Concentrations of ⁹⁰Sr and ¹³⁷Cs/⁹⁰Sr activity ratios in marine fishes after the Fukushima Dai‐ichi Nuclear Power Plant accident

Abstract

Strontium‐90 (⁹⁰Sr) was released together with radioactive cesium (Cs) from the Fukushima Dai‐ichi Nuclear Power Plant (FNPP) accident. Although the total amount of ⁹⁰Sr released into the marine environment from the FNPP was estimated to be a factor of 100 lower than that of ¹³⁷Cs, the public have been concerned about the safety of commercial species because of the tendency of accumulation in bones of organisms, its long physical half‐life and less information about ⁹⁰Sr concentrations in commercial species. In this study, we investigated the concentrations of ⁹⁰Sr in marine fishes off Japan before and after the FNPP accident. Except for within a 20‐km radius from the FNPP, ⁹⁰Sr concentrations higher than the background level (<0.046 Bq kg⁻¹ wet) were only detected in 4 of 26 specimens collected off Fukushima. Furthermore, ⁹⁰Sr concentrations in all samples collected outside of Fukushima were under the detection limit (<0.040 Bq kg⁻¹ wet). Also, the concentrations of ⁹⁰Sr in marine fishes were notably lower than those of ¹³⁷Cs, and thus the activity ratios of ¹³⁷Cs to ⁹⁰Sr in the whole body of teleost fishes were 5–190 times higher than that of before the accident. The activity ratio tended to decrease owing to a decrease in the ¹³⁷Cs concentration in marine fish. Our result together with ananalysis of other data sets indicated that the influence of the FNPP accident on ⁹⁰Sr in marine fishes was limited to the area near the FNPP and outside of the FNPP harbor area can be considered as negligible.

Introduction

Since the Fukushima Dai‐ichi nuclear power plant (FNPP) accident that occurred on 11 March 2011, various radioactive materials have been released into the sea directly or through the atmosphere. Two kinds of radioactive strontium (Sr)‐89 (physical half‐life; 50.5 days) and ⁹⁰Sr (28.8 years) were released from the FNPP, whereas other major released radioactive materials were radioactive iodine‐131 (8.02 days), and radioactive cesium (Cs)‐134 (2.06 years) and ¹³⁷Cs (30.1 years). The activity ratio of the ⁸⁹Sr/⁹⁰Sr ratio from the reactor core inventory on 11 March 2011 was estimated to be 11.5 (Nishihara et al., 2012; Povinec et al., 2012), although the amount of the two kinds of radioactive Cs was assumed to be approximately equal to each other (Chino et al., 2011).

The total ⁹⁰Sr released into the atmosphere from the FNPP accident was estimated to be approximately 0.14 PBq (Povinec et al., 2012) by a factor 100 lower than the ¹³⁷Cs, 13–20 PBq (Kobayashi et al., 2013; Aoyama et al., 2015), owing to its lower volatility than Cs (NERH, 2011). The amount of ¹³⁷Cs deposited on the ocean surface from the atmosphere was estimated as 7.6 PBq (Kobayashi et al., 2013) and 12–15 PBq (Aoyama et al., 2015), so this means that most of the ¹³⁷Cs was released into the ocean. In early April 2011, a leak of wastewater, which interacted with the ruptured nuclear fuel rods, included high concentrations of radioactive Sr and radioactive Cs became apparent. The principal direct release of ¹³⁷Cs was estimated to be 3.5 PBq (Tsumune et al., 2012). Based on the ¹³⁷Cs/⁹⁰Sr activity ratio at the beginning of the FNPP accident and 3.5 PBq of ¹³⁷Cs directly released into the ocean, the amount of ⁹⁰Sr in the wastewater was estimated to be approximately 0.04 PBq (Povinec et al., 2012).

Strontium‐89 and ‐90 are nuclear fission products as well as ¹³⁷Cs. The Chernobyl NPP accident, in 1986, released approximately 10 PBq of ⁹⁰Sr and 85 PBq of ¹³⁷Cs into the atmosphere (UNSCEAR, 2000), of which most polluted the land and freshwater environment of Europe and only a negligible level of ⁹⁰Sr and 16 PBq of ¹³⁷Cs was deposited on the ocean surface (IAEA, 2005). Before the FNPP accident, the main sources of ⁹⁰Sr in the North Pacific Ocean and the Sea of Japan were the global and close‐in radioactive fallout after the atmospheric nuclear bomb tests from 1945 to 1980 and the Chernobyl NPP accident (Bowen et al., 1980; UNSCEAR, 2000). The activity ratio of ¹³⁷Cs to ⁹⁰Sr (¹³⁷Cs/⁹⁰Sr) from the global fallout deposition was estimated to be about 1.6 (Krey and Krajewsky, 1970). Strontium migrates in soil columns faster than Cs, because Sr is chemically easier to elute than Cs in the soil column by rainwater (Miller and Reitemeier, 1963; Tsumura et al., 1984). Consequently, the ¹³⁷Cs/⁹⁰Sr activity ratio in the typical surface soil in Japan usually becomes higher than that in the global fallout deposition, and therefore freshwater in some rivers sometimes has been found to include high concentrations of ⁹⁰Sr (Igarashi et al., 2001; Ikeuchi, 2003). For marine organisms, they have different physiological mechanisms for the incorporation and excretion processes of Sr and Cs (Whicker et al., 1972; Young and Folsom, 1979). Also, the values of concentration factors for Sr and Cs in marine fishes are 3 and 100, respectively (IAEA, 2004).

The Japanese government and Fukushima Prefecture have carried out a monitoring program of radioactive Cs (including ¹³¹I until 1 April 2012) in marine fishes (Fisheries Agency of Japan, 2015a). The monitoring program has led to conclusions that the present contamination levels of radioactive Cs in marine fishes are low (Merz et al., 2015; Okamura et al., 2016). The target radioactive materials in the program did not include radioactive Sr, as the released amount was small, and the analytical method is more complex and time‐consuming than the analytical method for radioactive Cs. Therefore, the number of data for radioactive Sr is much less than that of radioactive Cs.

