Estimate of Fukushima-derived radiocaesium in the North Pacific Ocean in summer 2012
Distributions of radiocaesium (134Cs and 137Cs) derived from the Tokyo Electric Power Company (TEPCO) Fukushima Dai-ichi Nuclear Power Plant (FNPP1) accident in the North Pacific Ocean in the summer of 2012 were investigated. We have estimated the radiocaesium inventory in the surface layer using the optimal interpolation analysis and the subducted amount into the central mode water (CMW) by using vertical profiles of FNPP1-134Cs and mass balance analysis as the first approach. The inventory of the 134Cs in the surface layer in the North Pacific Ocean in August–December 2012 was estimated at 5.1 ± 0.9 PBq on 1 October 2012, which corresponds to 8.6 ± 1.5 PBq when it was decay corrected to the date of the FNPP1 accident, 11 March 2011. It was revealed that 56 ± 10% of the released 134Cs into the North Pacific Ocean, which was estimated at 15.3 ± 2.6 PBq, transported eastward in the surface layer in 2012. The amount of 134Cs subducted in the CMW was estimated to be 2.5 ± 0.9 PBq based on the mass balance among the three domains of the surface layer, subtropical mode water, and CMW.
KeywordsRadiocaesium FNPP1 accident Central mode water Subtropical mode water Inventory North Pacific Ocean Mass balance
As a result of an extraordinary earthquake and subsequent giant tsunami on 11 March 2011, Tokyo Electronic Power Company Fukushima Dai-ichi Nuclear Power Plant (hereafter referred to as FNPP1), locate 37.42°N, longitude 141.03°E, received serious damage. Because of the damaged FNPP1 reactors, large amounts of radiocaesium (137Cs, with a half-life of 30.2 years and 134Cs, with a half-life of 2.06 years) were directly released into the North Pacific Ocean by atmospheric deposition and direct discharge of liquid-contaminated stagnant water from the FNPP1 accident site, mostly in March and April 2011 [1, 2]. It was recorded that extremely high radiocaesium activity concentrations reaching 6.8 × 107 Bq m−3 were observed on 6 April 2011 at the site of the FNPP1 accident [3, 4]. From April to May 2011, a basin-scale assessment indicated that the FNPP1-derived radiocaesium activity concentrations in the surface waters in the North Pacific Ocean ranged from a few to 1000 Bq m−3 [1, 5]. Although the radiocaesium released by atmospheric deposition and direct discharge of liquid-contaminated stagnant water into the North Pacific Ocean by the FNPP1 accident were estimated by numerous studies, the median values of the released amount tended to converge on a range of 15–20 PBq [6, 7]. The directly discharged radiocaesium in the liquid-contaminated stagnant water into the North Pacific Ocean was estimated to be 3.5 ± 0.7 PBq using the 131I/137Cs activity ratio, which is a useful tracer to distinguish the release pathways between direct release and atmospheric deposition [2, 3]. The total 134Cs inventories ranged from 15.2 to 18.3 PBq . This estimation is in good consensus with the 134Cs inventories estimated by other methods, namely, 15.3 ± 2.6 PBq by optimal interpolation analysis  and 16.1 ± 1.4 PBq by the North Pacific model . It is considered that more than 75% of the radiocaesium released into the atmosphere from the FNPP1 event was deposited into the North Pacific Ocean.
Before the FNPP1 accident, 137Cs already existed in the North Pacific Ocean due to the large-scale atmospheric nuclear weapons testing that occurred in the late 1950s and early 1960s and the Chernobyl accident in 1986 [10, 11, 12, 13]. In the 2000s, 137Cs activity concentrations were 1.5–2 Bq m−3 with almost homogenous distribution; however, relatively high activity, which exceeded 2 Bq m−3, was observed in the western part of the subtropical gyre in the North Pacific Ocean . Because of its shorter half-life, 134Cs derived from the global fallout by large-scale nuclear weapons tests had already decayed in the seawater by 1993 . 134Cs, therefore, is regarded as an adequate chemical tracer released from the FNPP1 accident to investigate the transport of the FNPP1-derived contaminated seawater in the North Pacific Ocean. Taking into account that the 134Cs/137Cs activity ratio originating from the FNPP1 accident was almost 1 [15, 16], we can consider that FNPP1-137Cs caused an additional increase of approximately 22–27% in the 137Cs inventory that existed in the North Pacific prior to the FNPP1 accident, which was estimated to be 69 PBq from nuclear weapons testing .
