Natural and Fukushima-derived radioactivity in macroalgae and mussels along the Japanese shoreline

Introduction Conclusions References


Introduction
Following the devastation caused by the earthquake and tsunami on 11 March 2011, the Japanese public was anxious regarding radioactivity release from the damaged Fukushima Daiichi nuclear power plant (NPP).Radioactivity released onto land led to contamination in vegetable produce, dairy, and meat products, and similarly release into the coastal ocean led to contamination of seafood.Recognizing public concerns, the Ministry of Agriculture, Forestry and Fisheries (MAFF) initiated a monitoring program that set allowable radioactivity levels in seafood to 500 Bq kg −1 wet weight (wet wt., later lowered to 100 Bq kg −1 wet weight) for radioactive cesium (i.e., sum of 134 Cs and 137 Cs) (MAFF).Post-Fukushima academic studies have either attempted to predict consequences of the fallout for human and animal health (Kryshev and Sazykina, 2011;Garnier-Laplace et al., 2011;Murakami and Oki, 2012) or focused on estimating the quantities of released radioactivity and their atmospheric and oceanic dissipation (Buesseler et al., 2011(Buesseler et al., , 2012;;Tsumune et al., 2011;Yoshida and Kanda, 2012;Masson et al., 2011).However, peer-reviewed studies of radioactive concentrations in post-Fukushima marine animals and plants remain surprisingly scarce.Apart from detailed radioactivity measurements in small fish and crustaceans collected offshore during an international survey in June 2011 (Buesseler et al., 2012), peer-reviewed reports of radioactivity in economically important coastal organisms (e.g., macroalgae, bivalves) are missing.However, TEPCO (Tokyo Electric Power Company), MEXT (Ministry of Education, Culture, Sports, Science, and Technology) and MAFF have used the Internet as an outlet for continuous publication of radioactivity measurements in sediments, water and biota.While there is undeniably a value to a stream of data published on the web page, this approach is missing key information on methodology and therefore missing quality assurance (MEXT; MAFF; TEPCO, 2013).
The goal of this study was to quantify June 2011 levels of Fukushima-derived γ -emitting cesium isotopes (i.e., 134 Cs, 137 Cs) in mussels and macroalgae, collected from coastal locations north and south of the Fukushima NPP (Fig. 1).A second goal was to put these levels in the context of naturally occurring radioactivity due to γ -and β-emitting potassium ( 40 K) and 210 Pb and α-emitting polonium ( 210 Po) isotopes present in these organisms, which is directly relevant to assessing any increased radioactive dose for seafood consumers.

Sample collection
Samples of green (Ulva sp.) and brown (species were not identified) macroalgae as well as mussels (Mytilus galloprovincialis; average shell length 45.0 ± 3.8 mm) were collected by hand along several Pacific beaches south (2 June 2011) and north of Fukushima NPP (21 June 2011) (Fig. 1).Collection times were different for the northern and southern locations due to logistical capabilities.Southern sampling sites, i.e., Isohara port (36.81

Sample preparation
In the laboratory, mussel shells were separated from soft tissue, since only the latter was assumed to have accumulated radioactive Cs (Ke et al., 2000).Mussel tissue and macroalgae were placed into individual plastic containers and stored at −80 • C for 24 h prior to freeze-drying.All samples were then placed inside fast-freeze flasks and connected to a Labconco FreeZone Benchtop Freeze Dry System that was cooled to −52 • C, and pressurized to 0.12 mbar (Buesseler et al., 2012).After freeze-drying, samples were ground with a ceramic mortar and pestle to obtain homogeneously distributed samples that varied in dry weight (dry wt.) from 3.1 to 79.1 g.Samples were stored in straightside polypropylene 120 mL jars (Nalgene), which had been efficiency-calibrated to their specific geometry.