The levels of ⁹⁰Sr in marine fishes sampled from within the highly polluted FNPP harbor have been reported (Fujimoto et al., 2015). The Tokyo Electric Power Company Holdings, Inc. (TEPCO) has also regularly published data on ⁹⁰Sr concentrations in some marine fishes collected within a 20‐km radius from the FNPP (TEPCO, 2015). However, fewer data are available for ⁹⁰Sr in surveys of commercial species for the reasons described above. Strontium has similar chemical properties to calcium and is, therefore, accumulated in the bones of organisms (Whicker et al., 1972). Therefore, the public has been concerned about the safety of commercial species because of radioactive Sr and need the concentration data ⁹⁰Sr in commercial species.

In this study, we investigated the ⁹⁰Sr concentrations and the ¹³⁷Cs/⁹⁰Sr activity ratios in marine fishes caught off Fukushima and outside of Fukushima, North Pacific Ocean. We also determined the background level of ⁹⁰Sr and ¹³⁷Cs/⁹⁰Sr activity ratios in marine fishes off Japan before the FNPP accident based on the data obtained from the Nuclear Regulation Authority (NRA) database (NRA, 2015).

Materials and Methods

Analysis methods of ⁹⁰Sr and ¹³⁷Cs in marine fishes

The samples were collected commercially or using R/V Soyo‐maru belonging to Japan Fisheries Research and Education Agency (FRA). The sample location and information are shown in Table 1 and Fig. 1. The purpose of this study was to allay public concerns about the safety of commercial species for radioactive Sr. We selected the sample for the low ¹³⁷Cs concentration samples for analysis of the ⁹⁰Sr because the low ¹³⁷Cs concentration samples could be distributed to the market (Table 1). Basically, the samples for ⁹⁰Sr measurement were prepared from the whole body of fish, occasionally the internal organs and the muscles were removed from some samples before mincing. All the samples were dried at 105°C for 48–72 h and carbonized at 320°C for 4 h, 360°C for 4 h and 420°C for 4 h. The carbonized samples were finally ashed at 450–500°C for 48 h. The ash samples were ground to a fine powder and kept in a plastic container at room temperature.

Table 1. Concentrations of ⁹⁰Sr and ¹³⁷Cs in marine fishes caught in the North Pacific Ocean off Japan after the Fukushima Dai‐ichi Nuclear Power Plant (FNPP) accident