The basin-scale assessment indicated that the directly discharged and atmospheric deposited FNPP1-134Cs in the North Pacific Ocean was transported eastward in the surface layer by the Kuroshio and Oyashio Extensions and the North Pacific Current. In the summer of 2012, approximately 1.5 years after the FNPP1 accident, the main body of radioactive contaminated seawater was observed in the region of 40°N–45°N between 165°E and 170°W in the central North Pacific Ocean . The zonal speed of FNPP1 radiocaesium in the surface water at mid-latitude in the North Pacific Ocean was estimated to be approximately 8 cm s−1 until March 2012 [1, 5]. The FNPP1-134Cs was further transported eastward and to a higher latitude as subarctic currents with a zonal speed of 3.5 cm s−1 from March 2012 to August 2014 . The FNPP1-134Cs approached North America and diverged into the northward flowing Alaska Current and the southward flowing California Current beginning in February 2013 [18, 19, 20]. The arrival of FNPP1-radionuclides off the west coast of North America, on the Canadian continental shelf, was first reported in February 2015 [19, 20]. The North Pacific model also reproduced the eastward transport of FNPP1 radiocaesium in the North Pacific Ocean . In addition, FNPP1 radiocaesium was found south of the Kuroshio Front, with a subsurface maximum at a depth of approximately 200–600 m in the western subtropical region [1, 18, 21, 22, 23]. The subsurface maxima of 137Cs existed in the subtropical mode water (STMW) with a potential water density anomaly (σθ) = 25.0–25.6 kg m−3  and the central mode water (CMW) with σθ = 26.0–26.5 kg m−3 [25, 26]. The STMW is formed as a deep winter mixed layer south of the Kuroshio and Kuroshio Extension, which is located north of approximately 28°N from ~ 132°E to the dateline [27, 28, 29]. The CMW is characterized by a pycnostad in the lower pycnocline in the North Pacific subtropical gyre. The CMW is formed in the deep water mixed layer located north of the Kuroshio Extension [30, 31]. The subsurface peak of FNPP1 radiocaesium in the STMW suggests that atmospheric deposition occurred in the formation region of STMW, then subduction into the ocean interior occurred during the southwestward transport within 10 months after the FNPP1 accident . Kaeriyama et al.  estimated the amount of FNPP1-134Cs in the STMW at 4.2 ± 1.1 PBq in October–November 2012. However, the FNPP1 radiocaesium amount subducted into the CMW has not exactly been estimated.
The primary objectives of this study were as follows: (i) to investigate the distributions of radiocaesium in the surface seawater, (ii) to estimate the radiocaesium inventory in the surface layer, and (iii) to estimate the FNPP1 radiocaesium amount subducted into the CMW in the North Pacific Ocean in 2012. Furthermore, we discuss the fate of FNPP1 radiocaesium subducted into the ocean interior.
To elucidate the distributions of the radiocaesium activity concentrations, it is necessary to use as many data points as possible. We, therefore, compiled the data available from the literature and the monitoring results. Most of the data before the FNPP1 accident were included in the database, “Historical Artificial Radionuclides in the Pacific Ocean and its Marginal Seas (HAM database)” . The data observed after the FNPP1 accident were also shown in Aoyama et al. . Monitoring data were also available to use in this analysis [33, 34, 35].
In this study, we focused on the distribution of the central North Pacific Ocean in 2012, because the centre of the radioactive contaminated seawater in the surface layer was observed in the central region of the North Pacific Ocean, at approximately 180° in August–December 2012 . The data used in this analysis consisted of 236 records for 134Cs and 343 records for 137Cs. These records are listed in the supplementary Table S1. The activity concentrations of radionuclides in seawater are radioactive decay corrected to 1 October 2012 and 11 March 2011. The term “surface seawater” used in this study is defined as a sample collected at less than a 10 m depth. The vertical profiles of radiocaesium activity concentrations were used to estimate the radiocaesium inventory. The seawater was collected at 5 stations from 5 to 1006 m in depth between August and September 2012 during the research cruise of “Hakuho-maru” (HK12-4). The detection limits of 134Cs and 137Cs in the seawater samples collected by HK12-4 were 0.17 and 0.28 Bq m−3, respectively. In addition, we used vertical profile data observed on 8 September 2012 during the research cruise of “Mirai” (MR12-E03) reported by Kumamoto et al. . The detection limits of 134Cs and 137Cs were 0.1 and 0.17 Bq m−3, respectively. The details of the measurement were described in Kumamoto et al. [17, 18]. The activity concentrations of the radiocaesium in seawater are radioactive decay corrected to 11 March 2011. The seawater samples for obtaining the vertical profiles were collected at 58 points and are listed in the supplementary Table S2.
