Contamination of sea urchin Mesocentrotus nudus by radiocesium released during the Fukushima Daiichi Nuclear Power Plant accident.

Countless marine organisms were polluted with radioactive materials that were dispersed when the Fukushima Daiichi Nuclear Power Plant (FDNPP) was damaged in 2011 by the Great East Japan Earthquake. The aim of this study was to determine the degree to which marine herbivorous sea urchins, Mesocentrotus nudus, were contaminated with radiocesium because of the accident. We collected samples of sea urchins from four locations in Fukushima prefecture (at the coast and offshore from the Yotsukura and Ena stations) and investigated how the 137Cs activity concentrations changed. The biological half-life (Tbio) of 137Cs in the individual sea urchins was between 121 and 157 days. The ecological half-life (Teco) of 137Cs was 181-423 days and was high in places close to the FDNPP. The Teco values in the sea urchins were longer than previously reported. The results infer that the food sources of the sea urchins around the Fukushima coast strongly influenced their uptake of 137Cs.

Introduction

Massive quantities of radioactive materials leaked into the atmosphere after the Fukushima Daiichi Nuclear Power Plant (FDNPP) accident in Fukushima Prefecture, Japan, operated by the Tokyo Electric Power Company (TEPCO), sustained severe damage because of the Great East Japan Earthquake (Mw 9.0) and large tsunami occurred on 11 March 2011. About 80% (3.5–27 PBq) of the radioactive cesium (Cs) released into the atmosphere settled in the ocean [1, 2]. Radioactive Cs amounting to 3.5 PBq was transported directly out into the ocean [1]. The water mass with high levels of radioactive cesium flowed south from the FDNPP [1, 3–5] and contaminated many marine biota and seabed sediments along the southern coast and offshore area of Fukushima Prefecture [6-12]. Numerous researchers have observed temporal change in the radiocesium contamination of various marine organisms, including fish [13-15], invertebrates [16, 17], and algae [9, 18]. From monitoring data, these researchers showed that the concentrations tended to decrease over time, and that the radioactive concentrations in marine invertebrates decreased more quickly, such that the concentrations in many species were below the level of detection within three years of the FDNPP accident.

The variations in radioactive Cs concentrations in seawater, food, excreted material, and seafloor sediment (including reefs) need to be considered when analyzing the contamination of benthic invertebrates. In addition, researchers have examined how ingested contaminated sediments affected the digestive tract [7, 14, 17]. Elsewhere, researchers have reported that the uptake of radioactive Cs in benthic invertebrates depends on species-specific feeding habits, even when the surrounding sediments are highly contaminated [19]. Given the range of topics and findings so far, further research that includes food and the sediment environment is needed to clarify the mechanisms by which benthic organisms are contaminated by radioactive cesium.

Mesocentrotus nudus, a sea urchin, was an important fishery product in the Fukushima coastal area before the FDNPP accident. Monitoring data showed that the 137Cs activity concentration in M. nudus reached about 1000 Bq/kg-WW after the accident, then decreased gradually until it was below the detection limit (5.4–8.5 Bq/kg-WW) in 2015 [14, 20]. However, some studies reported that the decreasing trend in radiocesium concentrations in Echinoidea was not universal, and that the radiocesium concentrations in some species of this class, such as sea urchins, were relatively high [16, 17].

Echinoderms are primary consumers in ecosystems and may ingest food that is contaminated by radioactivity. We know that M. nudus mainly feed on large brown algae, but may be omnivorous, depending on the season [21]. It would be useful to have information about how the levels of radioactive materials in living organisms and their discharges changed in the period after the accident therefore, to understand whether radioactive materials have been transferred from food to primary consumers since the FDNPP accident.

The purpose of this study was to determine how the 137Cs contamination of the primary consumer sea urchin M. nudus from the FDNPP accident changed over time. To achieve this, the biological half-life (Tbio) and ecological half-life (Teco) of radioactive cesium in M. nudus were investigated over the period from 426–2726 days after the accident. We also examined whether M. nudus consumed food with high 137Cs activity concentrations. Through these investigations, we examined how radioactive Cs was transferred to primary consumers after the FDNPP accident.

Materials and methods

Sample collection and rearing the sea urchin

We estimated the biological half-life (Tbio) of 137Cs in sea urchins reared in laboratory conditions. Live M. nudus samples were collected from the Ena rocky reef fishing grounds (36.9413°N, 140.9481°E) in the northern part of Iwaki City, Fukushima. Once collected, the samples were immediately transported to the laboratory at the Fukushima Prefectural Fisheries Experimental Station (Shimokajiro, Iwaki City, Fukushima). At the laboratory, the sea urchins were cultured in an aquarium for 7 days to ensure the contents of the digestive tract were excreted before starting the experiment. The process to rear 7 sea urchin individuals was started on 31 May 2013. The rearing period was from May 2013 to August 2013, and 5 sea urchin samples were successfully reared in laboratory and survived for between 77 and 91 days. The 200-L rectangular rearing tank was divided into 6 sections, and one sea urchin was placed in each section. Each section measured 255 × 203 × 382 mm, and was filled to a depth of 300 mm with seawater that was pumped from the coast of Iwaki City. Kaeriyama (2017) reported the 137Cs activity concentrations of seawater at Onahama Port, Iwaki City and reported that the average concentration of 137Cs in seawater was 0.023±0.007 Bq /L during the sea urchin rearing period (S1 Fig) [22]. The seawater was filtered through sand and then poured into the tanks at a rate of 20–30 L/min. The sea urchins were fed until satiety with dried blades of kelp Laminaria spp., brought from Hokkaido, Northern Japan. No 137Cs was detected in the kelp.