Fish species	Sampling date	Sampling location	Locationa^a For location data see Fig. 1.	Fish body parts	^90Sr conc. Bq kg^−1 wet	^137Cs conc. Bq kg^−1 wet
Isurus oxyrinchus	30 Sep. 2014	North Pacific Ocean	42	Muscle	<0.013	0.89 ± 0.022
	30 Sep. 2014	North Pacific Ocean	42	Muscle	<0.010	1.3 ± 0.021
Squalus acanthias	8 Jul. 2014	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	35	Without internal organs	<0.0087	3.0 ± 0.038
Conger japonicus	9 Jul. 2014	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	36	Without internal organs	<0.011	2.7 ± 0.033
Conger myriaster	21 Dec. 2011	Off Fukushima	11	Whole body	0.043 ± 0.0066	13 ± 0.78
	7–9 Jul. 2014	Off Fukushima	‐	Without internal organs	<0.021c^c Average of 30 specimens.	3.0 ± 0.77c^c Average of 30 specimens.
Engraulis japonicus	14 Apr. 2011	Off Chiba	3	Whole body	<0.040	4.3 ± 0.38d^d Concentration of ^137Cs in the muscle.
	26 May 2011	Off Chiba	6	Whole body	<0.030	9.2 ± 0.37d^d Concentration of ^137Cs in the muscle.
	2 Sep. 2012	Off Kochi	22	Whole body	<0.018	<0.043
(Larvae)	14 Jul. 2015	Off Fukushima	46	Whole body	<0.0027	0.21 ± 0.0094
(Larvae)	14 Jul. 2015	Off Fukushima	46	Whole body	<0.0035	0.22 ± 0.013
Etrumeus teres	2 Sep. 2012	Off Kochi	22	Whole body	<0.018	0.082 ± 0.017
Sardinops melanostictus	6 Apr. 2011	Off Chiba	1	Whole body	<0.040	3.2 ± 0.41d^d Concentration of ^137Cs in the muscle.
	22 Jun. 2011	Off Ibaraki	7	Whole body	<0.030	11 ± 0.41d^d Concentration of ^137Cs in the muscle.
	20 Aug. 2012	Off Ibaraki	19	Whole body	<0.013	0.38 ± 0.020
Salangichthys ishikawae	18 Jan. 2012	Off Fukushima	12	Whole body	0.40 ± 0.024	39 ± 0.16
Oncorhynchus keta	1 Nov. 2012	Off Hokkaido	25	Whole body	<0.018	0.12 ± 0.0016
Coryphaenoides acrolepis	6 Aug. 2012	Off Ibaraki	18	Whole body	<0.028	0.44 ± 0.021
Gadus macrocephalus	21 Apr. 2011	Off Fukushima	4	Whole body	0.030 ± 0.0074	19 ± 0.13d^d Concentration of ^137Cs in the muscle.
	9 Jul. 2014	Off Fukushima	37	Without internal organs	<0.011	2.9 ± 0.047
	14 Apr. 2015	Off Aomori	41	Without internal organs	<0.0038	2.6 ± 0.039
	17 Apr. 2015	Off Fukushima	4	Without internal organs	<0.0061	0.28 ± 0.019
Cololabis saira	24 Jun. 2012	North Pacific Ocean	16	Whole body	<0.016	0.78 ± 0.029
	26 Jun. 2014	North Pacific Ocean	34	Without internal organs	<0.0077	0.49 ± 0.014
	4 Sep. 2014	North Pacific Ocean	39	Whole body	<0.010	0.12 ± 0.0075
Beryx splendens	17 Oct. 2012	Off Chiba	24	Without internal organs and muscle	<0.023	1.1 ± 0.041
Sebastes cheni	21 Dec. 2011	Off Fukushima	10	Whole body	1.2 ± 0.052	580 ± 4.9
	26 Dec. 2012	Off Fukushima	17	Whole body	<0.036	33 ± 1.3d^d Concentration of ^137Cs in the muscle.
	11 Sep. 2013	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	31	Without internal organs	0.21 ± 0.0015	51 ± 1.0
	15 Mar. 2015	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	43	Without internal organs	0.043 ± 0.0072	7.1 ± 0.23
	15 Mar. 2015	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	43	Without internal organs	0.049 ± 0.0076	9.7 ± 0.34
Sebastes schlegeli	5 Nov. 2012	Off Hokkaido	26	Without internal organs and muscle	<0.032	1.6 ± 0.054
Lepidotrigla microptera	30 Sep. 2013	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	32	Whole body	<0.015	2.2 ± 0.054
Myoxocephalus stelleri	9 Nov. 2012	Off Hokkaido	26	Without internal organs and muscle	<0.029	0.89 ± 0.035
Coryphaena hippurus	3 Sep. 2012	Off Kochi	22	Without internal organs and muscle	<0.029	0.28 ± 0.031
Trachurus japonicus	29 Aug. 2012	Off Ibaraki	20	Whole body	<0.018	0.94 ± 0.028
Etelis coruscans	15 Nov. 2012	Off Okinawa	27	Without internal organs and muscle	<0.023	0.10 ± 0.026
Pagrus major	1 Sep. 2014	Off Ehime	38	Without internal organs	<0.033	0.41 ± 0.020
	16 Sep. 2014	Off Ehime	38	Without internal organs	<0.033	0.12 ± 0.017
Evynnis tumifrons	24 Nov. 2013	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	33	Whole body	<0.024	2.9 ± 0.12
Ammodytes personatus	8 Apr. 2011	Off Ibaraki	2	Whole body	<0.020	44 ± 0.73d^d Concentration of ^137Cs in the muscle.
	12 Apr. 2011	Off Ibaraki	2	Whole body	<0.030	35 ± 0.92d^d Concentration of ^137Cs in the muscle.
(Larvae)	7 Apr. 2015	Off Fukushima	44	Whole body	<0.0056	0.59 ± 0.023
(Larvae)	7 Apr. 2015	Off Fukushima	44	Whole body	<0.0040	0.57 ± 0.022
(Larvae)	20 Apr. 2015	Off Fukushima	45	Whole body	<0.0045	0.73 ± 0.028
(Larvae)	20 Apr. 2015	Off Fukushima	45	Whole body	<0.0042	0.76 ± 0.028
Scomber japonicus	28 Oct. 2011	Off Ibaraki	9	Whole body	<0.025	5.8 ± 0.041
	12 Dec. 2012	Off Chiba	28	Without internal organs	<0.017	0.30 ± 0.021
Scomber australasicus	1 Jul. 2011	Off Ibaraki	8	Whole body	<0.030	3.9 ± 0.047d^d Concentration of ^137Cs in the muscle.
	21 Dec. 2011	Off Fukushima	11	Whole body	0.030 ± 0.0084	4.2 ± 0.46
	1 Feb. 2012	Off Miyazaki	13	Whole body	<0.015	1.1 ± 0.028
	29 Aug. 2012	Off Ibaraki	21	Without internal organs	<0.013	0.45 ± 0.021
	16 Oct. 2012	Off Ibaraki	23	Whole body	<0.017	0.11 ± 0.029
	28 Sep. 2014	Off Aomori	40	Without internal organs	<0.015	0.13 ± 0.012
	28 Sep. 2014	Off Aomori	41	Without internal organs	<0.016	0.29 ± 0.0091
Paralichthys olivaceus	29 Jul. 2013	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	30	Whole body	0.018 ± 0.0034	5.0 ± 0.19
	29 Jul. 2013	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	29	Whole body	0.016 ± 0.0035	2.7 ± 0.12
	30 Sep. 2013	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	32	Whole body	0.026 ± 0.0065	7.4 ± 0.10
Eopsetta grigorjewi	21 Dec. 2011	Off Fukushima	11	Whole body	0.094 ± 0.012	26 ± 0.13
Hippoglossoides dubius	22 Apr. 2011	Off Ibaraki	5	Whole body	<0.030	4.5 ± 0.033d^d Concentration of ^137Cs in the muscle.
Kareius bicoloratus	24 Nov. 2013	Off Fukushimab^b The coast of Fukushima within a 20 km radius of the (FNPP).	33	Whole body	<0.015	4.8 ± 0.14
Pleuronectes obscurus	5 Nov. 2012	Off Hokkaido	26	Without internal organs and muscle	<0.02	0.060 ± 0.017
Thamnaconus modestus	19 Feb. 2012	Off Ibaraki	14	Whole body	<0.023	3.6 ± 0.44
Sphoeroides pachygaster	21 Feb. 2012	Off Ibaraki	15	Whole body	<0.013	1.7 ± 0.038

• ^a For location data see Fig. 1.
• ^b The coast of Fukushima within a 20 km radius of the (FNPP).
• ^c Average of 30 specimens.
• ^d Concentration of ¹³⁷Cs in the muscle.

Figure 1

Monitoring sites for marine fishes in this study. Closed star shows the location of the Fukushima Dai‐ichi Nuclear Power Plant (FNPP) and the semicircle means the area within 20 km from the FNPP.

