Radiocesium biokinetics in olive flounder inhabiting the Fukushima accident-affected Pacific coastal waters of eastern Japan

Journal of Environmental Radioactivity 147 (2015) 130e141

Contents lists available at ScienceDirect

Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad

Radiocesium biokinetics in olive flounder inhabiting the Fukushima accident-affected Pacific coastal waters of eastern Japan Yutaka Tateda a, *, Daisuke Tsumune a, Takaki Tsubono a, Tatsuo Aono b, Jota Kanda c, Takashi Ishimaru c a b c

Environmental Science Research Laboratory, Central Research Institute of Electric Power Industry, 1646 Abiko, Abiko-shi, Chiba-ken 270-1194, Japan National Institute of Radiological Science, 14-9-1 Anagawa, Inage-ku, Chiba-shi, Chiba-ken 263-8555, Japan Tokyo University of Marine Science and Technology, 4-5-7, Konan, Minato, Tokyo, 108-8477, Japan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 October 2014 Received in revised form 22 May 2015 Accepted 27 May 2015 Available online xxx

Radiocesium (134Cs and 137Cs) originating from the Fukushima Dai-ichi Nuclear Power Plant (1FNPP) has contaminated coastal waters and been subsequently transferred to the marine biota along the Pacific coastal region of eastern Japan. To clarify the mechanism of radiocesium biokinetics in olive flounder, a commercially valuable and piscivorous predator, the biokinetics of 137Cs was simulated using a dynamic biological compartment model and then validated with the measured concentrations in available monitoring data. The 137Cs concentrations in seawater of the Pacific coastal sites of eastern Japan, from Kesen-numa (170 km north from the 1FNPP) to Choshi (190 km south from the 1FNPP), were reconstructed by fitting the simulated levels to the observed concentrations. Simulated values were verified by measured radiocesium levels in sedentary organism such as macro-algae and mussels inhabiting each study site which had accumulated radiocesium in their ambient environment from the beginning of the accident. Using reconstructed 137Cs concentrations in seawater, the 137Cs levels in olive flounder and its main planktivorous prey fish, e.g. anchovy, sand lance, whitebait, etc., were simulated and compared with observed concentrations to clarify the biokinetics of radiocesium in these organisms. This assessment showed that the determining factor for the maximum radiocesium concentrations in fish in the plankton food chain is likely to be the initial radiocesium concentration which they were exposed to during the contamination stage. Furthermore, the simulated 137Cs concentrations in gut contents of olive flounder were verified by measured 137Cs concentrations in the stomach contents of this fish collected within 30 km from the 1FNPP. These results indicated that the decrease of 137Cs levels in their prey organisms was the primary determining factor of radiocesium depuration, and the resultant ecological half-lives were 140e160 d in the olive flounder, by the simulation. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Fukushima reactor accident Dynamic biological compartment model 137 Cs Food chain transfer Coastal organisms

1. Introduction Radiocesium isotopes (134Cs and 137Cs) were released to the environment as a result of the accident at the Fukushima Dai-ichi Nuclear Power Plant (1FNPP) of the Tokyo Electric Power Company (TEPCO) following the Eastern Japan Earthquake and subsequent tsunami on 11 March 2011. A simulation which employed an ocean dispersion model (Tsumune et al., 2013) estimated the total 137Cs radioactivity directly leaked as effluent into the ocean to be 3.6 PBq. In the case of atmospheric deposition to the surface of ocean, 12 PBq

* Corresponding author. E-mail address: [email protected] (Y. Tateda). http://dx.doi.org/10.1016/j.jenvrad.2015.05.025 0265-931X/© 2015 Elsevier Ltd. All rights reserved.

was estimated by a simulation for the northwestern Pacific Ocean (Katata et al., 2014), however, air-borne 137Cs deposition to the coastal area has not been fully evaluated. In addition, riverine radioactive cesium discharge to coastal water was also suggested (Yamashiki et al., 2014), however the contribution ratio to the radiocesium inventory in coastal waters is not well documented. The radiocesium introduced to coastal waters was rapidly transferred to biota elevating 137Cs levels in a variety of marine organisms (Buesseler et al., 2011; Madigan et al., 2012). Furthermore, the radiocesium which was initially accumulated then decreased gradually over time in most of the coastal biota (Wada et al., 2013; Sohtome et al., 2014), while the 134þ137Cs levels in some demersal fish have still exceeded the Japanese regulatory limit of 100 Bq kg-wet1 even 3 y after the accident (MAFF, 2015). In

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

addition, the surrounding benthic environment with its contaminated sediment was suggested as a possible source which caused the delay of 137Cs depuration from benthic biota (Buesseler, 2012; Tateda et al., 2013). However, the exact mechanism has not been clarified. We reconstructed the temporal radioactive cesium concentrations in marine organisms inhabiting the southern coast of the Fukushima Prefecture (Tateda et al., 2013) by a simulation using a dynamic food chain transfer model, combined with a simulation of physical dispersion of radioactive cesium using the Regional Ocean Modeling System ROMS (Tsumune et al., 2012). The simulated radiocesium levels in the invertebrates, planktivorous fishes and their predator fish, e.g. olive flounder, in a coastal area 30 km south from the 1FNPP were explained by the established radiocesium biokinetics and food chain transfer through the ecosystem. However, the radiocesium levels in these organisms which inhabited the other coastal waters affected by the Fukushima accident were not well simulated, because of the insufficient reconstruction of the actual radiocesium levels in seawater. In this study, the 137Cs concentrations in the seawater at study sites in the Pacific coast of eastern Japan, from Kesen-numa (170 km north from the 1FNPP) to Choshi (190 km south from the 1FNPP), were reconstructed by fitting simulated levels to observed concentrations. These values were then verified by comparison with the measured radiocesium levels in the sedentary organism such as macro-algae and mussels which inhabited each study site and which had accumulated radiocesium in their ambient environment from the beginning of the accident. Using the reconstructed 137Cs concentrations in the Pacific coastal waters of eastern Japan, the 137 Cs levels in the typical piscivorous fish, olive flounder (Paralichthys olivaceus) and its typical planktivorous fish prey were simulated. Radiocesium transfer from seawater and through their food chain were considered in order to clarify the biokinetics of radiocesium transfer in these organisms. In addition, the 137Cs concentrations in gut contents of olive flounder inhabiting the area within 30 km of the 1FNPP were measured. These values were then used to verify the model which simulated transfer flux to olive flounder from prey, and evaluated the significance of the contribution of radiocesium transfer through the food chain during the depuration period in olive flounder. 2. Material and methods 2.1. Simulation The 137Cs levels in seawater which was contaminated by both atmospheric deposition and the direct leakage from 1FNPP accident were simulated as daily values in 20 layers from surface to bottom of a 1 km grid by using the Regional Ocean Model System (ROMS) for the area of 35 540 N e 40 000 N, 139 540 E  147 000 E during 1 March 2011 until 31 December 2012. With reference to our previous study (Tateda et al., 2013), 137Cs leakage from the source was assumed to continue at a rate of 50 GBq d1 after 1 November 2011 (Kanda, 2013). Daily 137Cs concentrations in surface and bottom water were simulated at the study sites in order to apply them to a simulation of 137Cs concentrations in organisms using a dynamic biological compartment model. The dynamic biological compartment model has 12 biological compartments with a detritus sub-model consisting of four subcompartments. Food web relations and food composition in the model were determined from values in the literature (Tateda et al., 2013; Fig. S1). In the biological simulation at each study site, temporal 137Cs concentrations in each organism compartment Bn(t) (Bq kg-1-wet) at time t (d) were determined by the 137Cs concentrations in seawater S(t) (Bq kg1) and in food Bsm(t) (Bq kg-1-wet)