The live sea urchin individuals were picked out from the rearing tank every week. On removal, the sea urchins were dried to remove any water, left in the air for 1 hour to drain as much of the internal water as possible, and then weighed. After measuring the body weight, each sea urchin was transferred into a 500 mL sealed container with a 90-mm inner diameter to measure the 137Cs counts. The shell length of the sea urchins on the first day of the experiment was 55.4±2.5 mm (S1 Table), and the individuals were almost the same size. The body weights increased and decreased during the experiment, but the shell length did not change significantly. One individual sea urchin was placed in a specimen bottle, so that the center of the bottle can be positioned in the center of the detector for the measurements. The 137Cs counts in the sea urchins was determined using gamma spectrometry in a closed-end coaxial high-purity germanium (HPGe) semiconductor detector (GEM20P4-70; Ortec, Tennessee, USA) at the Fukushima Prefectural Fisheries Experimental Station. The counting efficiency calibration of the HPGe detector was checked using volume standard sources (MX033U8PP for a 500 mL cylindrical container, Japan Radioisotope Assoc., Tokyo, Japan). The gamma rays from 137Cs were analyzed from the respective peaks in the energy spectrum for 137Cs (662 keV). Each sample weighed 0.063–0.086 kg and the measurement time for each sample was set at 5000 s. The radioactivity of a live sea urchin was measured weekly and the 137Cs counts represented the 137Cs activity [23].

The Tbio in days for each individual sea urchin was estimated from the total number of 137Cs counts in the 137Cs energy region (662 keV). We started to rear the sea urchins on 31 May 2013 (t = 0). The initial 137Cs counts (C0) and depuration rate constant (λ) were calculated from the exponential functional model in Eq (1) and the effective half-life (Teff) of 137Cs in the sea urchins was calculated from Eq (2). The Tbio of 137Cs in the live sea urchins was calculated from the Teff and the physical half-life (Tp) of 137Cs with Eq (3). where t denotes the number of elapsed days, Ct represents the 137Cs counts at elapsed day t, and C0 represents the initial 137Cs counts. λ is the depuration rate constant that allows the effective half-life (Teff) to be calculated (2). The Tbio was calculated from the Teff and the Tp (3). The Tp of 137Cs is 11013 days.

Radiocesium concentrations of wild samples

To measure the ecological half-life (Teco) of radiocesium (137Cs and 134Cs) in wild sea urchins (M. nudus), live samples were collected from 4 locations in Iwaki City, Fukushima Prefecture, that were between 35 and 50 km from the FDNPP. The samples were collected from different water depths at the coast and offshore, to permit comparison between shallow and deep areas. The sampling sites were Yotsukura coast (37.112°N, 140.995°E, 35 km south of the FDNPP, 1–2 m depth), Yotsukura offshore (37.101°N, 141.038°E, 25–35 m depth), Ena rocky coast (36.971°N, 140.958°E, 50 km south of the FDNPP, 6 m depth), and Ena rocky offshore (36.9413°N, 140.9481°E, 8–16 m depth) (Fig 1).

Fig 1

Sea urchin sampling sites on the Fukushima coast.

The Yotsukura and Ena stations are about 35 and 50 km south of the Fukushima Daiichi Nuclear Power Plant (FDNPP), respectively.

Sea urchin samples were collected on 12 occasions from 20 July 2012 until 29 May 2014 (426–1175 days after the FDNPP accident), in May, July, October, November, and December of 2012; in January, February, May, and October of 2013; and in January, February, and May of 2014. Further samples were collected from the Yotsukura coast on 27 August 2018. The details of the sampling program are provided in S2 Table. Consent and special permission to collect wild samples (including samples for laboratory experiments) were gained from the Iwaki City Fisheries Cooperative Association and Fukushima Prefecture, respectively.

Once collected, the sea urchin samples were immediately transported to the laboratory. The sea urchin samples were washed with seawater and many of them were dissected to extract the gonads. To examine the contamination pattern, the samples were divided into two categories, namely whole body and gonads. About 20 individual sea urchins/gonads were crushed and collected in one 100-mL plastic container (U-8, AS ONE, Inc., Osaka), and the weights of the individual samples were measured (23 whole body and 63 gonad samples). All the samples were kept at −18°C until the radioactivity was measured (N = 86).

The radiocesium (137Cs and 134Cs) concentrations in the whole body and gonad samples (86 samples) were measured using a germanium (Ge) semiconductor detector (GEM20-70, SEIKO EG&G CO. LTD, Tokyo; resolution of <2.0 keV/1.33 MeV). The detector efficiency was calibrated with volume radioactivity standard gamma sources (MX033U8PP, Japan Radioisotope Association, Tokyo, Japan). Gamma rays from 137Cs and 134Cs were analyzed from the relevant peaks in the energy spectrum for 137Cs (662 keV) and 134Cs (605 keV and 796 keV). The samples weighed from 0.022 to 0.167 kg and the measurement time was set to 7200 s per sample. The radiocesium concentrations in sea urchin whole body and gonad samples were decay-corrected to the sample collection date and were presented as Bq/kg-wet weight (WW) [23]. Sea urchin samples in which cesium was not detected (n = 4) were dried to remove the moisture and were further analyzed in a well-type coaxial germanium detector (EGPC 250-P21, Canberra France S.A.S, less than 2.1 KeV/1.33 MeV of resolution) for measurement times of 113, 960–229, 940 s per sample.