Before measuring ⁹⁰Sr, ¹³⁷Cs concentrations in all the samples were measured using a high‐purity germanium semiconductor detector equipped with a multichannel analyzer (MCA‐7600; Seiko EG&G, Chiba, Japan). The counting efficiency was calibrated using standard sources (MX033U8PP; Japan Radioisotope Association, Tokyo, Japan) with different heights (5, 10, 20, 30 and 50 mm). The counting times were more than 7200 s. The concentration of three standard deviations from counting error was defined as the ¹³⁷Cs detection limit.

After the ¹³⁷Cs measurement, ⁹⁰Sr was determined according to the method published by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan (2003), with slight modifications. Approximately 20 g of ash was dissolved in 20 mL of nitric acid using a microwave oven (Multiwave 3000; Anton Paar, Graz, Austria). To remove other ions coexisting with Sr, co‐precipitation with carbonate was done and then the co‐precipitation with oxalate was done. Finally, Sr was separated using a cation exchange resin column (Dowex, 50W X8, 26 × 3 cm i.d.) at a flow rate of 4–6 mL min⁻¹ with wash buffer, 15.4% (w/v) ammonium acetate solution‐methanol (1:1), and elution buffer, 15.4% (w/v) ammonium acetate solution. The co‐precipitation with iron was performed with an addition of 1 mL of 5 mg mL⁻¹ ferric chloride solution to remove ⁹⁰Sr from the sample solution. After standing for 2 weeks to ⁹⁰Y ingrowth, the co‐precipitation with iron was performed again, and newly generated ⁹⁰Y was collected on the filter paper by filtration. The radioactivity of ⁹⁰Y in the pellet was measured with an ultralow background gas flow counter (model LBC‐471‐Q; Aloka Co., Ltd, Tokyo, Japan) for 60 min. The concentration of stable strontium and calcium in ash was measured with an inductively coupled plasma atomic emission spectrometer (ICP‐AES; Optima 7300 DV, Perkin Elmer Corporation, MA, USA). In our study, the recovery rates of the Sr carrier were above 65.4–106%. The concentration of three standard deviations from the counting uncertainties for ⁹⁰Sr was defined as the ⁹⁰Sr detection limit.

Data sources

There is a huge database of radioactive materials in various samples collected in Japan operated by NRA (2015). We extracted the data of ⁹⁰Sr and ¹³⁷Cs concentrations in marine fishes of North Pacific Ocean off Japan collected before the FNPP accident from the NRA database, and used them for estimating the background concentrations of ⁹⁰Sr and ¹³⁷Cs in marine fishes (Fig. 2a,b). We also extracted data after the FNPP accident for visualizing temporal changes with our data (Fig. 3a,b). We referenced the data of ⁹⁰Sr and ¹³⁷Cs concentrations in marine fishes collected within a 20‐km radius from the FNPP reported by the TEPCO (2015), and used them for investigating temporal changes of ⁹⁰Sr and ¹³⁷Cs, estimating the ¹³⁷Cs/⁹⁰Sr activity ratios (Fig. 6), and revealing the source of pollution (Fig. 7).

Figure 2

Concentrations of ⁹⁰Sr (a) and ¹³⁷Cs (b) in marine fishes of the North Pacific Ocean off Japan caught before the Fukushima Dai‐ichi Nuclear Power Plant (FNPP) accident. All the data were obtained from the Nuclear Regulation Authority (NRA) database.

Figure 3

Concentrations of ⁹⁰Sr (a) and ¹³⁷Cs (b) in marine fishes after the Fukushima Dai‐ichi Nuclear Power Plant (FNPP) accident caught in the North Pacific Ocean off Japan. The data of closed diamonds and crosses were obtained in this study and from the Nuclear Regulation Authority (NRA) database (NRA, 2015), respectively. (A) The arrow indicates the upper concentration of the background of ⁹⁰Sr in marine fishes of the North Pacific Ocean off Japan.

Statistical analyses

We conducted all statistical analyses using the R Version 3.2.2 software (R Development Core Team, 2015). Non‐detected samples were excluded from the statistical analyses. The differences in ⁹⁰Sr and ¹³⁷Cs concentrations in marine fishes among the decade were compared using the Steel–Dwass test. The differences in ⁹⁰Sr and ¹³⁷Cs concentrations in marine fishes between before and after the FNPP accident and the differences in ⁹⁰Sr and ¹³⁷Cs concentrations between teleost fishes and elasmobranch fishes were compared using a Welch's two‐sample t‐test. We considered a P‐value <0.05 to be statistically significant.

Results

Background level of ⁹⁰Sr and ¹³⁷Cs in marine fishes of Japan

Figure 2a shows the ⁹⁰Sr concentrations in marine fishes caught in the North Pacific Ocean off Japan before the FNPP accident. The higher concentrations of ⁹⁰Sr in this data set were detected in flathead mullet, Mugil cephalus, collected from the coast off Aichi Prefecture in May 1966 (2.7 ± 0.10 Bq kg⁻¹ wet) and of the coast off Fukushima Prefecture in November 1972 (2.1 ± 0.044 Bq kg⁻¹ wet). We divided the dataset into five periods as below: May 1966 – 10 March 1971, 11 March 1971 – 10 March 1981, 11 March 1981 – 10 March 1991, 11 March 1991 – 10 March – 2001 and 11 March 2001 – 10 March 2011 (the decade just before the FNPP accident). The average concentrations of ⁹⁰Sr in each decade were 0.18 ± 0.50 Bq kg⁻¹ wet (means ± SD, n = 26), 0.067 ± 0.24 Bq kg⁻¹ wet (n = 76), 0.043 ± 0.027 Bq kg⁻¹ wet (n = 26), 0.026 ± 0.021 Bq kg⁻¹ wet (n = 38) and 0.024 ± 0.019 Bq kg⁻¹ wet (n = 24) in chronological order, respectively. The average concentration of ⁹⁰Sr in the earliest period (May 1966 – March 1971) was significantly higher than that of other periods (P < 0.001, Steel–Dwass test). The concentration of ⁹⁰Sr in marine fishes in the past two decades was significantly lower than others (P < 0.05, Steel–Dwass test). The average concentration of ⁹⁰Sr in marine fishes caught in the past two decades was 0.025 ± 0.021 Bq kg⁻¹ wet (n = 62), and therefore we defined this concentration (0.046 Bq kg⁻¹ wet) as the upper background level of ⁹⁰Sr concentration in marine fishes of the North Pacific Ocean off Japan.

