ in future studies. Our study showed that the 5-year trend in the 137 Cs activity concentration in the muscles of wild boar inside the DRZ was higher than that in wild boar from outside this area and confirmed the existence of seasonal variation in these activity concentrations and in the size effect of 137 Cs accumulation in the low-activity-concentration   www.nature.com/scientificreports/ ject in and around the "difficult-to-return zone" in Fukushima (2018-2020)". The capture of wild boar and the collection of muscle samples were conducted by the Japan Wildlife Research Center, which was commissioned to conduct these surveys. For all wild boar samples, the data collected included the GPS coordinates of the capture location, body length, body weight and sex. We used hind-leg muscle samples that were collected in five municipalities (Okuma, Katsurao, Tomioka, Namie, Futaba) from January in 2016 to November in 2020 (Fig. 1).

Measurement of muscle 137 Cs activity concentrations.
The surface and any connective tissue were removed from the muscle block samples that were collected for analysis. Raw muscle samples were either minced, sliced, or freeze-dried for several days and then crushed using a bottle blender (Osaka Chemical Co., Ltd., Osaka, Japan). The samples were then placed in standard U-8 containers (100 ml, ⌀56 mm × 68 mm) and gammaray emitting radionuclides were measured using a germanium semiconductor detector (Canberra GC3018, Meriden, USA). The 137 Cs activity concentration of freeze-dried muscle samples were calculated as fresh mass (hereinafter, "FM"). In addition, we used the 137 Cs activity concentration data for wild boar muscles that were collected from outside the DRZ area as part of a wild animal monitoring survey conducted by the Fukushima prefectural government 22 . The results obtained for the samples that were collected outside the DRZ were divided into three groups based on the region of collection, i.e., Hamadori excluding the DRZ (hereafter, Hamadori), Nakadori and Aizu (Fig. 2); the 137 Cs activity concentrations in these samples were compared against the results obtained for samples collected from wild boar inside the DRZ.
Calculation of 137 Cs deposited on soil. To examine whether there was any correlation between the 137 Cs activity concentrations in wild boar muscles and 137 Cs deposition on soil at the site of wild boar capture (Bq/m 2 ), 137 Cs activity concentrations were extracted from the 137 Cs ground deposition open data map compiled by the Japan Atomic Energy Agency's (JAEA) 5th Airborne Monitoring Survey (JAEA, 2012) 41 using ArcGIS Pro 3.1.6 (https:// pro. arcgis. com/ en/ pro-app/ latest/ get-start ed/ insta ll-and-sign-in-to-arcgis-pro. htm). The 137 Cs deposited on soil at each capture site was then estimated considering the decay rate over the number of days between the soil 137 Cs measurements and the capture date of the wild boar using a physical half-life of 137 Cs.

Statistical analysis.
To confirm the changes in the 137 Cs activity concentration in wild boar muscle samples collected in four regions (i.e., inside the DRZ and three regions outside the DRZ (Hamadori, Nakadori, and Aizu)) over time, we performed a regression analysis of the relationship between 137 Cs activity concentration in muscle and capture date (i.e., number of days to the capture date, with January 1, 1900 as the base date). Then, we excluded data that were below the detection limit [e.g. 42 ]. The effective half-life (T eff ) is one of the indicators of suitable measure to explain the behavior of radionuclides in various ecosystems and to predict future contamination levels 41 . T eff is calculated below 15,43 : Ecological half-life (T eco ) of 137 Cs was estimated for each region by T eco = ln2/λ. Estimates of λ were obtained from the slope of the natural-log regression of 137 Cs activity concentration versus time. T phys was the physical half-life of 137 Cs (30.2 year). Differences in the 137 Cs activity concentration in wild boar muscle for each sampling month were evaluated using the Kruskal-Wallis test. Then, the Steel-Dwass test was performed to evaluate the difference in the mean values for each month. These statistical analyses were performed using R4.0.3 (https:// www.r-proje ct. org/).
We performed the regression analysis based on mixed linear models for wild boar datasets inside the DRZ. We used the 137 Cs activity concentration in wild boar muscle as the response variable, and 137 Cs deposition on soil, body length and body weight as explanatory variables. The capture year was used for a random factor. These analyses were performed using the JMP 13.2.1 software package (SAS, Cary, NC, USA). Because the 137 Cs activity concentration in wild boar muscle exhibits seasonal variations 16 , we divided the data into three seasons, i.e., all seasons, high-activity-concentration season (December, January, February, March) and low-activityconcentration season (July, August, September). For all statistical analysis, we used the 137 Cs activity concentration in muscle and 137 Cs deposition on soil with log 10 transformation.

Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.