The Teco values of radiocesium (137Cs and 134Cs) in the sea urchins were calculated from the concentrations measured by the Ge semiconductor detector. The FDNPP accident occurred on 11 March 2011 (t = 0). The initial radiocesium concentration (C0) and the decay rate constant (λeco) were calculated from an exponential functional model (Eq 1). The Teff and the Teco were calculated using Eqs (2) and (4), respectively. Any data below the detection limit (ND) were excluded from the Teco calculations. The detection limit of the concentration was defined from the counting statistics as the concentration of three times the standard deviation. where t denotes the number of days since the FDNPP accident; C0 and Ct are the initial radiocesium concentrations on 11 March 2011 (t = 0) and the concentrations after t days, respectively. λ denotes the decay rate constant. 137Cs and 134Cs had T values of 11013 and 752.63 days, respectively.

Apparent concentration factor (ACF)

The apparent concentration factor (ACF) was determined from data collected by the NRA for the 137Cs activity concentrations in seawater (https://www.nsr.go.jp/english/index.html) from 2012–2014, and data from August 2018 during the sea urchin sampling periods. Sampling points of seawater measured by the NRA were Yotsukura station (T-12) and Ena station (T-20). Seawater samples were collected to a depth of about 3 m below the sea surface.

Statistical analysis

A generalized linear model (GLM) was used to determine the effects of the specific growth rate (hereinafter; SGR, % day−1) and elapsed days from the first measurement (31 May 2013) on the reduction rate of the Cs counts (D, day−1) in sea urchins during the rearing period. The sample ID (A–E) and SGR of the sea urchins and elapsed days were used as explanatory variables, and D in the sea urchins was the response variable in the GLM analysis. A normal distribution was applied to the response variable (D) and the best model was selected with the lowest Akaike information criterion (AIC). Moreover, a null model was built using the response variable only. The effect of the explanatory variables was determined by a maximum likelihood test with the best model and the null model. The correlations between the radiocesium concentrations in the sea urchins and elapsed days, and the ACF and elapsed days from the FDNPP accident, were determined with the Spearman’s rank test. Analysis of variance (ANOVA) was performed to demonstrate the spatial variation in the 137Cs activity concentrations in wild sea urchins. Different trends in the 137Cs counts in the rearing experiment and the 137Cs activity concentrations of the samples collected from the Yotsukura and Ena sites were tested by analysis of covariance (ANCOVA). Since the Tbio was calculated from the 137Cs counts and the Teco was calculated from the 137Cs activity concentrations, the relative value of the ANCOVA test was the tendency to decrease. All the statistical analyses were performed using Microsoft Office Excel 365 (Microsoft, USA) and JMP software (JMP 13, SAS Institute Inc. Cary, NC, USA).

Results

Biological half-life of 137Cs in sea urchins

The 137Cs counts were measured in five sea urchins during their rearing period (0–91 days) in the laboratory. The 137Cs counts in the sea urchins were between 29 and 44 at the start of the rearing period, and then decreased gradually until day 77–91 (Fig 2). We plotted the best fit exponential trend lines for the 137Cs counts in the five sea urchins (A–E) and listed the corresponding probabilities (p) in Table 1. The decreasing slopes of the 137Cs counts over time were significant in two of the five sea urchins (Pearson’s correlation coefficients of −0.601 for sample C (p < 0.05) and −0.581 for sample D (p < 0.05)). The decreases in the 137Cs counts for samples A, B, and E were not significant (p = 0.220, 0.324 and 0.218, respectively).

Fig 2

Temporal changes in the 137Cs counts in five sea urchins during the rearing period (31 May–30 August 2013) in laboratory conditions.

The solid and dashed lines represent statistically significant and insignificant regression slopes, respectively.

Table 1

Summary statistics for the regression slopes of the 137 Cs radioactivity (total number of 137 Cs counts), effective half-life (Teff, day), and biological half-life (Tb, day) of all 5 individuals.

Sample ID	Coefficient of determination (R2)	p-value	Coefficienta		Effective half-life, Tₑff (day)	Biological half-life, Tbio (day)
			C0 (Counts)	λbio (d-1)
A	0.12	0.220	42.80	-0.00243	-	-
B	0.06	0.324	22.98	-0.01617	-	-
C	0.39	0.023*	31.30	-0.00447	155	157
D	0.34	0.038*	31.27	-0.00580	120	121
E	0.17	0.218	37.76	-0.00386	-	-
All samples	0.12	0.007*	32.78	-0.00389	178	181

*Statistical significance (p < 0.05) of the decreasing trend in the data for the reared sea urchin individuals.

aCoefficient for the following exponential equation for the measured data: Ct = C0e(-biot). Ct represents the 137Cs counts at elapsed day t. C0 represents the 137Cs counts at the time of the first measurement (31 May 2013) and λbio is the depuration rate constant.

The Tbio of 137Cs in sea urchin samples C and D that were reared for about 3 months in the laboratory were estimated as 157 and 121 days, respectively (Table 1). The 137Cs counts of samples A–E were significantly correlated with the elapsed days and plotted along a single trend line (r = −0.326 and p < 0.05). We plotted the 5 samples on a single plot and estimated the Tbio of 137Cs as 181 days (Fig 2).

