Tritium activity concentration and behaviour in coastal regions of Fukushima in 2014

We observed 3H activity concentrations and the 137Cs activity concentrations during the SoSo 5 rivers cruise in 2014 and at the Tomioka port in 2014-2018. The 3H activity concentrations at coastal stations located close to the Fukushima coast ranged from 90 Bq m-3 to 175 Bq m-3, and decreased between 67 Bq m-3 to 83 Bq m-3 at the stations located 12–16 km from the coast. The 3H activity concentration at 20 the estuaries and ports, except at 56 north canal of the FNPP1 site, are around 200-500 Bq m-3 and slightly lower than the 3H activity concentration of 500-600 Bq m-3 observed in the rivers. These gradients of the 3H activity concentrations in the coastal region might indicate the large effect of 3H flux through the rivers. On the other hand, the 3H activity concentration at 56N of the FNPP1 site was significantly high compared to the 3H activity concentration in surrounding waters both north and south of the FNPP1 site and in river 25 waters. It should also be noted that the 3H activity concentrations were similar at the stations located both north and south of the FNPP1 site, while the 137Cs activity concentrations were lower at the stations north https://doi.org/10.5194/bg-2021-10 Preprint. Discussion started: 23 February 2021 c © Author(s) 2021. CC BY 4.0 License.