Optimal interpolation analysis methods
Optimal interpolation analysis (OI) is an effective analytical scheme for depicting the horizontal distributions of irregularly distributed measurement data to gridded analytical value . Detail of the analytical method has been already described in Inomata et al. [8, 13]. 134Cs and 137Cs activity concentrations estimated by OI (hereafter OICs134 and OICs137) were produced on a grid with a horizontal resolution of 3° longitude × 2° latitude, because the area is approximately equal to the geodetical area of one grid exactly, and the meridional transport prevails in the analytical area associated with the Kuroshio Current. The data used in the OI analysis were selected within the effective radius R; in other words, the data used in the OI analysis was located inside the circle with R. The background error at the observation point (σο) was considered. Based on a higher correlation coefficient for each gridded value estimated by the Akima spline method , which is also used as a conventional interpolation method and a gridded algorithm, and the OI, we selected 800 km and 100 as optimal values of R and σo, respectively .
Estimation of the water column inventory
Based on the estimated MPD, the water column inventory was calculated based on the assumption that radiocaesium is distributed constantly from 0 to 72 m and decreases monotonically from a depth of 72–144 m. The error in the OICs134 and OICs137 inventory was estimated as the square root of the sum of the analytical standard deviation, statistical error of OI, and error of mixed penetration depth estimation.
Mass balance analysis
The atmospheric chemical transport model and physical parameters in the North Pacific Ocean
It is very difficult to investigate the radiocaesium deposition into the North Pacific Ocean. To estimate the deposition amount of radiocaesium into the mode water formation region in the North Pacific Ocean, we used the following three atmospheric chemical transport models taking into account the uncertainty among the different models: the Model of Aerosol Species IN the Global Atmosphere (MASINGAR I), its newer version (MASINGAR MK-II), and the Meteorological Research Institute—Passive-tracers Model for radionuclides (MRI-PM/r). The simulation settings are similar to those of Aoyama et al. ; therefore, we have refrained from repeating the details here.
MASINGAR I is a global aerosol transport model, which is coupled online with atmospheric general circulation models (AGCMs) of the Japan Meteorological Agency (JMA) . MASINGAR I is coupled with an AGCM called MRI/JMA 98  and has been used as the operational dust forecasting model by the JMA. The model resolutions were set to a T106 Gaussian horizontal grid (approximately 1.125° × 1.125°) and 30 vertical layers from the surface to a height of 0.4 hPa. MASINGAR MK-II is also coupled with the global atmospheric model  as a component of the Earth system model of the Meteorological Research Institute, the MRI-Earth System Model . The model resolutions were set to a TL319 horizontal grid (approximately 0.5625° × 0.5625°) and 40 vertical layers from the ground surface to a height of 0.4 hPa. The MRI-PM/r is a regional-scale offline coupled meteorology-aerosol chemical transport model that was developed based on Kajino et al. . The Advanced Research Weather Research and Forecasting (WRF) model  was used to simulate the meteorological data. The global analysis data provided by the National Centers for Environmental Prediction (NCEP), ds083.2 (1° × 1°, 6-HR time interval), were used for the initial and boundary conditions of the WRF and for the analysis nudging method. The horizontal grid resolution is 60 km with a Mercator map projection covering the North Pacific Ocean and the surrounding countries. And they are vertically 20 layers from the ground to 10 km in height with a terrain-following coordinate.
The σθ was investigated using the basin-scale model of the Japan Coastal Ocean Predictability Experiment (JCOPE2). JCOPE2 provided physical parameters such as potential temperature, salinity, zonal velocity, and meridional velocity of the oceanography. The data was generated from the western North Pacific Ocean (10.5–62°N, 108–180°E) with a horizontal resolution of 1/12°. The JCOPE2 data assimilates the remote-sensing data of altimetry and surface temperature and in situ data of temperature and salinity profiles . The data from 21 March to 31 May, 2011 were used in this study. The σθ value corresponds to the STMW and CMW of the formation region, which were set to 25–25.6 and 25.8–26.4 kg m−3, respectively.
Distribution and inventory of radiocaesium in the surface layer from August to December 2012
Radiocaesium deposition in the mode water formation region in the North Pacific Ocean
The amount of 134Cs deposited in the North Pacific Ocean compared with the three models
Potential density anomaly (σθ) (kg m−3)
MASINGAR MK-II (PBq)
Evaluation of FNPP1 radiocaesium transported eastward in the surface layer
As described above, the OICs134 amount in the surface layer decay corrected to the accident date was estimated to be 8.6 ± 1.5 PBq. This corresponds to 56 ± 10% of the total amount of released 134Cs into the North Pacific Ocean. Although OI results did not cover the whole region in the North Pacific Ocean (Fig. 1), the OI results included the high radiocaesium activity concentration area in the North Pacific Ocean, located in a region from 25°N to 50°N and from 135°E to 135°W in April and May 2011 [1, 5, 8, 9]. Therefore, we think that the underestimation of OICs134 associated with the unavailable OI analysis would be small.