131

at the same time t (d). These values are derived by solution of the following equation simultaneously with food species compartments;

dBn ðtÞ=dt ¼ k0;n SðtÞ þ r1 d1;n Bs1 ðtÞ þ r2 d2;n Bs2 ðtÞ þ ð1  r1  r2 Þd3;n Bs3 ðtÞ  kn;0 Bn ðtÞ

(1)

where k0,n: uptake rate constant of 137Cs from seawater to organism Bn (d1) kn,0: excretion rate constant of 137Cs from organism Bn (d1) dm,n: uptake rate constant from food Bsm to organism Bn (d1), where dm,n ¼ an frn, where an ¼ assimilation rate of the 137Cs in organism Bn frn ¼ daily feeding rate of organism Bn (wet g of food) (wet g of Bn)1 d1 r1, 2, and 1-r1-r2: food composition rate of food organism Bsm to organism Bn (d1), The transfer parameters used in the calculations were the same as those used in the previous study (Table S1). For planktivorous fish of 2012 year class, those concentrations were simulated, assuming that they hatched at mid-January 2012. For olive flounder, the monitoring size was approximately >35 cm (body length), corresponding to > 3 y class (Kitagawa et al., 1994), which eliminated any year class hatched after 2012; thus the simulated value was understood to be the concentration in fish of the year class hatched before 2008.

2.2. Data for evaluation of simulated concentrations in seawater and organism 2.2.1. Radioactivity measurement of collected samples For the verification of simulated values in seawater and sedentary organisms, radiocesium concentrations were analyzed in surface seawater, several species of Sargassum and the purplish bifurcate mussel Septifer virgatus which had been collected at St. 1 to 8 (Fig. 1, Table S2) during April to September 2011. Surface seawater samples were collected in Teflon buckets followed by acidification with nitric acid. In the laboratory, an aliquot of 80 ml seawater was transferred to a 100 ml polystyrene container, and was directly measured with the gamma-spectrometer mentioned later. In case the radioactivity was below the detection limit, an aliquot of 1 L seawater were subjected to AMP extraction to concentrate radiocesium (Aoyama et al., 2012) using low background purified AMP (The General Environmental Technos Co., Ltd). Macro-algae and mussels were sampled by hand from the rocky shoreline of four and five study sites, respectively, during low-tide. Macro-algae of identified species were rinsed at the sampling sites with local seawater to remove attached epibionts, and wet weight was measured after wiping off the excess surface water. Algae samples were brought back to the laboratory, and dry weights were measured after freeze drying. Mussel samples were immediately transferred to the laboratory and dissected to separate soft parts from the shell. After measuring wet weight, soft parts were freeze-dried and re-weight to obtain dry weight. Within a 30 km area of the 1FNPP area, samples of stomach of olive flounder were collected during the dissection of fish muscle samples for radioactivity monitoring by the TEPCO during January to December 2013. The dissected stomachs were immediately frozen and sent to the laboratory where stomach contents were sorted to categorize food items as either fish, crustacean,

132

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

Fig. 1. Simulated and observed concentrations of 137Cs in seawater at the eight study sites (C) and 13 reference sites (-, scientific study site; △ and ,, monitoring sites) along the Pacific coasts of eastern Japan. Locations and codes of study and reference sites are shown in Table S2.

cephalopod or unknown, and the wet weight of each component was measured. After freeze drying to determine dry weight, prey food samples or total stomach contents were subjected to radioactivity measurement. Since, the radioactivity in fish was analyzed as a composite of three samples, stomach content samples were composited to correspond with the original fish sample data at each study site. All dried biological samples were

powdered and transferred to 100 ml polystyrene containers, and radioactive cesium was measured by a gamma-spectrometer (Gedetector GR2519, Inspector 2000, CANBERRA). The counting efficiency of sample were calibrated and determined by using calibration software (ISOCS, Genie 2000, CANBERRA), and a set of standard gamma volume sources (MX033U8PP, JRA). Measurement times were adjusted to obtain a standard deviation from

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

counting statistics less than 10%. Measurement accuracy was checked by using an IAEA standard reference material (Tuna fish, IAEA-352). 2.2.2. Published data sets of seawater and organisms Cesium-137 concentrations in seawater along the Pacific coastal waters of eastern Japan were monitored by the Ministry of Environment, Japan (ME, 2014), the Fukushima Prefecture (FP, 2014), and TEPCO (2015). Measurements from near the study sites have also been published by Oikawa et al. (2013) and Aoyama et al. (2012). Using latitude and longitude information for the reported data along with sampling depths, the nearest data sets were selected to compare with simulated 137Cs concentrations as reference sites (Table S2). To avoid using incorrect data due to improper background treatment in measurement of the TEPCO monitoring (NRA, 2013), we only used data of >1 Bq l1 before August 2011 and radiochemical analysis data after April 2012. Cesium-137 concentrations in macro-algae, some planktonic crustaceans, planktivorous fish, and olive flounder were monitored by the Fisheries Agency, Ministry of Agriculture, Forestry and Fisheries Japan (MAFF, 2015). Within a 30 km area of the 1FNPP, 137 Cs concentrations in muscle of planktivorous fish and olive flounder were monitored and reported by TEPCO (TEPCO, 2015). In figures, citation from monitoring data sets of ME (2014), FP (2014), TEPCO (2015), and MAFF (2015) are abbreviated as ME, FP, TEPCO, and MAFF. Measurement results for radiocesium in macro-algae and mussels (Baumann et al., 2013) and benthic invertebrate species (Sohtome et al., 2014) have also been reported. We used these data for verification of simulated 137Cs concentrations in these biota (macro-algae, three sites; plankton, two sites; mussel, two sites; planktivorous fish, 14 sites; olive flounder, 14 sites) at the reference sites (Table S2), which are close to the study sites of St. 1 to St. 8, and have similar temporal profiles of seawater concentrations (Fig. 1). 2.3. Statistical treatment The significance of the difference between observed data sets was examined by a KruskaleWallis test followed by a Wilcoxon rank sum test. The significance of the correlation was calculated by statistical analysis (t-test, 0.05 level of significance, KaleidaGraph, Synergy Software). 3. Results and discussions 3.1. Reconstruction of