The body weights (g) of the individual sea urchins were plotted against the rearing period (S2 Fig). There were significant correlations observed between the body weights of samples A, B, and D (p < 0.05) and the rearing period, but not between the body weights of samples C and E and the rearing period (p > 0.05). The specific growth rate (SGR) of the sea urchins was calculated as the percentage per day over a given time interval using the equation of Mahmud et al. (2016) [24], as follows: where W2 is the live body weight (g) at day d2, and W1 is the live body weight (g) at day d1.

The reduction rate in the 137Cs counts (D, day−1) was calculated with the formula used by Matsumoto et al. (2015) [15], as follows: where Cn and Cn+1 were the 137Cs counts for the nth and nth + 1 measurements, respectively, and dn and dn+1 were the number of days that elapsed between the nth and nth + 1 measurements when the initial time was set to 0 days.

The results from the generalized linear model (GLM) showed that the specific growth rate (SGR) of the sea urchins was significantly related to the rate of decrease in the 137Cs counts (D, day−1). The sample ID (A–E), elapsed days, and the SGR (%) per day were used as the explanatory variables in the GLM and the Akaike information criterion (AIC) was used to select the best model. The 137Cs counts reduction rate (D) per day in the sea urchins was the response variable. The result of a maximum likelihood test showed that the SGR of the sea urchins had a significant effect on D (ΔAIC = 1.85, χ2 = 8.72, p < 0.05) (S3 Table).

In S2 Fig, the relationships between the body weight and the rearing period of the individual sea urchin samples were depicted, and the growth tendency of only one sample (D) showed an opposite trend, and so had little effect on the GLM results. We used the data from all the samples in the GLM model to explore how the SGR affected the reduction rate (D) in the Cs counts during the rearing experiment. The GLM model showed that the 137Cs counts decreased significantly as the body growth of the sea urchin increased.

Ecological half-life of radiocesium in the sea urchins

We determined the concentrations of radiocesium in sea urchins from four fishing areas (Yotsukura coast, Yotsukura offshore, Ena rocky coast, and Ena rocky offshore) near the FDNPP in Fukushima prefecture. As shown in Fig 3, the 137Cs and 134Cs activity concentrations (converted to log scale) in the sea urchin gonads tended to decrease over time (426–1175 days after the FDNPP accident). Between 500 and 600 days after the FDNPP accident in 2012, the radiocesium (137Cs and 134Cs) activity concentrations in the gonads of sea urchins collected from the Yotsukura coastal area peaked at 139 and 105 Bq/kg-WW, respectively. The average 137Cs activity concentrations in the gonads of sea urchins collected from the Yotsukura coast, Yotsukura offshore, Ena rocky coast, and Ena rocky offshore area in 2013 and 2014 were 14.3, 56.5, 9.19, and 28.4 Bq/kg-WW, and 6.41, 71.1, 7.24, and 34.3 Bq/kg-WW, respectively. Between 2012 and 2014, the decrease in the radiocesium (137Cs and 134Cs) activity concentrations in the gonads of the sea urchins was greater in those from the coast areas (e.g. solid lines) than in those from the offshore areas (e.g. dashed lines in Fig 3). Both radiocesium (137Cs and 134Cs) activity concentrations in the gonads of sea urchins collected from the Yotsukura coast and Ena rocky coast decreased significantly (p < 0.05) (Table 2). By 2726 days after the FDNPP accident, the 137Cs activity concentrations were 0.13–0.17 Bq/kg-DW and 134Cs was not detected in the gonads of sea urchins collected from the Yotsukura coast site (S2 Table).

Fig 3

Spatial and temporal changes in the 137Cs and 134Cs activity concentrations in sea urchin gonads collected from four fishing grounds in Fukushima Prefecture after the FDNPP accident.

Solid and dashed lines represent statistically significant and insignificant regression slopes, respectively. Different trend lines indicate different sampling stations for 137Cs (—Yotsukura coast,—Ena rocky coast,—Yotsukura offshore, and ⏤ · Ena rocky offshore) and 134Cs (—Yotsukura coast,—Ena rocky coast,—Yotsukura offshore, and ⏤ · Ena rocky offshore), respectively. Data below the detection limit and dry weight (DW) basis are excluded.

Table 2

Summary statistics of the 137Cs and 134Cs radioactivity, effective half-life (Teff, day), and ecological half-life (Teco, day) in sea urchin gonads collected from the four sampling points (Yotsukura coast and offshore areas, and Ena rocky coast and offshore areas).

Study area	Radiocesium	Coefficient of determination(R2)	p-value	Coefficienta		Effective half-life (Tₑff) days	Ecological half-life (Teco) days
				C0 (Bq/kg)	λeco (d-1)
Yotsukura coast	137Cs	0.62	<0.01*	407.3	-0.0039	178	181
	134Cs	0.61	<0.01*	595.9	-0.0050	139	170
Yotsukura offshore	137Cs	0.01	0.26	166.1	- 0.0010	-	-
	134Cs	0.07	0.26	99.36	-0.0014	-	-
Ena rocky coast	137Cs	0.51	0.03*	44.19	- 0.0017	408	423
	134Cs	0.92	<0.01*	584.46	-0.0064	108	126
Ena rocky offshore	137Cs	0.18	0.72	54.72	-0.0006	-	-
	134Cs	0.56	0.16	54.15	-0.0014	-	-

*The decreasing trend was statistically significant (p < 0.05).

a Coefficient for the following exponential equation for the measured data: Ct = C0e(-ecot). Ct represents the 137Cs and 134Cs concentrations at elapsed day t, C0 represents the 137Cs and 134Cs concentrations at the time of first measurement, and λeco represents the decay rate constant.