Comparison with atmospheric deposition and directly discharged FNPP1 radiocaesium in the mode water formation region
FNPP1-derived radiocaesium released into the North Pacific Ocean originated from atmospheric deposition and direct discharge from the FNPP1. These were transported eastward along the surface currents. The amount of 134Cs shown in Table 1 are considered to be only atmospheric deposition. The directly released FNPP1 radiocaesium, which was estimated to be 3.5 ± 0.7 PBq, was mainly transported in the CMW formation region. Added to this directly released amount, the amount of FNPP1-derived radiocaesium in the North Pacific Ocean was 12.4 PBq for the Masingar I model, 13.6 PBq for the Masingar MK-II model, and 6.2 PBq for the MRI-PMr model. Considering that the OICs134 amount in the surface layer was estimated to be 8.6 ± 1.5 PBq, the FNPP1-derived radiocaesium amount transported eastward in the surface layer without subducting into the CMW was approximately 69% and 63% according to the Masingar I and MK-II models, respectively. For the MRI-PMr model, the deposition amount in the mode water formation region was smaller than that transported in the surface layer, which resulted in the deposition of a larger amount out of the JCOPE data region. Furthermore, the deposition into the STMW formation region was larger in the MRI-PMr model than that in the Masingar I and MK-II models. The difference in the deposition amount and deposited region among the three models would reflect the different transport and deposition schemes.
The fate of FNPP1 radiocaesium subducted into the ocean interior
It was reported that the main body of the FNPP1 radiocaesium-labelled STMW was observed in January 2012, approximately 10 months after the FNPP1 accident, and transported to south of 30°N in November 2012, approximately 20 months after the accident in the North Pacific south of Japan (NPSJ) [21, 49]. As shown in Fig. 3, a larger deposition of FNPP1 radiocaesium occurred three times from March to early April 2011. Considering that seawater was still in the colder season under the strong East Asia winter monsoon from March to early April 2011, the FNPP1 radiocaesium-contaminated surface layer in the STMW formation region, which is south of Kuroshio and the Kuroshio Extension, becomes colder and denser, resulting with the formation of FNPP1 radiocaesium-labelled STMW to be associated with the deep mixed layer. According to the previous research based on the physical parameters in seawater, during the winter, the STMW is capped by the seasonal thermocline and then transported by the southwestward Kuroshio recirculation until approximately 20°N to south and east of Taiwan [50, 51]. It was also found that the FNPP1 radiocaesium-labelled STMW was transported from NPSJ to the East China Sea (ECS), following then the Sea of Japan (SOJ) via the Tsushima Kaikyo, and transported into the North Pacific Ocean via the Tsugaru Strait within several years . The integrated amount of FNPP1-137Cs that returned to the North Pacific Ocean through the Tsugaru Strait until 2016 was estimated to be 0.09 ± 0.01 Bq , which corresponds to 2.1% of the estimated total amount of FNPP1-137Cs in the STMW . This indicates that a small amount of the FNPP1-134Cs went back to the surface layer in the North Pacific Ocean.
The subsurface maxima of radiocaesium activity in the CMW was observed in the seawater with σθ of 26.1–26.3 at 34°N–39°N, 165°E from October 2011 to June/July 2012 . These 134Cs activity concentrations in the CMW were higher than those in the STMW . This is consistent with this study, in which the FNPP1 radiocaesium amount, which is the sum of the direct release and atmospheric deposition amounts, in the CMW formation region was larger than that in the STMW formation region.
The subducted FNPP1 radiocaesium amount in the ocean interior estimated in this study was approximately 44% of the total amount injected into the North Pacific Ocean, and this estimation is consistent with an estimation from the high resolution ocean modelling study . In Kamidaira’s study, 42.7% of FNPP1-137Cs was penetrated below the mixed layer by eddies associated with geostrophic secondary circulations.