137

Cs concentration in seawater

Simulated 137Cs concentrations in surface water of study and reference sites of the eastern coast of Japan are shown in Fig. 1, along with the reported observed seawater concentrations (Aoyama et al., 2012; Oikawa et al., 2013), monitored values (ME, 2014; FP, 2014; TEPCO, 2015) at reference sites, and measurements made in this study (Table S3). The simulated profiles of temporal 137Cs levels at each study site were generally comparable to the observed 137Cs concentrations, however, the simulated levels were lower than the observed concentrations at St. 3, 4, and 5 (p < 0.001, p < 0.001, and p < 0.01, respectively), indicating that the input source was in deficit for the simulation of seawater levels. This result is compatible with the result at St. 4 which was evaluated during a 1.5 y data period by Tateda et al. (2013), thus confirming a source deficit in the surrounding coastal waters of the Fukushima Prefecture. In contrast, the temporal profiles of 137Cs levels at St. 1 and 2, the coastal areas of the Miyagi Prefecture, were comparable with the simulated values, suggesting the compatibility of simulated levels and observed concentrations in the

133

northern coastal area (>100 km north from 1FNPP). With regard to the coastal areas of St. 6 to St. 8, which are >120 km south of the 1FNPP, the profiles of simulated levels during May to July 2011 were lower than the observed concentrations (p < 0.001). The reason for this difference was presumed to result from an inadequate reproduction of the oscillation of the southward Oyashio current. Despite this disagreement, 137Cs levels in seawater after August 2011 were in near agreement with the observed temporal 137Cs levels in the coastal waters of Ibaraki, the southern prefecture adjacent to Fukushima. To derive seawater 137Cs concentrations which would be applicable for the biological model calculations, the simulated 137Cs levels were fitted to observed concentrations by approximation at each of the study and reference sites. The power approximation used to best fit the simulated levels to the observed concentrations (Fig. S2) employed in the following equation:

Sobserved ðtÞ ¼ aðSsimulated ðtÞÞb

(2)

where Sobserved(t) is the radiocesium concentration observed in seawater at day t, Ssimulated(t) is the radiocesium concentration simulated in seawater at day t, and t is the days after 1st March 2011. In this equation, a constant a is regarded as a correction factor for fitting the simulated levels to the observed concentrations. Using the correction factor a, the deficit in simulated level compared to the observed concentration can be derived by (a  1)/ a as the deficit ratio. The chemical form of cesium is ionic in seawater, thus the dissolved radiocesium introduced to the marine environment is not evidently adsorptive being considered to be conservative in its inventory within water column. Hence its inventory in a water body is expected to change only by diffusion. In this sense, the constant b, when it has changed significantly, is presumed to be the driving factor to represent additional local input to each of the coastal waters other than the initially introduced radiocesium from the accident. The adjusted seawater 137Cs concentrations were derived and the result of the one reference (St. M) and five study sites (St. 3, 4, 5, 6, and 8) are shown with observed values in Fig. 2(a). However, during 11e22 March 2011 measurements of radionuclides in the marine environment were not carried out and the monitoring started only after 23 March 2011, measurements were scarce and did not fully cover the entire coastal waters. Thus there may still be some uncertainty in using a limited set of observed concentrations for verification of the actual concentrations. In this study, by taking advantage of the sedentary organisms which were contaminated directly by seawater from the start of the accident, and retain their exposure record by way of their bioaccumulated tissue levels, the 137Cs concentrations in macro-algae and sedentary planktivorous bivalves were simulated by using the adjusted seawater concentrations. The simulated values were then compared with observed 137Cs concentrations in collected sedentary organisms such as sargassum and mussels inhabiting the rocky shoreline to verify the applicability of the adjusted seawater concentration used for calculation of the radiocesium concentration in organisms. The result of the simulated 137Cs concentrations in macroalgae which corresponds to sargassum at the reference and study sites are shown in Fig. 2(b), along with previously reported concentrations (Baumann et al., 2013; MAFF, 2015) and the concentrations measured in this study (Table S4). Similarly, the simulated 137Cs levels in planktivorous bivalves at four study sites (St. 3e6) are shown in Fig. 3(b) with the observed concentrations in mussel collected in this study (Table S5), and reported levels in mussel (Baumann et al., 2013; MAFF, 2015). To understand the radiocesium level in the food source for the mussels, the

134

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

Fig. 2. (a) Adjusted 137Cs concentrations in seawater at study sites (St. M, St. 3e6, 8), and observed ones in the study sitesa) and nearby reference sites. (b) simulated 137Cs concentrations in macro-algae at the study sites (St. M, St. 3e6, 8) and reference sites (T-12 and T-18), and observed ones in the sitesb). a) Including data at St. 8 by Aoyama et al. (2012) and at L1 by Oikawa et al. (2013). b) Including data by Baumann et al. (2013).

simulated 137Cs concentrations in zooplankton and detritus are also shown in Fig. 3(a). Also reported are concentrations in sand bottom mysid Acanthomysis mitsukurii, sand bottom shrimp Crangon uritai, and velvet shrimp Metapenaeopsis dalei (Sohtome et al., 2014) which will serve as analog organisms for zooplankton. As a result, the simulated temporal levels of the 137 Cs in macro-algae were compatible with the observed concentrations in macro-algae at the one reference and five study sites (Fig. 2(b)). Likewise, the simulated temporal 137Cs levels in mussels were comparable to observed values in mussels collected from the four study sites (Fig. 3(b)). The results showed a similar compatibility in the simulated values with the observed concentrations for zooplankton (Fig. 3(a)). The overall relationship between the simulated 137Cs concentrations and observed concentrations in these two sedentary organisms at seven sites is shown in Fig. 4. The simulated 137Cs concentrations in the sedentary organisms correlated with the observed 137Cs concentrations in corresponding organisms with a statistical significance at p < 0.001. This agreement demonstrated that the reconstructed 137Cs concentrations in seawater derived by approximation agree closely with the actual values and can be regarded as appropriate for calculating radiocesium concentrations in coastal organisms inhabiting the Fukushima 1FNPP accident affected coastal area.