The decreases in the slopes of the activity concentrations of both radionuclides in the sea urchins collected from the Yotsukura coast area were significant (p < 0.05) (Table 2). The λeco of 137Cs in the gonads of sea urchins was higher in the Yotsukura coast (0.0039 d−1) than in the samples from the other stations. Highest decay rate constant of 134Cs in sea urchin gonad was obtained at Ena rocky coast (λeco = 0.0064 d−1). The Teco and Teff of 137Cs and 134Cs in the gonads of sea urchins were estimated from the samples collected from the Fukushima coast and offshore areas (Table 2). The Teff and Teco of 137Cs in the sea urchins from the Yotsukura coastal area were calculated as 178 days and 181 days, respectively. Longer Teff and Teco were observed at Ena rocky coast (408 and 423 days, respectively). The Teff and Teco were not determined for the other stations as the 137Cs activity concentrations did not decrease significantly with time.

The Teff and Teco for 134Cs in the gonads of sea urchins from the Yotsukura coast and Ena rocky coast were 139 and 170 days, and 108 and 126 days, respectively, and the activity concentrations decreased significantly (p < 0.05) in the period since the FDNPP accident. We did not determine the Teff and Teco in the sea urchins from the Yotsukura offshore and Ena rocky offshore stations, as the decreases in the 134Cs activity concentrations in the sea urchin gonads from the Yotsukura offshore and Ena rocky offshore stations (p = 0.26 and 0.21, respectively) were not significant during the survey period.

The half-life of radioactive Cs was determined from the whole body for Tbio and from the gonad for Teco. For this comparison, the activity concentrations of both were measured at the same time. The 137Cs activity concentrations in the whole bodies and gonads at the Yotsukura coast were ND–5.9 and 11.9–59.4 Bq/kg-WW in December 2012, ND–4.1 and 4.56–20.8 Bq/kg-WW in February 2013, and 5.06–7.94 and 3.26–9.56 Bq/kg-WW in May 2014, respectively (S2 Table).

ACF in sea urchin gonads

The 137Cs activity concentrations in seawater at the Yotsukura coast and Ena rocky coast stations were 0.01–0.16 Bq/L and 0.01–0.10 Bq/L, respectively (S3 Fig). The regression slopes for 137Cs activity in seawater from the Yotsukura and Ena stations were significant (p < 0.05) (S4 Table). The depuration rate constants (λ) of 137Cs in seawater and in the gonads of sea urchins from the Yotsukura coast and Ena rocky coast stations were 0.0011 and 0.0015 d−1 and 0.0039 and 0.0022 d−1, respectively, so the values in the seawater were lower than those in the gonads. Moreover, the Teco values of 137Cs in seawater from the Yotsukura and Ena stations were 681 and 487 days, respectively, and were greater than the value for the gonads of sea urchins (181 days) at the Yotsukura coast.

The ACF in the gonads of sea urchins from the Fukushima coastal area varied widely (123.8–2648) between 426 and 1175 days after the FDNPP accident (Fig 4). The ACF of 137Cs in the gonads of sea urchins from the Yotsukura and Ena stations showed a decreasing trend, but the ACF was not significantly correlated with the number of days since the accident at either the Yotsukura stations (r = −0.24, p = 0.18) or the Ena stations (r = −0.10, p = 0.69). The ACF values at the Yotsukura and Ena stations had ranges of 123.8–2163 and 220.9–2648, and median values of 493.5 and 840.3, respectively. At the Yotsukura coast site, 2726 days after the FDNPP accident, the ACF was 118.0±15.99.

Fig 4

Spatial and temporal variations in the apparent concentration factor (ACF) of 137Cs in sea urchins.

Dashed lines represent statistically insignificant regression slopes.

Discussion

Radiocesium contamination after the FDNPP accident

Individuals of M. nudus were collected from the Ena coastal area on May 2013 after the FDNPP accident and were cultured in a water tank for up to 91 days. Our results showed that the 137Cs counts decreased significantly in two of the five individuals. As shown in S2 Fig, the body weights of samples A and B increased, but the body weight of sample D decreased (p < 0.05). We are not sure what caused the weight loss in sample D. The spawning season for sea urchins in the target sea area (off the coast of Fukushima Prefecture) is from July to August [25] and our rearing experiment was carried out between May and August, thus the sea urchin individuals may have spawned during the experimental period. Consequently, reared sea urchins may gain weight in the first half of summer and then lose weight due to spawning. Sample D, therefore, was in the weight loss phase in the latter half of summer. We did not confirm whether the samples were males or females, so we cannot say if there was a gender effect. The weight loss may have been related to the release of germ cells, a decrease in food intake because of an increase in the water temperature during summer, or stress in the breeding environment.

The decreases in the 137Cs counts were significant in some sea urchin samples (C and D) but were insignificant in other samples (Fig 2). The trends in the body weights of the individual sea urchins and the 137Cs counts were not uniform. To explore the relationship between the growth and the 137Cs counts in the sea urchins, the specific growth rates (SGR, % day−1) of sea urchins and reduction rates (D, day−1) in the 137Cs counts were calculated, and then a multivariate statistical model (GLM) was built using the data of all (five) sea urchin samples. The results of the GLM model showed that the SGR of the sea urchins was significantly related to the reduction rate in the 137Cs counts, i.e., when the growth rate of a sea urchin increased, then the reduction rate in the 137Cs counts also increased in our rearing environment. Other studies of different marine organisms showed that the body size had a significant effect on the depuration rate of 137Cs [15, 26, 27].