Figure 2b shows the ¹³⁷Cs concentrations in marine fishes caught in the North Pacific Ocean off Japan before the FNPP accident. The highest ¹³⁷Cs concentration was also detected in flathead mullet off Aichi caught in May 1966 (1.1 ± 0.086 Bq kg⁻¹ wet). We also divided the data set of ¹³⁷Cs into five periods as above described. The average concentrations of ¹³⁷Cs in the five periods were 0.21 ± 0.18 (n = 30), 0.30 ± 0.17 (n = 207), 0.22 ± 0.12 (n = 304), 0.15 ± 0.068 (n = 475) and 0.10 ± 0.045 (n = 496) in chronological order, respectively. We excluded the data of the earliest period, 1966–1971, because of the low number of data (n = 30) than the other periods. The concentration of ¹³⁷Cs has significantly decreased during the four periods (P < 0.001, Steel–Dwass test). Therefore, we defined the latest period of an average of ¹³⁷Cs concentration, 0.10 ± 0.045 Bq kg⁻¹ wet, as the background level of ¹³⁷Cs in marine fishes of the North Pacific Ocean off Japan.

Concentrations of ⁹⁰Sr in marine fishes after the FNPP accident except for off Fukushima

After the FNPP accident, the concentrations of ⁹⁰Sr and ¹³⁷Cs in marine fishes caught outside of Fukushima were examined in this study during April 2011 – April 2015 (Fig. 3a,b). Figure 3a,b include data obtained from the NRA database (NRA, 2015). We analyzed 37 specimens of 27 fish species, and the ⁹⁰Sr concentration in all samples were under the detection limit (<0.040 Bq kg⁻¹ wet; Fig. 3a). After the FNPP accident, ⁹⁰Sr was detected in only 3 out of 217 specimens in the NRA database (NRA, 2015; Fig. 3a). The concentrations of ⁹⁰Sr ranged from 0.0055 to 0.0098 Bq kg⁻¹ wet, which was within the background level. These results indicated that the FNPP accident had no influence on the ⁹⁰Sr concentration in marine fishes caught outside of Fukushima. In contrast, ¹³⁷Cs concentrations in most samples of this study and the NRA database were higher than the background level (Fig. 3b and Table 1). The concentrations of ¹³⁷Cs in samples of this study caught after the accident were under the detecion limit (<0.043) −44 Bq kg⁻¹ wet. Based on the NRA database, the average concentration of ¹³⁷Cs in marine fishes caught outside of Fukushima was 1.85 ± 4.1 Bq kg⁻¹ wet (n = 250; Fig. 3b), and was significantly higher than that of the background level (P < 0.001, Welch's t‐test).

FNPP‐derived ⁹⁰Sr in marine fishes off Fukushima

We measured ⁹⁰Sr concentrations in 27 specimens collected off Fukushima during April 2011 to September 2015. Strontium‐90 was detected in 12 out of the 27 specimens (Fig. 4a and Table 1). The ⁹⁰Sr concentrations in 4 of 12 specimens were higher than the upper concentration of the background level (0.046 Bq kg⁻¹ wet). The maximum concentration of ⁹⁰Sr in this study was 1.2 ± 0.052 Bq kg⁻¹ wet in the whole body of the Japanese rockfish, Sebastes cheni, collected in December 2011, whereas the concentrations in the whole body without internal organs of the same species collected in September 2013 and March 2015 were 0.21 ± 0.0015 Bq kg⁻¹ wet and 0.049 ± 0.0076 Bq kg⁻¹ wet, respectively. Furthermore, the concentration of ⁹⁰Sr in the Japanese rockfish collected on 26 December 2012 was undetectable (<0.036 Bq kg⁻¹ wet). Also, the concentration of ⁹⁰Sr in the round‐nose flounder, Eopsetta grigorjewi, collected in December 2011 and the Ishikawa icefish, Salangichthys ishikawae, in January 2012 were 0.094 ± 0.012 Bq kg⁻¹ wet and 0.40 ± 0.024 Bq kg⁻¹ wet, respectively. Concentrations of ¹³⁷Cs in all the samples collected off Fukushima ranged from 0.21 to 580 Bq kg⁻¹ wet, obviously higher than that of the background level.

Figure 4

Concentrations of ⁹⁰Sr (a) and ¹³⁷Cs (b) in marine fishes caught off Fukushima after the Fukushima Dai‐ichi Nuclear Power Plant (FNPP) accident. Open circles and closed squares indicate the sample collected off Fukushima and within 20 km from the FNPP obtained in this study, respectively. Open squares indicate marine fishes collected within 20 km from the FNPP obtained from TEPCO website (TEPCO, 2015).

Figure 5a shows the temporal variation of the ⁹⁰Sr concentration in the whole body of the teleost and elasmobranch fishes caught within a 20‐km radius from the FNPP during April 2012 to Jun 2015. The average concentrations of ⁹⁰Sr in the teleost fishes, 0.76 ± 0.94 Bq kg⁻¹ wet (n = 38), were significantly higher than that in the elasmobranch fishes, 0.27 ± 0.19 Bq kg⁻¹ wet (n = 18) (P < 0.01, Welch's t test). There was no correlation between the ⁹⁰Sr concentrations in the teleost fishes and the number of days since the FNPP accident occurred. In contrast, there was a negative correlation between the ⁹⁰Sr concentrations in the elasmobranch fishes and the number of days since the FNPP accident occurred (P < 0.01, Fig. 5a). Figure 5b shows the temporal variation of ¹³⁷Cs concentration in the muscle of the teleost fishes and the elasmobranch fishes. There was no significant difference between concentrations of ¹³⁷Cs in the teleost fishes, 390 ± 350 Bq kg⁻¹ wet (n = 38), and that in the elasmobranch fishes, 240 ± 230 Bq kg⁻¹ wet (n = 18). The concentrations of ¹³⁷Cs in the muscle of both the teleost and elasmobranch fishes declined since the FNPP accident occurred (P < 0.001, Fig. 5b).

