*5.3. Estimation of 90Sr Input to the Ocean from the FDNPP*

The continuous release from the FDNPP was the main source to the Fukushima coast. The amount of 137Cs released daily to the ocean was estimated to be from 8.1 GBq day−<sup>1</sup> [7] to 30 GBq day−<sup>1</sup> [8] in 2012 on the basis of simulation of the 137Cs activities of seawater in the harbor and at the north discharge gate, respectively. We examined the amount of released 90Sr based on that of 137Cs in 2013 by using Equation (1):

$$N\_{\rm Sr-90} = N\_{\rm Cs-137} \times \left(\frac{\text{C}\_{\rm Sr-90}}{\text{C}\_{\rm Cs-137}}\right)\_{\rm SW} = \text{C}\_{\rm Cs-137} \times F \times \left(\frac{\text{C}\_{\rm Sr-90}}{\text{C}\_{\rm Cs-137}}\right)\_{\rm SW} \tag{1}$$

where *N*, *C*, and *F* represent release rate, activity, and conversion factor from activity to daily release rate of 137Cs, respectively. For the estimation of daily released 137Cs, the activities of 137Cs at T2-1 300 m south of the south discharge gate were used (Table 3) [6]. Most of the 137Cs activities were lower than the MDA (1.2–1.5 Bq L<sup>−</sup>1). To avoid overestimation of the averaged 137Cs activity, we used only precise analysis data. The 137Cs activity at the T2-1 site ranged from 0.14 to 0.98 Bq L−<sup>1</sup> and showed considerable variation (mean value = 0.60 <sup>±</sup> 0.35 Bq L<sup>−</sup>1). The conversion factors, *F*, from activity to daily release rate of 137Cs were obtained on the basis of the amount of released 137Cs in the direct release event of March 2011 and 137Cs activities at the T2-1 site [7,20]. The conversion factor applied was 25.5 <sup>×</sup> 109 [20] for the T2-1 site.


**Table 3.** Estimation of the release rates for 90Sr and 137Cs into the ocean from the FDNPP site in May 2013. *C*: activity; *F*: conversion factor; *N*: release rate.

The resulting daily released amount of 90Sr was 9.1 <sup>±</sup> 6.1 GBq day−<sup>1</sup> during our sampling campaign in May 2013. The observed 90Sr activity in the coastal region was too low to disturb the ecological system and affect the background radiation dose, as mentioned above. Continuous release could increase the inventory of 90Sr in the Pacific Ocean. If the constant release (9.1 GBq day−1) continued over the year, the annual release rate would be estimated at 3.3 TBq yr−1, which is small relative to the inventory of 105 PBq in the ocean [3]. However, 90Sr in seawater should be closely observed to detect any unexpected release from the nuclear reactor buildings and the contaminated water storage tanks. This estimation needs to assume a stable release rate from the single source. As discussed above, the low 90Sr/ 137Cs source contributed to seawater around the FDNPP. Therefore, this result could be overestimated.

In this study, the 90Sr/ 137Cs activity ratio of 0.66 <sup>±</sup> 0.05, which was influenced by continuous release from the FDNPP, was distinguished based on precise 90Sr analysis. Buesseler et al. [11,30] suggested that the possible source of 137Cs was not only continuous release from the FDNPP but also the input from subsurface groundwater [31], river water [32], and desorption from the marine sediments in the coastal region [29,30,33]. The environmental migration of 137Cs through particulate and dissolved fluvial inputs, and remineralization from the sediments contaminated by direct discharge of stagnant water from 26 March to 6 April 2011, must also be taken into consideration. 90Sr/ 137Cs activity ratios could fluctuate according to the source in the FDNPP area and remobilization of 137Cs in coastal water. In addition to monitoring for ongoing release from the reactor buildings and possible leakage of stored contaminated water in tanks, continuous measurement of 90Sr is necessary for investigation of the migration of 137Cs in the marine environment. A combination of other fission product nuclides, 129I and 3H activity, will provide precise information for the current status of leakages from stagnant water, groundwater, and stored water in tanks.