3.2. Deficit ratio of radiocesium input to the Fukushima coastal water The correction factor a at four study sites and 15 reference sites were examined because of multiple sets of data of observed 137Cs concentration in seawater (Fig. 5(a)). Statistically significant (p < 0.05) correction factors (a in Eq. (2)) was derived at four study and ten reference sites and are shown in Fig. 5(b) and Table 1. The median value of the correction factors was 2.0 (the 1st and 3rd quartiles were 1.6 and 3.8) (Fig. 5(b)). The calculated deficit ratio was 0.6 at St. 3 in the south of Miyagi Prefecture, and was 0.7 at T-22 in northern coastal areas of Fukushima. In contrast, the deficit ratio was 0.8 as a maximum at T-M10 and T-18 in the southern coastal area of Fukushima (Table 1). According to the reconstructed 137Cs levels in surface seawater at each study site, the maximum concentrations along the Pacific coastal waters of eastern Japan were derived and are also shown in Table 1. The maximum 137Cs concentrations in surface water were higher (0.9e2 kBq l1) in front of the 1FNPP and to south area (T-S4, T-S8, T-S7, T-S5, and St. 4) than those (0.1e0.4 kBq l1) in the southern area (T-17-1, T-M10, T-18, T-20 and St. 5), and were 0.01e0.03 kBq l1 in the northern Fukushima (T-22, T-MA, and St. 3) and northern Ibaraki (St. 6) coastal waters. Considering that these maximum concentrations were the consequence of a current-

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

Fig. 3. (a) Simulated 137Cs levels in zooplankton and detritus along with observed concentrations in shrimp and mysid, and (b) simulated concentrations in the purplish bifurcate mussel Septifar virgatus.

driven contaminated water plume which was transported from the waters just off the 1FNPP, dilution ratios (as a ratio of (maximum concentration of seawater at reference/study site)/(maximum concentration of seawater at the landing pier of the 1FNPP port)) were estimated as being approximately 2e5  105 at St. 3 to T-MA

Fig. 4. The relationship between simulated 137Cs concentrations and observed concentrations in sedentary organisms (A, macro-algae; M, mussel) using adjusted seawater concentrations at St. M to St. 8 along the Pacific coasts of eastern Japan.

135

137

Cs levels in bivalve with observed

in the northern area, 1e3  103 off T-S4 to the more southerly St. 4 area, and 2e6  104 at T-17-1 to St. 5 in the southern area, respectively (Table 1). Assuming that the 137Cs activity of 3.6 PBq was a contamination source for water in front of the landing pier in the 1FNPP port (660 kBq l1 at 6 April 2011 by TEPCO (2015), the order of magnitude of the 137Cs inventory as a supplementary source at each site was derived in order to compensate for the deficit observed in surface water concentration as shown in Table 1. The supplementary source off the 1FNPP down to the southern coastal area (T-S4, T-S8, T-S7, and T-S5) were estimated to range from 5 to 10 GBq and were higher than the 0.5e2 GBq in the southern coastal area (T-17-1 to St. 5), and were below 1 GBq order of magnitude in the northern coastal area (St. 3 to T-MA). Although, this estimation should be revised after the identification of any unknown source, the preliminary evaluation of supplementary source will aid in the discussion of radiocesium redistribution between land and coastal waters. An additional 137Cs input event to coastal water was reported during September 2011 as being associated with riverine suspended fine particle discharge, following heavy rain (Nagao et al., 2013), with a proposed total discharge of 137Cs of approximately 5 TBq via the Abukuma river in the south of the Miyagi during the period August 2011 to May 2012 (Yamakishi et al., 2014). The exchangeable 137Cs associated with this material was examined by 1 M ammonium acetate extraction and reported to be <5% of the total content in the sediment particles (Otosaka and Kobayashi,

136

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

that the input occurred only during the initial weeks immediately after the accident at each of the coastal waters investigated. A supplementary source probably in dissolved form of 5e10 GBq and 1e3 GBq in the south and southern coastal waters, respectively, of the Fukushima Prefecture can be explained by radiocesium being loosely attached to terrestrial components which may have been transported just after the initial accident, probably during March to April 2011. Spherical Cs-bearing particle deposition which was observed in the terrestrial environment (Adachi et al., 2013) is another possible initial source even for the coastal area. However, the contribution to surface water by this particle deposition as the initial input is not yet documented, and is thus a topic for future study. 3.3. Temporal prey

Fig. 5. (a) The 19 sites examined for correction and change factors (a and b) by fitting of simulated seawater concentration to observed seawater concentration (C, sampling study site; ,, monitoring sites). (b) The Box and whisker plot of statistically significant results (p < 0.05) of 14 sites in Table 1. Box shows upper and lower quartiles with an inside bar of the median, and whiskers show minimum and maximum.

2013). If it is assumed that there is an exchangeable activity of <250 GBq (5 TBq  <0.05) input from riverine particles discharged from Abukuma river mouth, 30 km north from the Fukushima border, then this source is likely to comparable with the estimated local supplement source of 0.04e0.2 GBq to northern Fukushima coastal waters. This comparison assumes a likely 103 times dilution from the river mouth. This estimation additionally suggests that terrigenous exchangeable radiocesium is also likely to be the regional source to south Fukushima coastal water from rivers within Fukushima Prefecture. The seawater simulation in this study includes input sources both from the direct leakage and the atmospheric deposition originating from the 1FNPP accident. Hence, the possible additional input source to compensate for the deficit in seawater levels is presumed to be caused by the released radiocesium which has been redistributed from contaminated land. The values of change factor b at four study and six reference sites were statistically significant (p < 0.05) (Table 1), with the median of the change factor b being 1.0 (the 1st and 3rd quartiles were 1.0 and 0.97, Fig. 5(b)). This result indicated that the time-dependent change of local input flux from the regional source was negligible for each coastal water during the 3 y period. Therefore, it is likely