134 Cs, 137 Cs and 40 K analysis
We used a planar, low energy germanium detector (Canberra model GLP 3830) and Genie 2000 software (Canberra) to measure and analyze the spectrum of γ emissions in our samples.Average counting time was 27 h, and the average counting uncertainties were 4 % for 134 Cs and 137 Cs, and 8 % for 40 K.The detector efficiency was determined as previously described by Buesseler et al. (2012) for individual radioisotopes with varying geometries as follows: 1.7-3.5 % for 134 Cs (at 604.7 keV peak), 1.6-3.2% for 137 Cs (at 661.7 keV peak), and 0.8-1.7 % for 40 K (at 1461 keV peak).Calculations of 134 Cs, 137 Cs, and 40 K concentrations in the samples assumed negligible self-absorption of γ rays for these isotopes, given that they all emit at energies > 200 keV (Hurtado et al., 2007).Correction for coincidence summing was omitted as a low efficiency planar detector was used for sample analyses.Measured 134 Cs and 137 Cs were decay-corrected to  210 Po in macroalgae and mussels sampled in June 2011 along the east coast of Japan.Activities were decay-corrected to the date of sample collection, except for 40 K given its very long half-life (1.25 × 10 9 yr).σ is the propagated counting error (Bq kg −1 ). the date of sample collection (taking into consideration their half-lives of 2.07 and 30.08 yr, respectively).

Analyses of 210 Pb and 210 Po
Small sub-samples (0.5-1.0 g) were spiked with known amounts of 209 Po (0.07 Bq) and stable Pb (0.1 mg) as internal tracers of the radiochemical analysis.After the addition of tracers, sub-samples were acid-digested at 80-90