Figure 5

Temporal variation of ⁹⁰Sr (a) or ¹³⁷Cs (b) concentrations in the teleost and elasmobranch fishes. The data set of samples caught within 20 km from the FNPP obtained from the TEPCO website (TEPCO, 2015). Open and gray squares indicate the teleost and elasmobranch fishes, respectively. The black and the gray dashed lines indicate the regression lines for the data of the teleost fishes and the elasmobranch fishes, respectively.

Activity ratios of ¹³⁷Cs to ⁹⁰Sr in marine fishes

We used the data obtained from the NRA database to estimate the background ratio of ¹³⁷Cs to ⁹⁰Sr in marine fishes off the North Pacific Ocean off Japan. The data were chosen as follows: the data were obtained within the past decade before the FNPP accident and had both the detected ⁹⁰Sr and ¹³⁷Cs concentrations. The ¹³⁷Cs/⁹⁰Sr activity ratios before the FNPP accident ranged from 2.6 to 18 (whole body; 13 specimens of Japanese icefish, Salangichthys microdon; 3 of Japanese anchovy, Engraulis japonicus; 1 of bananafish, Pterocaesio digramma; 1 of brown sole, Pleuronectes herzensteini), 7.7–25 (muscle; 1 of Alaska pollock, Theragra chalcogramma; 1 of fat greenling, Hexagrammos otakii) and 0.90 (¹³⁷Cs muscle/⁹⁰Sr bone; 2 of red seabream, Pagrus major). Because the concentration of ¹³⁷Cs in marine fishes markedly increased after the accident, the ¹³⁷Cs/⁹⁰Sr activity ratios of teleost fishes ranged from 98 to 480 (whole body; n = 9), 79 to 5800 (¹³⁷Cs muscle/⁹⁰Sr whole body; n = 38) and 140 to 240 (whole body without internal organs; n = 4) (Fig. 6). The elasmobranch fishes of ¹³⁷Cs/⁹⁰Sr activity ratios were 150–5600 (¹³⁷Cs muscle/⁹⁰Sr whole body; n = 18). Focusing on the teleost fishes, the ¹³⁷Cs/⁹⁰Sr activity ratios in the whole body, those in the whole body without internal organs, and the ¹³⁷Cs in muscle/⁹⁰Sr in the whole body caught off Fukushima were 5–190, 8–90, and 4–2200 times higher, respectively, than the background of ¹³⁷Cs/⁹⁰Sr activity ratio in the whole body. The ¹³⁷Cs in muscle/⁹⁰Sr in whole body for the teleost have now approximately the background level (R = 0.748, P < 0.001, Fig. 6).

Figure 6

The ¹³⁷Cs/⁹⁰Sr activity ratios in marine fishes collected off Fukushima after the Fukushima Dai‐ichi Nuclear Power Plant (FNPP) accident. Open circles and triangles indicate the ¹³⁷Cs/⁹⁰Sr activity ratios in whole body and in whole body without internal organs, respectively. Open and gray squares indicate the ¹³⁷Cs in muscle/⁹⁰Sr in whole body. Open and gray squares indicate the teleost and elasmobranch fishes, respectively, which were obtained from the TEPCO website (TEPCO, 2015). The black and the dashed lines indicate the regression lines for the ¹³⁷Cs/⁹⁰Sr activity ratio of the teleost fishes and of both teleost and elasmobranch fishes, respectively. Range of lines mean the ¹³⁷Cs/⁹⁰Sr activity ratio in marine fishes caught past decade before the accident. They indicate ¹³⁷Cs in muscle/⁹⁰Sr in bone (*1), and ¹³⁷Cs/⁹⁰Sr in muscle (*2), and ¹³⁷Cs/⁹⁰Sr in whole body (*3) from left to right.

Discussion

The upper background level of ⁹⁰Sr and ¹³⁷Cs in marine fishes off the North Pacific Ocean off Japan were determined to be 0.046 Bq kg⁻¹ wet and 0.15 Bq kg⁻¹ wet, respectively, based on the NRA database (NRA, 2015). Although the concentrations of ⁹⁰Sr in muscle or bone of marine fishes off the North Pacific Ocean and the Bering Sea were undetectable owing to the high level of the detection limit, 2.1 Bq kg⁻¹ wet (Burger et al., 2007), those in the whole body of flatfish off the coast of Korea were estimated to be 0.0021–0.013 Bq kg⁻¹ wet (Yang et al., 2002), which fell within our determined background of the ⁹⁰Sr concentration. It is difficult to determine the background level of ⁹⁰Sr in marine fishes of other oceans because the reports are so few. In contrast, there are many reports of the ⁹⁰Sr concentrations in freshwater fishes affected by the Chernobyl NPP accident. Stronium‐90 concentrations were estimated to be 2.3–240 Bq kg⁻¹ wet in the edible part of perch, Perca fluviatilis, collected in the lakes in southern and central Finland from 1987 to 1997 (Outola et al., 2009). As far as we know, the maximum ⁹⁰Sr concentration in fish influenced by the FNPP accident was 170 ± 1.2 Bq kg⁻¹ wet in the whole body without internal organs of the Japanese rockfish, Sebastes cheni, of which the ¹³⁷Cs concentration in muscle was 34 500 ± 250 Bq kg⁻¹ wet, collected in the FNPP harbor on 12 February 2012 (Fujimoto et al., 2015). Based on the NRA database, the highest concentrations of ⁹⁰Sr before the FNPP accident were detected, 2.7 ± 0.10 Bq kg⁻¹ wet, in the flathead mullet collected from the coast of Aichi in May 1966. This source was thought to be from global fallout after the atmospheric nuclear weapon test from 1945 to 1980 (Bowen et al., 1980; Morita et al., 2010). In Japan, the global fallout also increased the ⁹⁰Sr concentration in freshwater fishes more than the Chernobyl NPP accident. In the 1960s to 1970s, ⁹⁰Sr and ¹³⁷Cs were detected in many freshwater fishes of Japan, and the average values of ⁹⁰Sr and ¹³⁷Cs concentrations were 4.0 ± 3.9 Bq kg⁻¹ wet (n = 116) and 0.43 ± 0.31 Bq kg⁻¹ wet, respectively (n = 128) (NRA, 2015).