It is thought that any decreases in the variables measured in the sea urchins from the sampled sea areas reflect the habitats. In this study, the Tbio was determined from the decreasing tendency in the overall 137Cs counts. When the data for all five individuals are included, the excretion rate coefficient λbio was 0.00389 d−1 and the Tbio was 181 days (Table 1). This result, calculated from the data of all the individuals, was used as the excretion rate coefficient of 137Cs of M. nudus and agrees well with the results from Aomori Prefecture (1988) [28], who reported an excretion rate coefficient of 0.0042 d−1 for 137Cs of M. nudus taken from the food.

The radiocesium (137Cs and 134Cs) activity concentrations in sea urchins M. nudus were monitored for the period between 426 and 2726 days after the FDNPP accident (11 March 2011). The maximum 137Cs activity concentration in the gonads of sea urchins on 25 October 2012, at 139 Bq/kg-WW, was greater than the Japanese regulatory limit of radiocesium for foodstuffs (100 Bq/kg-WW), but was lower than the maximum value (290 Bq/kg-WW) reported in sea urchins by Sohtome et al. (2014) [17]. The activity concentrations decreased gradually over time in the area close to the FDNPP, such that, in 2018, the 137Cs activity concentrations in the gonads of sea urchins were 0.13–0.17 Bq/kg-DW), and 134Cs was not detected at the Yotsukura station (S2 Table).

The 137Cs activity concentrations in M. nudus in the Fukushima coastal area were examined, and the Teco was calculated from the λeco. The highest activity concentration of radiocesium was observed at Yotsukura coast. The trendline of the temporal changes in the 137Cs activity concentrations at the coast areas (Yotsukura and Ena rocky) showed a significant exponential correlation (p < 0.05) (Table 2). The λeco and Teco of 137Cs at the Yotsukura coast were 0.0039 d−1 and 181 days, respectively, and were 0.0017 d−1 and 423 days at the Ena rocky coast, respectively. Iwata et al. (2013) reported a value of 126±12 days for the Teco of 137Cs in M. nudus [16]. Their results were derived from over 1 to 2 years, and so are useful for understanding how the 137Cs activity concentrations decreased in the period immediately after the accident. Our results present the changes in the 137Cs activity concentrations between 2 and 4 years after the accident. The difference in the study period affects the Teco. The Teco of 137Cs was 181 days at the Yotsukura coast sampling point, which was close to the FDNPP. It was reported that, during the FDNPP accident, highly contaminated water flowed directly out of the plant, diffused, and flowed to the south [3]. After the highly contaminated water had passed, the 137Cs activity concentrations in seawater immediately decreased (< 0.10 Bq/L) [29, 30]. Therefore, we infer that the organisms near the FDNPP were in contact with highly contaminated water. Thus, the spatial differences in the Teco may be a legacy of the initial contamination.

The Teco of 137Cs of the sea urchin from the Ena rocky coast (423 days) was much longer than that of the sea urchin from the Yotsukura coast (181 days) (Table 2). The faster depuration rate of 137Cs in the sea urchin from Yotsukura may reflect the ongoing gradual decreases in the 137Cs activity concentrations in seawater and sediment at the Yotsukura station (S3 Fig). Also, the difference between the Teco values may be attributable to the difference in the distance of the sampling stations from the FDNPP (Fig 1), number of samples analyzed, amount of 137Cs deposited on the sediments, and the uptake of 137Cs contaminated food. Matsumoto et al. (2018) reported that the residual half-life (Tre) of 137Cs was much longer in sediments collected from the Ena rocky station (736 days) than in those from Yotsukura (482 days) [12]. This may help explain the longer Teco of 137Cs in the sea urchin caught from the Ena rocky coast of Fukushima. Because the organic matter in sea sediment contains high 137Cs that would gradually translocate to the lithogenic fraction [14] which might be ingested by sea urchin. In another study, Shigeoka et al. (2019) reported that the 137Cs activity concentrations in some seagrasses, which are important food source for sea urchins, were higher at Ena station than at the Yotsukura station [9]. Thus, the spatial difference in the Teco might be attributed by deposited 137Cs in sediment which was transferred through food uptake (e.g., organic matter and marine plants) by the sea urchins.

From their sea urchin rearing experiment, Nakamura et al. (1986) reported that each sea urchin had a unique distribution of radionuclides in the shell, spine, digestive tract, and gonad [31]. They also clarified that the whole-body (spine and shell) contamination of a sea urchin was mainly from radionuclides in seawater, and that the contamination of the digestive tract and gonad was from both seawater and food uptake. The 137Cs activity concentrations in seawater decreased very quickly after the FDNPP accident, so the concentrations were low during the survey period (< 1 Bq/L) (S3 Fig) and the 137Cs activity concentrations were low in the whole body of the sea urchins. However, the 137Cs activity concentrations in the food of sea urchins contributed significantly to the 137Cs accumulation in the gonad [31], which may explain why the 137Cs activity concentrations were higher in the gonad samples than in whole body samples.