Cesium in macroalgae and mussels
All samples of post-Fukushima mussels and macroalgae contained 134 Cs (range: 51-393 Bq kg −1 dry wt.) and 137 Cs (range: 60-463 Bq kg −1 dry wt., Table 1).Measured 137 Cs concentrations in mussels and macroalgae prior to the Fukushima accident (background) were two to three orders of magnitude lower.For example, brown macroalgae collected off the Japanese coast in 2007-2008 had 137 Cs concentrations of 0.038-0.058Bq kg −1 wet wt.(or 0.14-0.21Bq kg −1 dw assuming a wet/dry ratio of 0.27; Morita et al., 2010;Carvalho, 2011), and mussels (Mytilus edulis) collected from the south-western Korean Sea in 1988 had 0.04 Bq kg −1 dry wt.In the case of the shorter lived 134 Cs isotope (half-life = 2.07 yr), no other sources than the Fukushima accident could have produced the detected 134 Cs levels in mussels and algae.In addition, the ratio of 134 Cs : 137 Cs in our macroalgae samples was approximately 1 (mean ± SD = 0.85 ± 0.003 and 0.88 ± 0.10 for the northern and southern sampling locations, respectively), consistent with ratios found in marine plankton (Buesseler et al., 2012) and similar to measured concentrations in the water following the radioactive discharge from the Fukushima NPP (Buesseler et al., 2011;MEXT;Aoyama et al., 2012).Furthermore, our analyses revealed consistently lower radiocesium levels (i.e., the sum of 137 Cs and 134 Cs) in samples from northern locations than those in the southern locations.However, even in the more contaminated southern locations none of the samples exceeded the 500 Bq kg −1 wet wt.safety limit.Differences in Cs contamination between the two regions were greatest for mussels.M. galloprovincialis from the site north of the Fukushima NPP contained 7.5 times less radiocesium than southern samples (north: mean of 110 Bq kg −1 dry wt.; south: mean of 822 Bq kg −1 dry wt.).Green macroalgae from northern locations contained on average 141 Bq kg −1 dry wt. of radiocesium, which was 5.7 times less than in the southern sample (804 Bq kg −1 dry wt.), while brown macroalgae from the south contained approximately twice as much radiocesium as northern samples.
Given that Cs uptake by macroalgae is proportional to the Cs concentrations in water (Gutknecht, 1965), we conclude that radiocesium levels were indeed higher in coastal waters south of Fukushima than in waters north of Fukushima at the time of sampling, likely due to the 19-day lag between sampling at these locations.Indeed, MEXT-online reports show a ∼ 30 % decline in 137 Cs seawater concentration at Dai-ichi within this time interval (i.e., 2-21 June 2011).Some of the difference in 137 Cs concentrations in biota collected at these two time points (i.e., the 137 Cs concentration difference between biota collected north on 21 June 2011 vs. those collected south on 2 June 2011) can be accounted for by the general decline of radioactivity at its discharge.Any further difference could be accounted for by the southward shore current along the Pacific coast of Honshu (Buesseler et al., 2012).Cs concentrations in biota can be combined with published concentration factors (CFs, i.e., the wet tissue to water activity ratio; L kg −1 ) to estimate Cs concentrations in the water, assuming that the organisms were in equilibrium with ambient water (IAEA, 2004).Given that the biological half-lives of Cs in green and brown macroalgae and in mussels are 5-8 and 8 days, respectively (Gutknecht, 1965;Dahlgaard, 1981), it would appear that the assumption of equilibrium was met.However, for longer biological half-life values (e.g., as reported by Dahlgaard and Nolan, 1989, for Cs in M. edulis -17.5 d), assumption of equilibrium could lead to a greater error.
While CFs might be useful in approximating the concentration levels in the water, it is important to realize that these approximations come with an increased chance of error.For example, CFs of 137 Cs in various brown macroalgae varied up to 100 % and in mollusks up to 67 % in a study following radioactive waste disposal in Arctic seas (Fisher et al., 1999).The variation of CF values of 137 Cs in green mussels -Perna viridis -has been also related to body size, water pumping rate, and salinity (Ke et al., 2000).Due to observed variability of 137 Cs CFs in algae and mussels, which however is smaller for an individual species, estimates of 137 Cs concentrations in seawater based on literature CFs are therefore expected to fall within fourfold of directly measured values.
By applying our measured Cs concentrations in mussels and using a Cs wet wt.CF for mussels as 60 (IAEA, 2004), we estimate total ambient radiocesium levels in northern and southern coastal sampling sites in June 2011 to have ranged from 2.9 × 10 2 Bq m −3 to 22.1×10 2 Bq m −3 , respectively (Table 2).Estimates for seawater 134 Cs concentrations using our mussel concentrations of 134 Cs were 1.4 × 10 2 north of Fukushima NPP and 9.9 × 10 2 -10.5 × 10 2 Bq m −3 south of Fukushima NPP (Table 2).Reported CF values for green and brown algae are 60 and 34, respectively (IAEA, 2004).Using these CFs and a dry to wet weight ratio of 0.18 for green algae and 0.27 for brown algae (Carvalho, 2011), we estimate seawater 134 Cs concentrations of 1.8 × 10 2 -2.0 × 10 2 Bq m −3 north of Fukushima and 11 × 10 2 Bq m −3 south of Fukushima based on green algae measurements (Table 2).Brown-algae-based measurements estimated 134 Cs concentrations of 6.4 × 10 2 -10.1 × 10 2 Bq m −3 north of Fukushima and 19 × 10 2 Bq m −3 south of Fukushima (Table 2).Green-algae-based estimates of 137 Cs in seawater north of Fukushima are 2.2 × 10 2 -2.5 × 10 2 Bq m −3 and 13.1 × 10 2 Bq m −3 south of Fukushima, while brown-algaebased estimates of 134 Cs in seawater north of Fukushima are 6.9 × 10 2 -12.1 × 10 2 Bq m −3 and 22.4 × 10 2 Bq m −3 south of Fukushima (Table 2).These estimated 137 Cs concentrations in coastal seawater are generally about 2 orders of magnitude lower than the maximum 137 Cs concentrations recorded by TEPCO and presented by MEXT in mid-April at 30 km offshore Fukushima (∼ 200 000 Bq m −3 ; Buesseler et al., 2011).It is practically impossible to relate our lower estimates at northern locations to those reported by MEXT, because 137 Cs detection limits declared in these reports were high (i.e., 6 Bq L −1 or 6 × 10 3 Bq m −3 ).Therefore, any Cs at concentrations lower than 6 × 10 3 Bq m −3 would not have been detected. 137Cs concentrations predicted by us for the southern location on 2 June 2011 are at least sevenfold lower than 15 × 10 3 Bq m −3 reported online for seawater collected on 2 June 2011 near our sampling site (MEXT). 134Cs concentrations measured in coastal seawater Onahama (Aoyama et al., 2012), near the southern sampling site in the present study, show up to threefold higher than the maximum CF-estimated values in the present study (Table 2).This agreement between our CF-estimation of seawater Cs and an actual measurement is consistent with an up to fourfold error (see above).
The variability in the 40 K and 210 Po levels for brown macroalgae (Table 1) is not unusual (Thompson et al., 1982;Ugur et al., 2002;Goddard and Jupp, 2001).Thompson et al. (1982) showed that for just one species of a brown macroalgae, Fucus vesiculosus, collected from the Irish Sea, 40 K concentrations varied twofold.Another study showed that 40 K concentrations can vary within algal divisions (Goddard and Jupp, 2001).In that study 40 K concentrations showed a 44-fold variation and ranged from 100 to 4400 Bq kg −1 dry wt. in green algae, and 125-5400 Bq kg −1 dry wt. in brown algae (Goddard and Jupp, 2001).Published 210 Po concentrations for green algae are scarce, but one study reported 210 Po concentration of 15 Bq kg −1 dry wt., which falls within the range of 210 Po concentration in green algae in this study (Carvalho, 2011).As in the present study, variability for 210 Po concentrations in mussels has been reported.For example, 210 Po concentrations showed 26-fold (52-1300 Bq kg −1 dry wt.) variation for Mytilus galloprovincialis from various coastal locations of the Aegean Sea (Ugur et al., 2002).Wildgust et al. (1998) suggested that variation of 210 Po concentrations in organisms could be linked to environmental conditions such as salinity or temperature, and to biological parameters such as the reproductive cycle phase.The ratios of 210 Po : 210 Pb for green and brown macroalgae fit within the range of previously reported ratios, but for mussels 210 Po : 210 Pb ratios are lower in this study (Carvalho, 2011).
From a human-health perspective, it is important to assess levels of natural radionuclides such as 40 K, 210 Po and 210 Pb as these radionuclides are always present in seafood.Because 40 K is the dominant γ -emitter in the ocean and in marine biota, its evaluation as a potential radioactive dose contributor is important.The dose attributable to 40 K is however predominantly due to its β emissions (UNSCEAR, 1993). 210Po releases α-particles with its decay, and α emissions are 20 times more damaging to cells than γ and β emissions (Eckerman et al., 2012).In fact 210 Po has been shown as the radionuclide producing by far the greatest radioactive dose in marine biota, as it also is the greatest radioactive dose contributor to seafood consumers (Aarkrog et al., 1997).Therefore even though its radioactivity is lower, doses from the naturally occurring 210 Po in biota, including those from the Japanese shore (Table 1), can be of greater radiological importance than 134+137 Cs.