In this study, we could not detect ⁹⁰Sr in 37 specimens of 27 fish species collected outside of the Fukushima Prefecture (detection limit, <0.040 Bq kg⁻¹ wet), whereas higher ¹³⁷Cs concentrations than the upper background concentration were detected from 62 of 64 specimens. Strontium‐90 is easier to remove from surface water than ¹³⁷Cs, and consequently the ⁹⁰Sr derived from the FNPP in seawater spread across the ocean and immediately diluted to the background level (Povinec et al., 2012). Therefore, the FNPP‐derived ⁹⁰Sr would not be detected in marine fishes caught outside of Fukushima.

Strontium‐90 was detected in 11 out of 26 specimens, and all of these detected specimens were collected from the southern area off Fukushima, whereas ⁹⁰Sr could not be detected in the Conger japonicus specimen from location 36 in the northern area within a 20‐km radius of FNPP (Fig. 1 and Table 1). The FNPP‐derived radionuclides including ⁹⁰Sr in seawater samples were higher in southern than in northern areas of Fukushima owing to influence by the Oyashio current (Casacuberta et al., 2013). This current also influenced the ¹³⁷Cs contamination of marine organism and sediment (Wada et al., 2013; Ambe et al., 2014).

The maximum concentration of ⁹⁰Sr in this study was 1.2 ± 0.052 Bq kg⁻¹ wet in Japanese rockfish collected on 26 December 2011 (location 10). On 4 December 2011, TEPCO announced that a leakage of approximately 150 L of the treated water from the evaporative concentration apparatus, which removes the radioactive Cs from the contaminated water, was found. Just after the leak, the ⁹⁰Sr concentration in the surface water sampled near the southern discharge channel of the FNPP was 400 Bq L⁻¹ on 5 December 2011, although it decreased to 9.6 Bq L⁻¹ on 10 December 2011, and 0.45 Bq L⁻¹ on 24 December 2011 (TEPCO, 2012). Strontium‐89 was also included in the wastewater. The concentrations of ⁸⁹Sr in the surface water on each of the sampling dates were 140 and 2.5 Bq L⁻¹, and not detected, respectively. Strontium‐89 was successfully detected in Japanese rockfish with 1.2 Bq kg⁻¹ wet for ⁹⁰Sr, and the concentration was 0.25 ± 0.059 Bq kg⁻¹ wet (Fisheries Agency of Japan, 2015a). The ⁸⁹Sr/⁹⁰Sr activity ratio was 0.21 ± 0.050, in good agreement with the activity ratio (0.26–0.35) in the surface seawater near the southern discharge channel of the FNPP in December 2011. However, Fig. 7 shows that the Japanese rockfish was not influenced a great deal by the wastewater release in December 2011. We also detected the ⁹⁰Sr concentration over the background level only in the Japanese rockfish in September 2013. In August 2013, the wastewater including high levels of ⁹⁰Sr was leaked from a tank of FNPP, although TEPCO did not announce any input of β‐emitters in contaminated water from the tank into the sea (TEPCO, 2014). Although we measured ⁹⁰Sr in 4 species totaling 10 specimens in 2015, only 2 specimens of the Japanese rockfish were detectable for ⁹⁰Sr and the concentration was the same as the background level. The migration area of the Japanese rockfish is known to be limited showing a largely territorial and sedentary behavior (Utagawa and Taniuchi, 1999). Thus, the leaked wastewater from the FNPP would directly affect the local area where the Japanese rockfish inhabited. For instance, the otolith of the Japanese rockfish is useful for detecting ⁹⁰Sr contamination from the harbor of the FNPP (Fujimoto et al., 2015). Our results suggested that this fish species might be a good indicator for ⁹⁰Sr contamination.

Figure 7

The relationship between the ⁹⁰Sr and ¹³⁷Cs concentrations in marine fishes. The lines show the ¹³⁷Cs/⁹⁰Sr activity ratios in marine fishes predicted from the ¹³⁷Cs/⁹⁰Sr activity ratios in seawater (FNPP fallout = 100; the wastewater release spring 2011 = 39; the wastewater release December 2011 = 0.016) (Castrillejo et al., 2016) and each concentration factor (Cs: Sr = 100: 3). Red, green, and blue lines and values mean the predicted lines and ratios from the ratio of Fukushima Dai‐ichi Nuclear Power Plant (FNPP) fall out, wastewater release spring 2011, and wastewater release December 2011, respectively. Open circles indicate the fish data.

In the data of TEPCO and our study, the maximum concentration of ⁹⁰Sr in fish outside of the FNPP harbor was 6.0 Bq/kg‐wet in the marbled sole, Pleuronectes yokohamae, of which ¹³⁷Cs concentration in muscle was 1100 Bq kg⁻¹ wet, collected at 3 km from the FNPP on 13 December 2012 (TEPCO, 2015). In Fig. 7, the plot of the marbled sole clearly deviated from the levels observed in other samples. Also, the concentration of ⁹⁰Sr in it also deviated from the others whereas the concentration of ¹³⁷Cs in the marbled sole was not different from that in others collected 650 days after the FNPP accident (Fig. 4a,b). Therefore, the marbled sole could be influenced by the treated water that leaked in December 2011.