137

Cs concentrations in olive flounder and its typical

Using the reconstructed 137Cs concentrations in surface seawater at St. 1 to St. 7, 137Cs levels in planktivorous fish, which is the dominant prey of olive flounder, were simulated and shown in Fig. 6(a) along with observed concentrations in corresponding species such as anchovy Engraulis japonicus, sand lance Ammodytes personatus, whitebait etc. (MAFF, 2015) collected from fishing areas off the study sites (Table S2). In addition, simulated results for the 2012 year class planktivorous fish are also shown to evaluate any age effect on their concentrations. Simultaneously, the 137Cs concentrations in olive flounder, which represents piscivorous fish species, were also simulated from reconstructed bottom seawater concentrations and are shown in Fig. 6(b) along with observed concentrations in monitored olive flounder (MAFF, 2015). To verify the effect of fish movement on the concentration, the simulated results of reference sites at close proximity to a study site, e.g. at St. 3 to St. 5, and St. 7, are also shown. Simulated temporal profiles and levels of 137Cs in planktivorous fish contaminated in 2011 were comparable to observed concentrations in corresponding species, indicating that 137Cs levels in these fish species at each study site could be explained by modeled uptake both from the contaminated water and from the planktonic matter during several months after the accident. The agreement of the simulated profiles of exponential decrease with the observed depuration of 137Cs concentrations in monitored planktivorous fish species of the 2011 contaminated group were also interpreted to be a result of the metabolic excretion of radiocesium (Tateda et al., 2013). This is due to the radiocesium content in their planktonic matter food being sufficiently depleted and not a significant radiocesium source for this fish (Fig. 3(a)). In contrast, the 137Cs concentrations in planktivorous fish of the 2012 year class were simulated as being stabilized in the range from 0.1 to 5 Bq kg-wet1 (Fig. 6(a), Table 2) which suggests equilibration with the 137Cs in seawater (0.003e0.2 Bq l1) at the end of 2012 (Fig. 2(a)). The derived concentration factor (CF) of the 2012 year class planktivorous fish was estimated to be 30 (median value with range of 8e230), which is comparable with a measured CF in Japanese fish (Tateda and Koyanagi, 1996). Concerning the olive flounder, calculated 137Cs levels (Fig. 6(b)) were nearly comparable with observed concentrations (MAFF, 2015) at all of the reference sites. Compared to the sedentary organisms inhabiting a single location, movement of olive flounder may affect the radiocesium level in this species by temporal changes of exposure to different levels of radiocesium in seawater. Though the reported migration distance was ca. 50 km for the olive flounder with a body length <20 cm in the Pacific coastal waters in eastern Japan (Ishido, 1990), the evaluated group of >3 y body length of >35 cm in this study was reported to stay within their habitat after the migration of young stages (Takeno, 2010). The

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

137

Table 1 Correction factor, change factor, derived deficit ratio and simulated137Cs maximum concentration in surface seawater at the study and reference sites with calculated dilution ration and estimated GBq for supplement source. Prefecture Location Miyagi Watari Fukushima off Souma off Kashima off 1FNPP off Kumagawa off 2FNPP off Kidogawa Yotsukura off Natsuigawa off Iwaki off Toyoma off Onahama Ibaraki Nakoso Hitachinaka

Study site Reference site

Correction factor a

Change factor b

Deficit ratio (a1)/a

Maximum137Cs concentration in surface seawater (kBq l1)

Dilution ratioa

St.3

2.7

0.9

0.6

0.01

2  105

0.05

5

          

Estimated supplement sourceb (GBq)

T-22 T-MA T-S4 T-S8 T-S7 T-S5 St.4 T-17-1 T-M10 T-20 T-18

3.0 2.6 1.5 1.5 1.4 1.2 3.8 2.0 4.1 3.9 4.1

0.9 0.9 (1.0)c (1.0) (1.0) (1.0) 0.9 0.7 0.9 0.9 1.1

0.7 0.6 0.3 0.3 0.3 0.2 0.7 0.5 0.8 0.7 0.8

0.01 0.03 1 0.9 1 2 1 0.1 0.1 0.2 0.4

2 5 2 1 2 3 2 2 2 3 6

10 105 103 103 103 103 103 104 104 104 104

0.05 0.2 5 5 5 10 5 0.5 0.5 1 2

St.5 St.6

2.8 1.8

1.3 1.0

0.6 0.4

0.1 0.01

2  104 2  105

0.5 0.05

Constants at St. 1 Kesennma, St. 2 Ishinomaki, T-13-1 Niidagawa, T-S1 Ootagawa, T-S2 Odaka, T-S3 Ukedo, St.7 Kashima, and St. 8 Choshi were not insignificant in regression analysis. a Dilution ratio ¼ (maximum137Cs concentration in surface seawater at landing pier in the port of 1FNPP: 660 kBq l1, TEPCO (2015))/(maximum137Cs concentration in surface seawater of reference study site). b Supplement source GB ¼ 3.6 PBq/(dilution ratio)  106. c According to the results of other study sites, changing factor b was assumed as 1.0.

difference of 137Cs levels simulated at T-12 and T-M10 (St. 4 in Fig. 6(b)) and the corresponding distribution range of observed concentrations may suggest some effect of fish movement (30 km distance between T-12 and T-M10) on the radiocesium concentrations in olive flounder in this area. Similarly, the lower 137Cs concentration in the monitored fish group compared to the calculated levels simulated at T-18 and T-20 (St. 5 in Fig. 6(b)) raise the possibility that the monitored value originated from the southern fishery area. However, 137Cs levels simulated in the overall areas demonstrated that the observed 137Cs concentrations in olive flounder generally agreed with the simulated temporal change at each study site, indicating that the fish movement was not critical in determining radiocesium concentrations in the case of mature olive flounder. The simulated result for olive flounder in this study was derived from the general food composition ratio (planktivorous fish, 0.33; cephalopod, 0.33; carnivorous fish, 0.34: Table S1), and was sufficiently comparable to the observed 137Cs concentration in olive flounder inhabiting the Pacific ocean coast along eastern Japan. On the other hand, the food regime of olive flounder in the Fukushima coastal water reported by Tomiyama and Kurita (2011) was a food composition dominated by planktivorous fish of 0.8 (cephalopod, 0.1; other fish, 0.1). Although the above site specific food composition should have been used for the olive flounder in proximity to Fukushima, a trail simulation was not successful in this study. The reason was likely to be a result of the unsuccessful reconstruction of radiocesium level for “other fish, 0.1”, which was underestimated in this study. Thus “carnivorous fish, 0.34” used in this study was considered to compensate for the deficit of 137Cs transfer from prey, especially around the Fukushima area in this study. This hypothesis is supported by the requirement for additional radiocesium transfer to demersal fish, which was raised in previous reports (Tateda et al., 2013). There is still uncertainty in the radiocesium transfer flux from prey of each food category for olive flounder living around Fukushima. However, the application of our approach in other areas demonstrated the sufficient applicability of the dynamic food chain compartment model to the planktonic organisms, planktivorous

fish and their predator olive flounder for the accident-affected area of the Pacific coastal waters of eastern Japan. With regard to the coasts within 30 km from 1FNPP (T-S1, T-S2, and T-S3 (the northern Fukushima coastal area); T-S4 and T-S8 (the coastal area just off 1FNPP); and T-S7 and T-S5 (the southern coastal area)), the 137Cs concentration in the predator fish was simulated with the use of reconstructed 137Cs concentrations and the results are shown in Fig. 7 along with the monitored values in surface seawater (TEPCO, 2015) and olive flounder (TEPCO, 2015; MAFF, 2015) collected in each area. Simultaneously, simulated 137Cs concentrations in the stomach contents of olive flounder are shown with measured values from this study (Table S6). The results showed that the 137Cs concentrations in olive flounder inhabiting within 30 km of the 1FNPP decreased exponentially both in the simulated levels and the observed concentrations, as shown in subfigures (3) in Fig. 7 (a), (b) and (c). Simultaneously, the simulated 137Cs levels in stomach contents of olive flounder decreased over time, a finding which was confirmed by a decrease in the measured 137Cs concentrations in corresponding samples at these sites (subfigures (2) in Fig. 7 (a), (b) and (c)). The simulated maximum 137Cs levels in planktivorous fish are shown in Table 2. These values indicate maximum concentrations of 2 Bq g-wet1 in the area south of the 1FNPP (T-S7 and T-S5), whereas they were 0.2e0.6 Bq g-wet1 in waters off 1FNPP (T-S3, TS4 and T-S8) and in the southern area (St. 4 and St. 5) during AprileMay 2011. In contrast, the simulated 137Cs levels in planktivorous fish inhabiting the northern Fukushima waters of T-S1 and T-S2 and the southern waters of St. 6 and St. 7 were lower than the simulated 137Cs levels in planktivorous fish inhabiting the Fukushima coastal waters. In addition, the simulated results demonstrated that the maximum 137Cs concentrations in olive flounder were the same as those in planktivorous fish at all the reference sites, although the maximum concentrations occurred later than those of planktivorous fish, i.e. from August to September 2011 in the Fukushima coastal area and October to December 2011 in neighboring prefectures. The 137Cs concentrations in olive flounder and planktivorous fish were estimated not to exceed 0.05