#### *5.4. Estimation of E*ff*ective Dose Rate by Ingestion from Marine Products*

90Sr dispersion to the coastal area is the most serious issue for fisheries due to its radiotoxicity. We estimated the dose impact to human health from marine products. The highest 90Sr activity (29.13 mBq L−<sup>1</sup> at AN7; Table 1) was comparable to typical levels for North Pacific surface seawater in the early 1960s during nuclear weapons testing [34]. Taking into consideration the processes in the food chain and the highest activity in the coastal water observed in this study (29.13 mBq L−<sup>1</sup> at AN−7), we obtained Equation (2):

$$D = \mathbb{C} \times \mathbb{C}F \times I\mathbb{R} \times F \tag{2}$$

where *D* is representative of the annual dose rate. *C*, *CF*, *IR*, and *F* are representative of the 90Sr activity in seawater, the concentration factor from seawater to marine products (5–10 [35]), the intake rate of marine products (28.4 kg yr−<sup>1</sup> [36]), and dose coefficient (2.8 <sup>×</sup> <sup>10</sup>−<sup>8</sup> Sv/Bq [37]), respectively. It should be noted that these concentrations are quite small (0.23 μSv yr−1) compared with the International Commission on Radiological Protection (ICRP) limit of 1 mSv yr−<sup>1</sup> for a member of the general public. Much higher 90Sr activities were observed at monitoring points near the south (150−670 mBq L<sup>−</sup>1) and north (260−5800 mBq L<sup>−</sup>1) discharge gates [6]. Even this anomalously high 90Sr activity (5800 mBq L−1) close to the FDNPP would contribute 46 μSv yr−<sup>1</sup> to the annual effective dose rate by marine products.

## **6. Conclusions**

90Sr is useful as a tracer for continuous releases from the FDNPP site. We reported 90Sr data in seawater along with 134Cs and 137Cs in samples collected in the coastal area off Fukushima Prefecture. Released 90Sr was dispersed along the Fukushima coast, and the highest 90Sr activity was 29.13 mBq L−<sup>1</sup> at a sampling site 16 km south of the FDNPP. FDNPP site-derived 90Sr/ 137Cs ranged from 0.16 to 0.64 and the slope of a linear regression fitting of the relationship of Fukushima site-derived 90Sr and 137Cs was 0.66 <sup>±</sup> 0.05, which was similar to the ratio of contaminated water in the FDNPP reactor and turbine buildings. These results suggest that the major contamination source is contaminated water in the FDNPP buildings. On the other hand, the l37Cs-rich source could also affect seawater and cause temporal and spatial variations. The estimated release rate of 90Sr (9.6 <sup>±</sup> 6.1 GBq day−1) was small relative to the inventory of 90Sr in the Pacific Ocean. Release of 90Sr has been controlled by the water shielding wall between the reactor buildings and the harbor since 2015. However, our results imply that if any accidental release of radionuclides, including 90Sr from the FDNPP, occurs during decommissioning of the reactors, the coastal area can be exposed to a high activity plume.

**Author Contributions:** Conceptualization, H.T. and M.Y.; methodology, H.T. and H.O.; investigation, H.T; data curation, H.T., T.Y., K.T., and H.N.; writing—original draft preparation, H.T.; writing—review and editing, H.O, D.T., and J.K.; visualization, H.T.; supervision, M.Y.; project administration, H.T.; funding acquisition, H.T. All authors interpreted the data. All authors provided final approval of the version of the manuscript for publication and agreed to be accountable for all aspects of the work.

**Funding:** This work was supported by JSPS KAKENHI (grant numbers: 24110004 and 26340019).

**Acknowledgments:** We would like to thank the captain, crew, and scientific party of the UM-13-5 cruise by the RTV *Umitaka-Maru* for their collaboration in sampling.

**Conflicts of Interest:** The authors declare no conflicts of interest.