At the beginning of this survey, the 137Cs activity concentrations in the sea urchins were much higher in the coastal samples than in the offshore samples (Fig 3). The 137Cs activity concentrations decreased significantly in the coastal samples over the period, but the 137Cs activities did not change significantly in the offshore samples (Table 2). In contrary, the average 137Cs activity concentrations in the sea urchins collected from the Yotsukura offshore (67.5±46.6 Bq/kg-WW) and Ena rocky offshore areas (31.6±10.9 Bq/kg-WW) were higher than those of coastal areas (Yotsukura = 37.9±41.8 Bq/kg-WW, Ena = 10.9±6.65 Bq/kg-WW) but the spatial differences were insignificant (ANOVA; p = 0.053). These results indicate that the 137Cs activity concentrations in sea urchins might be influenced by the consumption of 137Cs contaminated food by sea urchins, and may have been related to the ingestion of organic matter in seabed sediment with high 137Cs activity concentrations in the offshore area to the south of the FDNPP [10, 11]. Additionally, the high 137Cs in sea urchins may be related to the high 137Cs activity concentrations in a rocky reef in the offshore area, as reported by Suzuki et al. (2019) [32].

The Tbio was determined from the cesium counts in the whole body, and the Teco was determined from the cesium activity concentrations in the gonad. As mentioned in the results section (line 336–339), the 137Cs activity concentrations in the gonads tended to be high relative to those in the whole body until 2013, but were similar in 2014. Very little cesium had accumulated in the whole body until 2013. This suggests that the change in the concentrations of radioactive substances in the whole body was due to the change in the radioactivity in the gonads; that the samples were from different parts of the sea urchins had little effect on the half-life.

The Tbio values of 137Cs of the individual sea urchin samples were between 121 and 157 days, while the value for all the reared samples was 181 days (Table 1). The Tbio values were similar or close to the Teco value (181 days) of 137Cs determined from the gonads of the sea urchins at the Yotsukura coast but shorter than the value for the Ena coast (423 days). It was also revealed that there was a significant difference (ANCOVA, p < 0.0001) between the lab experiment and the field sample (Ena and Yotsukura coasts) (S5 Table).

We compiled information about the 137Cs activity concentrations in seawater (Bq/L), marine plants (Bq/kg), and sediments (Bq/kg) reported for the period from April 2012 to May 2014 by Shigeoka et al. (2019) and Matsumoto et al. (2018) [9, 12]. We compared their values with those from this study and plotted exponential trend lines over time in a depuration graph of the 137Cs activity concentrations (natural log scale) in the gonads of sea urchins, seawater, marine plants (brown algae, red algae, and seagrass), and sediment at the Yotsukura and Ena stations (S3 Fig). The 137Cs activity concentrations at both sampling stations were ranked as follows: sediment > sea urchin > marine plants > seawater.

The apparent concentration factor (ACF) measured in our study was much higher (123.1–2648) than the values reported in previous studies (ACF = 10–20) [28, 31]. The ACF was high after the accident, and then decreased gradually, but the decrease was not significant (Fig 4). The ACF value was still high (118.0±15.99) compared with the previously reported values even 2726 days after the accident, when the 137Cs activity concentrations in seawater were low [31]. Thus, in 2018, the sea urchin 137Cs activity concentrations were probably related to uptake from seawater and food.

Cesium is mainly ingested into sea urchins through food and water, and the food can have more effect on the accumulation of 137Cs than the seawater [31]. Types of kelp or seaweed, the main foods of sea urchins, contribute to the contamination in the edible part of sea urchins. Of 16 species of marine plants contaminated by radiocesium [9], 2 species (Eclonia bicyclis and Phyllospadix iwatensis) may have contributed to the contamination of sea urchins when ingested [21, 31]. The uptake of 137Cs will differ depending on the species of seaweed consumed as food. Tateda et al. (2013) reported that the 137Cs depuration rate was slower for sea urchins (by 2.1 times) than for other algae feeding invertebrates, because of the contribution of contaminated food (suspended particles attached to the algae and algal fragments) in their digestive tract [33].

Causes of long term 137Cs contamination in sea urchins

In this study, the biological half-life and the ecological half-life were determined simultaneously, so we were able to calculate the 137Cs activity concentration in the food of M. nudus collected from the Fukushima coast over the study period. Here, we adopted a compartment model for primary consumer herbivorous invertebrates [34], shown in Eq (7), and used the λbio and λeco values from this study (Tables 1 and 2) to calculate the 137Cs concentration in the food of M. nudus in 1 day. where S(t) is the 137Cs concentration in seawater (Bq/L), and B1(t) and B2(t) indicate the 137Cs concentrations of a producer and a primary consumer (Bq/kg-WW), respectively. t is the time (days). k02, k20, and k12 denote the uptake rate coefficient of the primary consumer from seawater (d−1), excretion rate coefficient (d−1), and the uptake rate coefficient from food (d−1), respectively.

Eq (7) was therefore transformed into a formula for the 137Cs concentration of food, (Eq (7)′)

dB2(t)/dt was calculated from B2 (t + 1) − B2 (t) * exp (−λeco). The values of k02 and k12 were 0.1568 and 0.0042, respectively [28]. k20 was set to λbio; 0.003889, i.e., the result from this study. The 137Cs activity concentrations in the food of sea urchins at the Yotsukura station up to 700 days after the accident and at about 1000 days after the accident were estimated as 116–19037 Bq/kg and 4483–31417 Bq/kg, respectively (S4 Fig).