138

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

Fig. 6. (a) Simulated 137Cs levels in planktivorous fish with observed concentrations in sand lance (Ammodytes personatus), Ishikawa shirauo (Salangichthys ishikawae), whitebait, sardine (Sardinops melanostictus) and anchovy (Engraulis japonica). Simulated 137Cs levels in planktivorous fish at T-12 and T-20 are not shown because of their similar profiles to those of adjacent sites. (b) Simulated 137Cs levels in olive flounder (Paralichthys olivaceus) along the Pacific coast of eastern Japan. The data of below detection limit (BDL) are shown by a bar symbol (). After, the first occurrence of a BDL, measured value are shown by triangle symbol ( ).

▵

Bq g-wet1 (which is below 100 Bq kg-wet1 as total radiocesium of the Japanese regulatory limit for seafood product) to the north of T-MG1 and south of St. 6. These results demonstrated that the determining factor for the maximum radiocesium concentrations in fish in the planktivore food chain is likely to be the initial radiocesium concentration they were exposed to during the contamination stage. As shown in Fig. 6 and Table 2, the maximum concentrations in both fish groups were of the same order of magnitude regardless of food habits, but instead depend upon the distance from the 1FNPP. Since the maximum radiocesium concentrations in each coastal water were determined primarily by the distance from the 1FNPP (Table 1), the conventional method to estimate the maximum radiocesium concentration in planktivorous fish and their predator fish may be empirically derived from maximum seawater concentration record at the coastal area where these fish collected. For example, the maximum

137 Cs concentrations of 2 Bq g-wet1 in the planktivorous fish are the same as the 2 Bq g-wet1 concentrations in predator piscivorous fish at T-S5, and 7 (Table 2), and are of the same order of magnitude as the maximum 137Cs concentration of 1e2 kBq l1 in seawater (Table 1). Exposure to the maximum concentration of radiocesium in seawater is suggested as the determining factor for the radiocesium concentration in fish, indicating the importance of maintaining the seawater concentration as low as possible during the initial contamination phase to reduce the resultant radiocesium increase and the maximum levels in fish.

3.4. Depuration of radiocesium from olive flounder and influential factors Cesium-137 depuration from the 2011 contaminated planktivorous fish had resulted in reductions in levels which were similar to

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

139

Table 2 Simulated maximum137Cs concentration in planktivorous fish and olive flounder at study and reference study sites within the accident-affected area in the Pacific coast of eastern Japan. Prefecture Location

Study site Reference site

Maximum137Cs concentration Planktivorous fish

Olive flounder

2011 year class

Miyagi Kesennuma Ishinomaki Watari Fukushima off Ootagawa off Odaka off Ukedo off 1FNPP off Kumagawa off 2FNPP off Kidogawa off Yotsukura Ibaraki off Nakoso Hitachinaka Kashima Choshi

2012 year class

Concentration (Bq g-wet1)

Date (month/year)

Concentration (mBq g-wet1)

Date (month/year)

Concentration (Bq g-wet1)

Date (month/year)

St.1 T-MG1 T-22

0.006 0.03 0.05

06/2011 07/2011 06/2011

0.1 0.2 0.7

07/2012 08/2012 08/2012

0.005 0.004 0.08

10/2011 10/2011 10/2011

T-S1 T-S2 T-S3 T-S4 T-S8 T-S7 T-S5 St.4

0.04 0.07 0.4 0.4 0.6 2 2 0.4

06/2011 05/2011 04/2011 04/2011 04/2011 04/2011 04/2011 04/2011

1 1 1 1 1 4 4 5

12/2012 12/2012 12.2012 12/2012 12/2012 12/2012 12/2012 05/2012

0.05 0.1 0.4 0.5 0.6 2 2 0.5

10/2011 09/2011 08/2011 08/2011 08/2011 08/2011 08/2011 09/2011

St.5 St.6 St.7 St.8

0.2 0.02 0.01 0.007

04/2011 04/2011 04/2011 09/2011

2 0.7 0.3 0.3

04/2012 04/2012 03/2012 04/2012

0.3 0.03 0.02 0.01

09/2011 10/2011 10/2011 12/2011

the concentrations in planktivorous fish hatched after 2012. A similar depuration of radiocesium is seen in olive flounder inhabiting along the Pacific coasts of eastern Japan (Fig. 6) and within a 30 km area of the 1FNPP (Fig. 7), however their depuration rates were slower than those simulated for planktivorous fish. The slower depuration in olive flounder compared to that of planktivorous fish is most likely due to the contribution of radiocesium transferred from food. Because of the dominance of radiocesium transfer flux from food noted at St. 4 in a previous study (Tateda et al., 2013), the 137 Cs transfer from food to olive flounder is believed to be the primary source of this radionuclide during the first several months in all of the simulated areas in this study. Simulation at all study sites also indicated that radiocesium transfer from food even during the depuration stage is occurring and serves to slow the rate of depuration in this species (Fig. S3 (a)). Thus, the determining factor for radiocesium depuration from olive flounder is consequently influenced by the radiocesium depuration rate in its food organisms as shown in subfigures (2) in Fig. 7 (a), (b), and (c). Cesium-137 depuration states in olive flounder inhabiting the contaminated coastal waters by the simulation are summarized in Table 3. Briefly, the depuration of radiocesium from olive flounder inhabiting waters in front of the 1FNPP and in the southern coastal waters was estimated to begin after September 2011. In the Miyagi and Ibaraki Prefectures, depuration began later, i.e. after November 2011. Adopting a depuration rate from Tateda et al. (2013), namely the depuration rate derived as a constant by approximation after the depuration starting date (Fig. S3(b)), the ecological half-life (Teco1/2) for olive flounder during the second year after the accident was calculated. The observed 137Cs concentrations in olive flounder were not exactly correct in strict sense to use in Teco1/2 calculation after 2 y later, because the monitoring measurements were below the detection limit (BDL) (MAFF, 2015). This makes the approximation of the depuration rate from monitoring values problematic due to increasing incidences of BDL's especially in the later phase of depuration after the second year. Thus, in this case the model simulation is useful to derive the Teco1/2 by using the calculated values during the depuration phase in which the 137Cs levels in olive flounder can be regarded as exponentially reduced (Fig. S3(b)). For olive flounder in the Fukushima coastal waters, the