Distributions of the FNPP1 radiocaesium released into the North Pacific Ocean were investigated, and the amounts were estimated in the surface layer, STMW, and CMW in the summer of 2012. The centre of FNPP1-derived contaminated seawater was located in the region approximately 40 and 50°N between 165°E and 178°W. The estimated 134Cs in the surface layer was 5.1 ± 0.9 PBq on 1 October 2012 and 8.6 ± 1.5 PBq decay corrected on 11 March 2011. For 137Cs, the inventory was estimated to be 13.3 ± 2.3 PBq on 1 October 2012 and 13.9 ± 1.8 PBq on 11 March 2011. Approximately 56 ± 10% of the released amount of 134Cs in the North Pacific Ocean was mainly transported eastward in the surface layer. The remaining 44% of the 134Cs amount released into the surface seawater was subducted into the ocean interior. Considering that the amount of 134Cs injected into the North Pacific Ocean and subducted into the STMW obtained in the previous study were estimated to be 15.3 ± 2.6 PBq and 4.2 ± 1.1 PBq, the amount of 134Cs subducted into the CMW was estimated to be 2.5 ± 0.9 PBq based on the mass balance among the three domains; that is, the surface layer, STMW and CMW.
This research was financially supported by Grant-in-Aid for Scientific Research on Innovative Areas, “Interdisciplinary study on environmental transfer of radionuclides from the Fukushima Dai-ichi NPP Accident” (Project No. 25110511) of the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT). This research was also supported by the cooperation program by Institute of Radiation Emergency Medicine, Hirosaki University (JFY2016, 2017).
- 1.Aoyama M, Hamajima Y, Hult M, Uematsu M, Oka E, Tsumune D, Kumamoto Y (2016) 134Cs and 137Cs in the North Pacific Ocean derived from the March 2011 TEPCO Fukushima Dai-ichi Nuclear Power Plant accident, Japan. Part one: surface pathway and vertical distributions. J Oceanogr 72:53–65CrossRefGoogle Scholar
- 6.Aoyama M, Kajino M, Tanaka TY, Sekiyama TT, Tsumune D, Tsubono T, Hamajima Y, Inomata Y, Gamo T (2016) 134Cs and 137Cs in the North Pacific Ocean derived from the March 2011 TEPCO Fukushima Dai-ichi Nuclear Power Plant accident, Japan. Part two: estimation of 134Cs and 137Cs inventories in the North Pacific Ocean. J Oceanogr 72:67–76CrossRefGoogle Scholar
- 7.Buesseler K, Dai M, Aoyama M, Benitez-Nelson C, Charmasson S, Higley K, Maderich V, Masque P, Oughton D, Smith JN (2016) Fukushima Daiichi–derived radionuclides in the ocean: transport, fate, and impacts. Annu Rev Mar Sci 9:1–31Google Scholar
- 8.Inomata Y, Aoyama M, Tsubono T, Tsumune D, Hirose K (2016) Spatial and temporal distributions of 134Cs and 137Cs derived from the TEPCO Fukushima Daiichi Nuclear Power Plant accident in the North Pacific Ocean by using optimal interpolation analysis. Environ Sci Process Impacts 18:126–136CrossRefGoogle Scholar
- 24.Masuzawa J (1969) Subtropical mode water. Deep Sea Res 16:463–472Google Scholar
- 33.Japan Coast Guard (2012) Radioactivity monitoring in Tokyo Bay, off of Fukushima and Ibaraki Prefectures. http://www.kaiho.mlit.go.jp/info/kouhou/h24/k20121106/k121106-1.pdf (in Japanese). Accessed 20 Aug 2018
- 34.Ministry of Education, Culture, Sports, Science and Technology (2012) Readings of seawater and dust monitoring in sea area. http://radioactivity.nsr.go.jp/en/list/259/list-1.html. Accessed 20 Aug 2018
- 35.Ministry of Environment (2013) The first marine environmental monitoring in the tsunami-affected area in 2012. http://www.env.go.jp/water/kaiyo/monitoring/eq/attach/result_20131105-1.pdf (in Japanese). Accessed 20 Aug 2018
- 39.Aoyama M, Inomata Y, Tsumune D, Tateda Y (2018) Fukushima radionuclides in the marine environment from coastal region of Japan to the Pacific Ocean through the end of 2016. Prog Nucl Sci Technol (accepted)Google Scholar
- 45.Yukimoto S, Yoshimura H, Hosaka M (2011) Meteorological Research Institute-Earth System Model Version 1 (MRIESM1)—model description. Technical reports of the Meteorological Research Institute 64Google Scholar
- 47.Shamarock W, Klemp J, Dudhia J (2008) A description of the advanced research WRF version 3. NCAR technical note NCAR/TN/u2013475+STRGoogle Scholar
- 49.Kumamoto Y, Aoyama M, Hamajima Y, Oka E, Murata A (2018) Time evolution of Fukushima-derived radiocesium in the western subtropical gyre of the North Pacific Ocean by 2017. J Radioanal Nucl Chem 312:1703–1710Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.