Introduction

Following the 2011 accident at the Tokyo Electric Power Company Fukushima Daiichi Nuclear Power Plant (FDNPP), residents of surrounding municipalities were forced to evacuate for prolonged periods [1]. Even nearly 15 years after the accident, many residents of Okuma Town, Fukushima Prefecture, continue to live outside the town. Okuma Town is the host municipality of the FDNPP, and a full evacuation order was issued immediately after the accident [1].

Since then, evacuation orders have been lifted in a stepwise manner. In 2019, evacuation orders were lifted in parts of the mountainous areas of the town (Ogawara district and Chuyashiki district), and in 2022, evacuation orders were also lifted in the central area of Okuma Town (the Specified Reconstruction and Revitalization Base Areas) [2]. Nevertheless, approximately 50.9% of the town area (4,004 ha) remains designated as difficult-to-return zones (DRZs), and many residents remain evacuated [2]. At present, approximately 1,300 residents live in Okuma Town, corresponding to only about 10% of the pre-accident population [2].

For the long-term recovery of the town, it is important to create conditions under which former residents are able to return if they wish to do so. Achieving this requires not only the restoration of living environments, but also the provision of objective, easily understandable information regarding the radiation environment. Against this background, the Japanese government has established Specified Living Areas for Returnees (SLARs) within DRZs and has progressively implemented decontamination activities and infrastructure development to enable residents with an intention to return to resume their lives [3]. Compared with other DRZs that include extensive forested areas, the SLARs are areas with a higher likelihood of future habitation and therefore occupy an important position from the perspectives of radiation protection and recovery policy [4].

Previous studies have extensively investigated the radiation environment in Fukushima Prefecture following the nuclear accident, including large-scale monitoring of ambient dose equivalent rates and evaluations conducted after the lifting of evacuation orders [1,5,6]. These studies have contributed substantially to understanding regional trends in radiation exposure and the effectiveness of remediation measures. However, few studies have reported evaluations of environmental radioactivity based on field measurements conducted in SLARs located within DRZs prior to the full-scale implementation of decontamination and demolition activities. Moreover, data focusing on prospective living environments, such as public roads and individual residential properties in which residents are expected to spend substantial time after returning, remain scarce.

Against this background, the present study aimed to evaluate environmental radioactivity in the SLARs of Okuma Town based on field measurements conducted before the progression of decontamination and demolition activities. Specifically, we measured ambient dose equivalent rates on public roads using a car-borne survey and within individual residential properties to characterize the spatial distribution of radiation levels and to estimate potential radiation exposures under scenarios of resident returns. By providing baseline data prior to remediation, this study seeks to contribute to the assessment of future dose reduction effects and to support ongoing discussions of radiation protection and recovery planning in post-accident settings.

Materials and methods

Survey location

The FDNPP (37° 25’ N, 141° 02’ E) is located across Okuma Town and Futaba Town in Fukushima Prefecture, Japan. Although evacuation orders for Okuma Town were partially lifted in 2019, approximately 50.9% of the town remains designated as DRZs. Within these zones, approximately 440 ha have been designated as SLARs [4].

In this study, SLARs adjacent to the Specified Reconstruction and Revitalization Base Areas were divided into three areas according to differences in the status of decontamination (Fig 1): Area 1 (Ottozawa district), in which little to no decontamination had been conducted; Area 2 (Kuma and Kumagawa districts), in which decontamination was ongoing; and Area 3 (Nogami and Shimonogami districts), in which major roads had already been decontaminated following the lifting of the evacuation order.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Mapped ambient dose equivalent rate distribution (September 2025).

Ambient dose equivalent rate and geographic location data were obtained from the RadiProbe software and replotted by the authors to create this replacement figure on a public-domain base map from the Natural Earth (https://www.naturalearthdata.com/). This figure is similar to, but not identical to, the original RadiProbe software image and is provided for illustrative purposes only. In addition, area lines were added by the authors to indicate the three study areas: Area 1, the Ottozawa district, where little or no decontamination had been conducted; Area 2, the Kuma and Kumagawa districts, where decontamination was ongoing; and Area 3, the Nogami and Shimonogami districts. This figure was created by the authors and is published under the CC BY 4.0 license.

https://doi.org/10.1371/journal.pone.0354457.g001

Effective dose calculation

In this study, environmental radiation was evaluated in accordance with the indicators for assessing additional radiation doses provided by the Ministry of the Environment. Accordingly, the annual external effective dose was calculated using the method presented by the Ministry [8].