The public has been concerned about the safety of commercial species for ⁹⁰Sr, because ⁹⁰Sr tends to persist in the bones of organisms for a long time (Whicker et al., 1972) and is not a target radioactive material in the monitoring program in Japan. Also, it was pointed out to be the presence of the ongoing input of ¹³⁷Cs from the FNPP (Kanda, 2013). It was reported that the ongoing inputs of ⁹⁰Sr from the FNPP releases would be in order of 2.3–8.5 GB day⁻¹ in September 2013 (Castrillejo et al., 2016). However, our results showed no influence in fish collected outside of the FNPP harbor by the ongoing inputs of ⁹⁰Sr from the FNPP. Figures 4a and 5a show the ⁹⁰Sr concentrations were relatively constant in the teleost fishes and a decreasing tendency in the elasmobranch fishes but did not show any increasing trend in both groups of fishes.

The results indicate that the elasmobranch fishes might have a lower concentration of ⁹⁰Sr than the teleost fishes (Fig. 5a). The skeleton of the elasmobranch fish is made of cartilaginous material, and that of the teleost fish is mainly made of bone tissue, meaning that the calcium concentration in elasmobranch fishes is lower than that in teleost fishes. However, only 30% of ⁹⁰Sr in the whole body of teleost fish accumulates in bone (Johansen et al., 2015).

It was difficult to assess the ⁹⁰Sr contamination precisely to marine organism owing to the FNPP accident based on the concentration because TEPCO selected the ⁹⁰Sr analysis samples from samples with relatively high concentrations of radioactive Cs in the sampling. However, overall, the concentrations of ⁹⁰Sr in fish were lower than those of ¹³⁷Cs. The activity ratio of ¹³⁷Cs/⁹⁰Sr has been used to obtain evidence of the origin of these radionuclides (Igarashi et al., 2001). This study showed that the ratios of ¹³⁷Cs/⁹⁰Sr in marine fishes ranged from 71 to 5900 after the FNPP accident. The activity ratios were clearly higher than that before the accident. There is a large difference between the value of the concentration factor for Sr, 3, and that for Cs,100, in marine fishes (IAEA, 2004). After the FNPP accident, the ¹³⁷Cs/⁹⁰Sr activity ratios in marine fishes were similar to the predicted relationship between the ⁹⁰Sr and ¹³⁷Cs concentration of the wastewater release in spring 2011 considering each concentration factor (Fig. 7, Castrillejo et al., 2016). Our results clearly showed that the source of detectable ⁹⁰Sr over background was the FNPP. The ¹³⁷Cs in muscle/⁹⁰Sr in the whole body for the teleost fishes are now approximately the background level (R = 0.748, P < 0.001, Fig. 6). The contaminated water with a lower ¹³⁷Cs/⁹⁰Sr ratio and ⁹⁰Sr concentration than those immediately after the accident continued to be released (Castrillejo et al., 2016). However, the decreasing ¹³⁷Cs/⁹⁰Sr activity ratios in this study was not as a result of the contaminated water with a lower ¹³⁷Cs/⁹⁰Sr, because the ⁹⁰Sr concentration in marine fish did not tend to decrease (Fig. 4). Recently in TEPCO's report announced on 17 December 2015, ¹³⁷Cs concentrations in marine fishes caught within 20 km from the FNPP were less than 70 Bq kg⁻¹ wet (TEPCO, 2015). Therefore, the decreasing ¹³⁷Cs/⁹⁰Sr activity ratios in this study would mainly depend on the decrease in the ¹³⁷Cs concentration in marine fish.

Since 1 April 2012, the Japanese standard limits of radioactive Cs in foods are 100 Bq kg⁻¹ wet (MHLF, 2015). These limits set radioactive Cs as the representative radionuclide considering plutonium, ⁹⁰Sr, and ruthenium (Ru)‐106. The limits for commercial species were calculated based on the assumption that the effective dose from radioactive Cs is the same as the effective dose from other radionuclides. Plutonium and ¹⁰⁶Ru from the FNPP accident were hardly detected in seawater, marine sediments or marine organisms (Buesseler et al., 2012; Bu et al., 2014; Steinhauser, 2014; Fisheries Agency of Japan, 2015b; JAEA, 2015). In the FNPP accident, two kinds of radioactive Cs were assumed to be released to the environment with the amount released approximately equal to each other (Chino et al., 2011). Therefore, hypothesizing the presence of only ⁹⁰Sr and radioactive Cs in commercial species, the activity ratio of ¹³⁷Cs/⁹⁰Sr assumed in the standard limits was below 2.15. Our results clearly show that the ratio observed in marine organisms was above 2.15, indicating that the hypotheses in the standard limits were sufficiently verified.

In conclusion, the concentrations of ⁹⁰Sr in all the samples collected outside of Fukushima were under the detection limits (<0.040 Bq kg⁻¹ wet). Although we detected ⁹⁰Sr in marine fishes collected off Fukushima, a small number of species were higher than the background level (>0.046 Bq kg⁻¹ wet). The maximum concentration of ⁹⁰Sr in fish collected outside of the FNPP harbor was 6.0 Bq kg⁻¹ in December 2012. Our results indicated that the influence of the FNPP accident on ⁹⁰Sr in marine fishes was limited to the area near the FNPP and can be considered as negligible.

Acknowledgements

We appreciate the captain and crew of the R/V Soyo‐maru belonging to Japan Fisheries Research and Education Agency (FRA) for collecting the samples after the FNPP accident. We would like to thank the fishery workers in Fukushima who caught marine fishes after the FNPP accident. We would also like to thank the staff of the Radioecology Group, a research center for fisheries oceanography and marine ecosystem, National Research Institute of Fisheries Science, FRA. This study was supported financially by the Fisheries Agency of Japan.