Various researchers have reported the concentrations of 137Cs in seaweed along the Fukushima coast after the accident [9, 14, 16]. Shigeoka et al. (2019) investigated the 137Cs activity concentrations in 15 species of algae and 1 seagrass species and found that the 137Cs activity concentrations in algae at 500 and 1000 days after the accident were 2.86–12.3 Bq/kg-WW and 0.22–0.58 Bq/kg-WW, respectively (S3 Fig) [9]. The 137Cs activity concentration in food calculated in this study (S4 Fig) was 20–2554 times higher than the concentration in algae. These results therefore suggest that the 137Cs contamination in the sea urchins was not exclusively from seaweed but was also from other food consumed.

As sea urchins are bottom dwelling invertebrates, higher 137Cs activity concentrations in sea sediment (S3 Fig) may transfer to them through benthic food web. Otosaka and Kobayashi (2013) demonstrated that organically bound 137Cs in sediment collected from the coastal area 70 km south of the FDNPP contributed about 20% of sedimentary 137Cs, even though the sediment had a relatively low proportion of organic matter (4–6%) [35]. These results suggest that the bioavailable fraction (e.g., detritus) of 137Cs in sediment is an ongoing source of radioactive cesium to the sea urchin.

Suzuki et al. (2019) showed that the 137Cs activity concentrations on reefs were still high after the accident, which suggests that the high levels of 137Cs are sustained by ongoing contamination [32]. Sea urchins also eat coralline red algae that adhere to the rocky reef. There are no published investigations of the concentrations of radioactive substances in coralline, so we can reasonably suspect that some foods of sea urchin are still severely contaminated. It has been reported that about 3–12% of the seabed sediments is organic matter [12]. Sea urchins, therefore, eat organic matter in the sediments with unknown radiocesium concentrations and seaweed, resulting in a long ecological half-life.

In this study, we investigated the temporal changes in the 137Cs activity concentrations in the sea urchin M. nudus after the FDNPP accident and determined the biological and ecological half-lives. From these results, we examined the relationship between the changes in the 137Cs activity concentrations in the marine environment after the accident and the sea urchin contamination. We found that the changes in the 137Cs activity concentrations in the sea urchins reflected the decrease in the 137Cs activity concentrations in the food rather than the decrease in the seawater concentrations. These results reinforce that, to reduce the concentrations of radioactive materials in organisms in the Fukushima area, the concentrations of radioactive materials in ecosystems, including those in prey organisms, must first decrease.

Conclusions

After the Great East Japan Earthquake in 2011, radiocesium was dispersed into the ocean environment near the FDNPP. In this study, radiocesium contamination of the sea urchin M. nudus reared in laboratory and field conditions was monitored. The Tbio and Teco of radiocesium were also ascertained. The Tbio values of 137Cs in individual sea urchins reared in the laboratory were between 121–157 days and were shorter than the Teco of 137Cs in gonads from sea urchins collected from the marine environment (181–423 days). The Teco measurements reflect direct contamination by radiocesium from the surroundings (e.g., water, sediment) and food habits (e.g., kelp, seaweed). The 137Cs activity concentrations in the sea urchins reflect the 137Cs activity concentrations in the food rather than the seawater radioactivity. However, we did not identify the type of food. It would be useful to carry out further studies of the Teco and Tbio of various species to improve our understanding of how seafood ingest and eliminate 137Cs.

Shell length of sea urchin samples at the beginning of the rearing experiment.

(DOCX)

Click here for additional data file.

137Cs and 134Cs activity concentrations measured in sea urchins from the Fukushima area.

(DOCX)

Click here for additional data file.

Results from GLM analysis of the 137Cs counts reduction rate (D) per day in sea urchins, and a maximum likelihood test against the null model for the specific growth rate of the sea urchins.

AIC = −220.05.

(DOCX)

Click here for additional data file.

Summary statistics for the regression slopes of the 137Cs radioactivity (total number of 137Cs concentration), effective half-life (Teff, day), and ecological half-life (Teco, day) in seawater from the Yotsukura and Ena stations.

(DOCX)

Click here for additional data file.

Analysis of covariance (ANCOVA) for the relative value of the 137Cs activity (natural log scale) changes that were used to calculate the half-lives (Tbio and Teco) in sea urchins.

(DOCX)

Click here for additional data file.

Trend of 137Cs activity concentrations (Bq/L; Mean±SD) in seawater during the rearing experiment of sea urchins (Kaeriyama, 2017).

(DOCX)

Click here for additional data file.

Temporal changes in the body weights of sea urchins during the rearing period (31 May–30 August 2013) in the laboratory.

Solid and dashed lines represent statistically significant and insignificant regression slopes, respectively.

(DOCX)

Click here for additional data file.

Spatial and temporal changes in the 137Cs activity concentrations in sea urchin gonads (Bq/kg-WW), seawater (Bq/L), marine plants (Bq/kg-WW), and sediment (Bq/kg-DW) collected from two fishing areas in the Fukushima Prefecture after the FDNPP accident.

Filled and empty squares in the water data show our survey values and the NRA data, respectively. Solid lines indicate fitted exponential functions for 137Cs concentrations in sea urchins, seawater, marine plants, and sediment. Data below the detection limit were excluded.

(DOCX)

Click here for additional data file.

Uptake of 137Cs through food in sea urchins (Bq/Kg-WW) collected from the Yotsukura fishing ground in the Fukushima Prefecture after the FDNPP accident.

(DOCX)

Click here for additional data file.