The annual external effective dose was calculated using the following equations [8]:

E _i ＝ E _out ＋E _in

E _out＝ D ·T_out

E _in= D· T _in· r

where E is the annual external effective dose (mSv); E _i is the estimated daily external effective dose (mSv); E _out and E _in are the outdoor and indoor effective doses; D are ambient dose rates measured at 1 m above ground outdoors (µSv/h); T represents occupancy time (16 h indoors, 8 h outdoors per day), based on the guidelines of the Ministry of the Environment; and r = 0.4 is the gamma-ray attenuation factor for wooden houses [9].

Results

The car-borne survey showed that the proportion of locations with an ambient dose rate of 0.42 µSv/h or higher decreased from 55.6% in November 2024 to 40.2% in September 2025 (Fig 2). When analyzed by area, the proportion of locations with an ambient dose rate of 0.42 µSv/h or higher showed a significant decrease over time in Areas 2 and 3. In contrast, no significant temporal change in dose levels was observed in Area 1 (Fig 3).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Temporal variation in the ambient dose equivalent rate on public roads in the SLAR.

https://doi.org/10.1371/journal.pone.0354457.g002

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Temporal variation in ambient dose equivalent rate on public roads by area in the SLAR.

https://doi.org/10.1371/journal.pone.0354457.g003

In September 2025, the median ambient dose equivalent rate on public roads was 0.34 µSv/h. The median ambient dose equivalent rates on public roads were 1.05 µSv/h in Area 1, 0.37 µSv/h in Area 2, and 0.24 µSv/h in Area 3 (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Ambient dose equivalent rates on public roads by area in the SLAR.

https://doi.org/10.1371/journal.pone.0354457.t001

Ambient dose equivalent rates were also measured within the premises of 107 residential properties. The median ambient dose equivalent rate in front of house entrances was 0.96 µSv/h. By area, median ambient dose equivalent rates in front of house entrances were 3.86 µSv/h in Area 1, 0.90 µSv/h in Area 2, and 0.45 µSv/h in Area 3 (Table 2). The median ambient dose equivalent rate of all measurement locations was 1.61 µSv/h, with values of 6.54 µSv/h in Area 1, 1.46 µSv/h in Area 2, and 0.65 µSv/h in Area 3. At all measurement points, including the house entrance, rear side of the house, right side, and left side of the house, ambient dose equivalent rates were highest in Area 1, whereas significantly lower values were observed in Areas 2 and 3 (Fig 4).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Ambient dose equivalent rates and effective doses at residential properties from November 2024 to February 2025.

https://doi.org/10.1371/journal.pone.0354457.t002

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Ambient dose equivalent rates at measurement points in residential properties.

https://doi.org/10.1371/journal.pone.0354457.g004

Discussion

In this study, we evaluated ambient dose equivalent rates on public roads using a car-borne survey, as well as ambient dose equivalent rates within residential properties, in the SLAR in Okuma Town. The results demonstrated that the overall proportion of relatively high-dose locations on public roads decreased over the study period (Table 1, Figs 2 and 3). In particular, significant temporal decreases were observed in Areas 2 and 3, whereas no significant change was detected in Area 1 (Table 2, Fig 4). A similar pattern was observed in residential properties, whereambient dose equivalent rates were significantly lower in Areas 2 and 3 compared with Area 1. A comparison between measurements obtained on public roads and those within residential properties enables a more realistic understanding of the radiation environment relevant to returning residents. In this study, ambient dose equivalent rates measured within residential properties tended to be higher than those measured on public roads. This finding may be explained by the fact that radiocesium is more readily removed from paved surfaces through weathering processes such as rainfall and surface runoff, whereas higher dose rates tend to persist in residential and soil-covered areas.

Furthermore, a consistent spatial pattern was observed in both road-based and residential measurements, with higher dose levels in Area 1 and lower levels in Areas 2 and 3. This consistency suggests that relative differences in radiation levels among administrative areas are maintained even across different measurement environments. These findings support existing knowledge regarding spatial heterogeneity in the post-accident radiation environment and provide important field-based data for assessing radiation exposure in areas in which the progress of decontamination varies. These findings indicate that, even within the SLARs, the radiation environment is not homogeneous and that substantial differences exist among administrative areas in terms of dose distribution and the progress of dose reduction [1,5]. In Areas 2 and 3, decontamination activities and building demolition have already been initiated, and the dose reduction effects of these measures may have influenced the observed dose levels [1,6]. This spatial heterogeneity has important implications for radiation protection and reconstruction efforts. Rather than applying uniform countermeasures across the entire designated area, considering area-specific approaches based on the local radiation environment may be reasonable [10,11]. In areas where dose levels have relatively decreased, it may be possible to advance measures aimed at supporting the return of residents and the restoration of daily life, such as the relaxation of access restrictions, whereas areas with persistently higher dose levels may require continued environmental management and monitoring [12,13].