Teco1/2 ranged from 140 to 160 d, whereas they were >160 d at >30 km in the north area in Miyagi, and at the >50 km south coastal waters in Ibaraki. The longer Teco1/2 of olive flounder inhabiting areas further distant is attributable to the larger contribution of radiocesium transfer from food relative to the excretion flux, being 1.3 (0.4/0.3) times greater than that in the Fukushima coastal area (Fig. S3(a)). The Teco1/2 of 140e160 d equivalent to a depuration rate of 0.0043e0.0050 d1 for olive flounder in the Fukushima coastal waters during second year 2012 are longer than the biological half-lives (Tb1/2) of 63e73 d (equivalent to a metabolic rate of 0.0095e0.011 d1) previously reported for this species (Suzuki et al., 1992). The difference between Teco1/2 and Tb1/2 is due to the radiocesium transferred from food to flounder during the depuration phase (Fig. S3(a)). In the case of the T-S4 area, the simulated flux of transfer from food was on the order of 0.1e1 Bq kg-wet-1d1. Assuming a daily food intake of 0.05 (kg-wet1food) / (kg-wet1predator) and an assimilation rate of 0.5, the 137Cs concentration in food of flounder at T-S4 was 4e40 Bq kg-wet1food (estimated radiocesium concentration in food (Bq kg-wet1food)) ¼ (simulated transfer flux from food: 0.1e1 Bq kg-wet1predator d1)/(daily food intake: 0.05)/(assimilation rate: 0.5) during 2012. This estimation agrees with the observed 137Cs concentrations in the stomach contents of flounder at T-S4 (Fig. 7(c) (2)). The rate of 137Cs depuration in olive flounder within a 30 km area of 1FNPP was simulated to be still controlled by radiocesium transfer through food chain. However, overall radiocesium levels in olive flounder, as a typical piscivorous fish in the Fukushima coastal waters, was mostly depurated as shown by the fish monitoring data not only in Fukushima coastal waters but also within 30 km area of 1FNPP, with the exception of those fish collected from inside the port (TEPCO, 2015). 4. Conclusions Using the observed and simulated 137Cs concentrations in sedentary organisms, based on the temporal record of introduced radiocesium from the Fukushima accident, the seawater concentrations of the Pacific coasts along eastern Japan were reconstructed by approximation of simulated levels compared to observed values.

140

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

Fig. 7. (1) Reconstructed 137Cs seawater level by simulation with observed concentration, (2) simulated 137Cs level in stomach content of olive flounder with measured concentration, and (3) simulated and observed 137Cs concentration in muscle of the olive flounder, collected in (a) north, (b) front, (c) south area within a 30 km of the 1FNPP. The data of below detection limit (BDL) are shown by a bar symbol (). After, the first occurrence of a BDL, measured value are shown by triangle symbol ( ).

▵

Table 3 Estimated start time of radiocesium depuration in olive flounder with ecological half-life (EHL) derived by simulation. Prefecture Location Miyagi Kesennuma Ishinomaki Watari Fukushima off Ootagawa off Odaka off Ukedo off 1FNPP off Kumagawa off 2FNPP off Kidogawa off Yotsukura Ibaraki off Nakoso Hitachinaka Kashima Choshi

Site

Starting month of depuration (month/year)

Ecological half-life (d)

St.1 T-MG1 T-22

11/2011 11/2011 11/2011

410 160 170

T-S1 T-S2 T-S3 T-S4 T-S8 T-S7 T-S5 St.4

11/2011 09/2011 09/2011 09/2011 09/2011 09/2011 09/2011 09/2011

160 150 140 150 140 140 140 140

St.5 St.6 St.7 St.8

10/2011 11/2011 10/2011 11/2011

160 180 170 190

Y. Tateda et al. / Journal of Environmental Radioactivity 147 (2015) 130e141

Through the reconstruction procedure, deficits of 137Cs concentrations in seawater were identified. These deficits suggested an additional source to compensate the observed 137Cs levels in coastal water mostly in front of and to the south of the 1FNPP. A supplementary local source to the Fukushima coastal waters other than atmospheric deposition and direct leakage was estimated to be within a range of 1e10 GBq. The simulated 137Cs concentrations in planktivorous fish e.g. anchovy, whitebait, sand lance, and sardine etc., were deduced and the simulated 137Cs levels in 2012 year class which hatched after the accident converged to the similar concentrations. The maximum 137Cs concentrations in plantivorous fish and their predator fish such as olive flounder showed similar levels to the initial seawater concentrations to which they were exposed. The ecological half-lives for olive flounder during the second year after the accident ranged within 5e6 months, and food composition was the determining factor of radiocesium depuration rate in this species inhabiting the Pacific coast of eastern Japan. Acknowledgment We express our gratitude to the radioecologists of the former Nakaminato Marine Radioisotope Facility, National Institute of Radiological Science, Japan, and those in the Radioecology Laboratory of the former IAEA Marine Environmental Laboratories in Monaco for their extensive radiotracer studies aimed at producing radio nuclide transfer parameters. For confirming the modelling strategy, the unpublished data comparisons made within the research group of the Interdisciplinary Study on Environmental Transfer of Radionuclides from the Fukushima Daiichi NPP Accident (Grant-in-Aid for Scientific Research on Innovative Areas 24110005) were helpful. We also thank the TEPCO and Tokyo Power Technology for sharing the stomach samples collected through their fish monitoring activities. This study was supported by the Central Research Institute of the Electric Power Industry. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jenvrad.2015.05.025. References Adachi, K., Kajino, M., Zaizen, Y., Igarashi, Y., 2013. Emission of spherical cesiumbearing particles from an early stage of the Fukushima nuclear accident. Sci. Reports 3, 2554. http://dx.doi.org/10.1038/srep02554. Aoyama, M., Tsumune, D., Uematsu, M., Kondo, F., Hamajima, Y., 2012. Temporal variation of 134Cs and 137Cs activities in surface water at stations along the coastline near the Fukushima Dai-ichi nuclear power plant accident site, Japan. Geochim. J. 46, 321e325. Baumann, Z., Casacuberta, N., Baumann, H., Masque, P., Fisher, N., 2013. Natural and Fukushima-derived radioactivity in macroalgae and mussels along the Japanese shoreline. Biogeosciences 10, 3809e3815. Buesseler, K., 2012. Fishing for answers off Fukushima. Science 338, 480e482. Buesseler, K., Jayne, S., Fisher, N., Rypina, I., Baumann, H., Baumann, Z., Breier, C., Douglas, E., George, J., Macdonald, A., Miyamoto, H., Nishikawa, J., Pike, S., Yoshida, S., 2011. Fukushima-derived radionuclides in the ocean and biota off Japan. Proc. Natl. Acad. Sci. 109, 5984e5988. FP, 2014. The Results (Seawater) of Environmental Radioactivity Monitoring (Port and Fishery Waters) by Fukushima Prefecture, XML, CSV File. Japan Atomic Energy Agency. http://emdb.jaea.go.jp/emdb/portals/70100000012/ (in Japanese).