Summary
This is the first report in the peer-reviewed literature on the levels of anthropogenic and natural radioactivity in the biota from coastal Japan following the Fukushima NPP failure.At the time of sample collection in June 2011, concentrations of Fukushima-derived 137 Cs and 134 Cs were elevated, and when combined (i.e., 134+137 Cs) they did not exceed the 500 Bq kg −1 wet wt.recommended safety level for seafood consumption.Biota concentrations of Cs were higher in the southern locations than in the north, likely due to the 19day lag between sampling and the southward flow of coastal currents (Buesseler et al., 2012).Concentrations of individual (i.e., 210 Po and 40 K) natural radionuclides showed levels that were 20-fold lower or 5-fold higher than those of 134+137 Cs.A concentration-factor-based approach estimated that radioactive Cs concentrations in coastal waters at the northern location in the second half of June could be in the range of 2.9 × 10 2 -22.2 × 10 2 Bq m −3 , and at the southern www.biogeosciences.net/10/3809/2013/Biogeosciences, 10, 3809-3815, 2013 locations earlier on were either similar or up to 14-fold higher.

Fig. 1 .
Fig. 1.Sampling locations for mussels and macroalgae along the east coast of Honshu Island of Japan.

Table 2 .
Estimates of134Cs and 137 Cs concentrations in coastal seawater at the locations north and south of the Fukushima NPP based on concentration factors (CFs; L kg −1 ) derived from the literature(IAEA, 2004).Ranges are provided based on the maximum and minimum concentration values for the biota.
a North (N) or south (S) relative to the Fukushima nuclear power plant (NPP).b Indicates one value instead of a range when concentrations were available for only one sample.