SLARs were established based on the intentions of residents to return, with priority given to those administrative areas in which a higher proportion of residents expressed a desire to resume living. Compared with other DRZs that include extensive forested land, these designated areas are therefore more likely to be inhabited in the future. The present results, demonstrating reductions in ambient dose equivalent rates in parts of the designated area, provide scientific information that may help inform ongoing assessments and discussions on appropriate radiation protection and recovery planning in post-accident settings [10].

At the same time, previous studies have suggested that reductions in radiation dose alone do not necessarily lead directly to residents’ decisions to return. Even when levels decrease, decisions regarding return are influenced by multiple factors, including the restoration of infrastructure, access to medical and welfare services, educational environments, and employment opportunities [14,15]. Accordingly, radiation dose management should be implemented as part of comprehensive support packages that integrate infrastructure development and social support measures [16,17].

In this study, ambient dose equivalent rates at front entrances tended to be lower than those at other measurement points within residential properties. This is likely because front entrances are often covered with concrete or other impermeable surfaces. Previous studies have shown that radioactive cesium deposited on impermeable surfaces such as roads and roofs is efficiently removed by rainfall and surface runoff, leading to faster decreases in ambient dose rates compared with soil-covered areas [13]. In contrast, most other measurement points were located on soil-covered ground, where radioactive cesium tends to be retained for longer periods, resulting in relatively higher ambient dose equivalent rates.

Notably, the present study also identified locations in Area 1 where ambient dose equivalent rates remained relatively high, even approximately 15 years after the accident. This finding suggests that time-dependent decay and general remediation measures alone may be insufficient to achieve dose reduction in certain locations [1,18]. Continued monitoring focused on such localized higher-dose areas, together with targeted environmental management, remains important for long-term radiation protection and recovery [11,12].

This study has several limitations. First, ambient dose equivalent rates within residential properties were measured at a limited number of points within each property, and these measurements thus do not necessarily adequately represent the status of entire premises or future residential environments. In particular, when residents return in the future, existing houses are expected to be demolished, followed by decontamination of the land. The rates measured in this study thus do not directly reflect radiation exposure levels during future residential use. Nevertheless, the measurements taken in the course of this study provide valuable empirical data on the radiation environment prior to demolition and decontamination. These pre-remediation measurements provide an important baseline for evaluating changes in ambient dose equivalent rates before and after remediation activities. Moreover, site-specific measurements based on actual field surveys contribute to the characterization of localized dose distributions that may not be fully captured by model-based assessments alone [18]. In addition, annual external effective doses in this study were evaluated using the calculation method adopted by the Ministry of the Environment, Japan for regulatory and administrative purposes. This method treats ambient dose rates as a proxy for external effective dose and does not apply an effective dose conversion coefficient. Although calculations based on ICRP Publication 144 were not performed in the present study, such approaches may be useful for more realistic effective dose assessments [19].. Further, the measurement period of this study was relatively short and did not account for seasonal or long-term temporal variations in ambient dose equivalent rates. Previous studies have shown that ambient dose rates can change over time due to factors such as weather conditions, radioactive decay, and environmental processes [20]. As a result, continuous or long-term monitoring would be necessary to better understand temporal trends and improve long-term dose assessments.

Conclusions

This study evaluated ambient dose equivalent rates on public roads and within residential properties in SLAR in Okuma Town, Fukushima Prefecture, to assess potential radiation exposure under scenarios of resident return. The results showed an overall decreasing trend in ambient dose equivalent rates on public roads during the study period.

In contrast, estimated external dose within residential properties prior to remediation and demolition remained below the reference level of 20 mSv for post-accident exposure situations, while indicating the continued importance of remediation and other dose reduction measures. As demolition and decontamination activities are currently ongoing in the study area, continued dose assessments will be essential to support the optimization of radiation protection and to provide a scientific basis for future recovery and reconstruction efforts.

Supporting information

Acknowledgments

The authors are grateful to the staff of Okuma Town and the related administrative organizations for their support and cooperation in conducting the field surveys in the Specified Living Areas for Returnees. We also sincerely thank the residents who provided access to their properties for radiation measurements.