141

Ishido, Y., 1990. Distribution and migration of the young Japanese flounder (Paralichtys olivaceus) along northeastern coast of Japan. Bull. Tohoku Natl. Fish. Res. Inst. 52, 33e43 (in Japanese). Kanda, J., 2013. Continuing 137Cs release to the sea from the Fukushima Dai-ichi nuclear power plant through 2012. Biogeoscience Discuss. 10, 3577e3595. Katata, G., Chino, M., Kobayashi, T., Terada, H., Ota, M., Nagai, H., Kajino, M., Draxier, R., Hort, M., Malo, A., Torii, T., Sanada, Y., 2014. Detailed source term estimation of the atmospheric release for the Fukushima Daiichi nuclear power station accident by coupling simulations of atmospheric dispersion model with improved deposition scheme and oceanic dispersion model. Atmos. Chem. Phys. Discuss. 14, 14725e14832. Kitagawa, D., Ishido, Y., Sakurai, Y., 1994. Age, growth and maturation of the Japanese flounder Paralichthys olivaceus in coastal waters of the northern Sanriku district. Bull. Tohoku Natl. Fish. Res. Inst. 56, 69e76 (in Japanese). Madigan, D., Baumann, Z., Fisher, N., 2012. Pacific bluefin tuna transport Fukushima-derived radionuclides from Japan to California. Proc. Natl. Acad. Sci. 109, 9483e9486. MAFF, 2015. Results of the Monitoring on Radioactivity Level in Fisheries Products. List of Monitoring Results, Cesium. Excel File. Japan Fisheries Agency. http:// www.jfa.maff.go.jp/e/inspection/index.html. ME, 2014. The Results of Monitoring (Seawater) by Ministry of the Environment, XML, CSV File. Japan Atomic Energy Agency. http://emdb.jaea.go.jp/emdb/ portals/20760702/ (in Japanese). Nagao, S., Kanamori, M., Ochiai, S., Tomihara, S., Fukushi, K., Yamamoto, M., 2013. Export of 134Cs and 137Cs in the Fukushima river systems at heavy rains by Typhoon Roke in September 2011. Biogeosciences 10, 2767e2790. NRA, 2013. Synopsis of Public Hearing on Monitoring by TEPCO, 22 November 2013. Nuclear Regulatory Authority, Japan. http://www.nsr.go.jp/data/000048869.pdf (in Japanese). Oikawa, S., Takata, H., Watabe, T., Misonoo, J., Kusakabe, M., 2013. Distribution of the Fukushima-derived radionuclides in seawater in the Pacific off the coast of Miyagi, Fukushima, and Ibaraki prefecture, Japan. Biogeoscience 10, 5031e5047. Otosaka, S., Kobayashi, T., 2013. Sedimentation and remobilization of radiocesium in the coastal area of Ibaraki, 70 km south of the Fukushima Dai-ichi nuclear power Plant. Environ. Monit. Assess. 185, 5419e5433. Sohtome, T., Wada, T., Mizuno, T., Nemoto, Y., Igarashi, S., Nishimune, A., Aono, T., Ito, Y., Kanda, J., Ishimaru, T., 2014. Radioecological impact of TEPCO's Fukushima Dai-ichi nuclear power plant accident on invertebrates in the coastal benthic foodweb. J. Environ. Rad. 138, 106e115. Suzuki, Y., Nakamura, K., Nakamura, R., Nakahara, M., Ishii, T., Matsuba, M., Nagaya, Y., 1992. Radioecological studies in the marine environment. In: Proc. Intern. Conf. Rad. Eff. Protec. Japan Atom. Energ. Res. Inst. Tokyo, pp. 484e491. Takeno, K., 2010. Studies on the Fisheries Biology of Japanese Flounder Paralichthys olivaceus in Western Wakasa Bay. Special Report No. 8. Kyoto Institute of Oceanic and Fishery Science, p. 74 (in Japanese). Tateda, Y., Koyanagi, T., 1996. Concentration factors for 137Cs in Japanese coastal fish (1984e1990). J. Radiat. Res. 37, 71e79. Tateda, Y., Tsumune, D., Tsubono, T., 2013. Simulation of radioactive cesium transfer in the southern Fukushima coastal biota using a dynamic food chain transfer model. J. Environ. Rad. 124, 1e12. TEPCO, 2015. Archive of the Result of Radioactive Substance Analyses Around the Fukushima Daiichi Nuclear Power Plant, Seawater CSV File. Tokyo Electric Power Company. http://www.tepco.co.jp/decommision/planaction/monitoring/ index-j.html (in Japanese). Tomiyama, T., Kurita, Y., 2011. Seasonal and spatial variations in prey utilization and condition of a piscivorous flatfish Paralicthys olivaceus. Aquat. Biol. 11, 279e288. Tsumune, D., Tsubono, T., Aoyama, M., Hirose, K., 2012. Distribution of oceanic 137Cs from the Fukushima dai-ichi nuclear power plant simulated numerically by a regional ocean model. J. Environ. Radioact. 111, 100e108. Tsumune, D., Tsubono, T., Aoyama, M., Uematsu, M., Misumi, K., Maeda, Y., Hayami, H., 2013. One-year, regional-scale simulation of 137Cs radioactivity in the ocean following the Fukushima Dai-ichi nuclear power plant accident. Biogeoscience 10, 5601e5617. Wada, T., Nemoto, Y., Shimamura, S., Fujita, T., Mizuno, T., Sohtome, T., Kamiyama, K., Morita, T., Igarashi, S., 2013. Effects of the nuclear disaster on marine products in Fukushima. J. Environ. Rad. 124, 246e254. Yamashiki, Y., Onda, Y., Smith, H., Blake, W., Wakahara, T., Igarashi, Y., Matsuura, Y., Yoshimura, K., 2014. Initial flux of sediment-associated radiocesium to the ocean from the largest river impacted by Fukushima Daiichi nuclear power Plant. Sci. Rep. 4, 3714. http://dx.doi.org/10.1038/srep03714.