Introduction
Tritium ( 3 H) is a radioactive isotope of hydrogen with a half-life of 12.4 yr. The major sources of 3 H are cosmogenic, nuclear weapons tests, military production, and nuclear reactor operation, particularly heavy-https://doi.org /10.5194/bg-2021-10 Preprint.  water reactor operation and nuclear fuel reprocessing plant (Galeriu and Melintescu, 2011). 55 Approximately 78 PBq ( PBq = 10 15 Bq) of 3 H is produced each year by cosmic ray spallation and the inventory of 3 H in the atmosphere is 1.3 EBq ( EBq = 10 18 Bq); further, it was estimated that 240 EBq of 3 H was injected by nuclear weapons tests (Galeriu and Melintescu, 2011). From the nuclear fuel reprocessing plants, 0.276 EBq was discharged mainly into the Atlantic ocean between 1970 and 2008 . In the nuclear reactor operations, during 1990-1997, 51.8 PBq of 3 H was released into 60 the atmosphere, and 79.2 PBq of 3 H was released into the ocean, as stated in Tables 32 and Table 35 in annex C of the UNSCEAR 2000 report (UNSCEAR, 2000), most of which was released by heavy-water reactors. Before the commencement of atmospheric testing of thermonuclear weapons in 1952, the 3 H content of precipitation was in the range of 180-1000 Bq m -3 (Galeriu and Melintescu, 2013); this background concentration resulted from cosmic ray spallation and showed a maximum at the mid-65 latitudes of both the hemispheres. While elevated 3 H levels have been measured in the atmosphere since 1952, 3 H produced by thermonuclear testing has been the dominant source of 3 H in precipitation (Katsuragi et al., 1983;Galeriu and Melintescu, 2011).
The peak activity concentration of 201600 Bq m -3 was recorded in the precipitation in Tokyo in the northern hemisphere in March 1964(Katsuragi et al., 1983. After the atmospheric nuclear weapons tests 70 were partly banned in 1963, the 3 H levels in the precipitation began to decline gradually because of radioactive decay and atmospheric dispersion, and was transferred into the ocean and groundwater via the water cycle among the domains of atmosphere, ocean, and land with an apparent residence time of 23.0 ± 2.3 months (Katsuragi et al., 1983). In the southern hemisphere, 3 H levels in the precipitation in Australia were measured over the past 50 years (Tadros et al., 2014), demonstrating that the elevated levels of 3 H in 75 the environment were due to last century's atmospheric thermonuclear testing. The peak in the 3 H activity concentration reached a maximum of 189000 Bq m -3 (160 TU ) in 1963 owing to the peak in the atmospheric tests, and then, from 1963 to present, a rapid drop in 3 H activity concentration was observed.
This drop cannot entirely be from natural decay; therefore, it is attributed to the washout of 3 H into the oceans and groundwater. Since 1990, the levels of the 3 H activity concentration have declined globally 80 and regionally; currently the levels of 3 H in Australia are stable and in the range of 240-350 Bq m -3 (2-3 TU), suggesting that today the 3 H levels in the precipitation are predominantly due to naturally occurred https://doi.org /10.5194/bg-2021-10 Preprint. (Tadros et al., 2014). Some observations suggested that the effect of atmospheric weapon tests in the 3 H activity concentration in the precipitation disappeared after 2000.
Further, explosions from the nuclear tests led to tritium emissions in the atmosphere as tritiated hydrogen 85 (HT) and methyl tritium gas (CH3T) were rapidly oxidised and converted into tritiated water molecules (HTO) that closely follow the whole water cycle and water masses (Ducros et al., 2018). Sugihara et al. (Sugihara et al., 2008)  They also reported that the 3 H activity concentrations in the river waters showed a meridional 90 distribution, and the 3 H activity concentration was high at 35-43 °N and low in the south of 35 °N. In open-water, 3 H was traditionally used as a tracer for ocean circulation, largely in the Pacific ocean (Michel and Suess, 1975;Van Scoy et al., 1991;Stark et al., 2004) and Sea of Japan (Kajì et al., 2005).
Since the largest source of 3 H in the environment atmospheric weapon tests, as stated previously, the 3 H activity concentration in the surface seawater showed a similar trend of freshwater on land, as expected. 95 Therefore, a peak in the 3 H activity concentration in the range of 500-12000 Bq m -3 in the surface seawater was recorded in many places in 1963 in the northern hemisphere (Stark et al., 2004). The 3 H activity concentration in the surface seawater decreased rapidly as along with the trend of the 3 H activity concentration in the precipitation/river water. Povinec et al. (Povinec et al., 2013) estimated the pre-Fukushima 3 H activity concentration in the surface water in the north-western Pacific ocean as 70 ± 10 Bq 100 m -3 based on the data stored in the GLOMARD/MARIS database. In certain coastal regions, the 3 H activity concentration is affected by the liquid and gaseous discharges from the nuclear fuel reprocessing plants Masson et al., 2005) and the operation of nuclear power plants. Momoshima et al. (Momoshima et al., 1987) measured the 3 H activity concentration in various environmental materials around a typical nuclear power station, the Genkai Nuclear power plant, in Japan in 1983 and 1984. They 105 found that the elevated 3 H activity concentration was observed on one occasion in pine needles and surface soil after noting a high 3 H activity concentration of 220-10000 Bq m -3 in the seawater . However, no incidental increase in 3 H levels was observed in atmospheric water vapour, hydrogen, and methane. Masson et al. (Masson et al., 2005) presented free-water 3 H and organically bound 3 H levels in the French coastal marine environment, from Concarneau to Gravelines, along with the 3 H levels in the seawater. 110 The matrices selected for their specific survey included seawater, seaweed, molluscs, crustaceans, and https://doi.org /10.5194/bg-2021-10 Preprint.  They also reported the 3 H activity concentrations of approximately 150 Bq m −3 south of Ireland and in the Atlantic Ocean. This activity concentration of 150 Bq m −3 is similar to, but slightly higher than, those reported in the north-western Pacific Ocean (70 ± 10 Bq m −3 ; Povinec et al., 2013) and may have been affected by the heavy discharges from the Europian fuel reprocessing plants of La Hague and Sellafield.
In this paper, we present results of the 3 H activity concentration observed during the SoSo 5 rivers cruise 120 and at the Tomioka port and Hasaki and discuss the behaviour of 3 H in the coastal region of Fukushima.
We also present results on the 3 H contents found in the fish filet collected close to the FNPP1 site. These results are also discussed using the already published 3 H activity concentrations of river and open-ocean waters.

Sampling
In October 2014, the SoSo 5 rivers cruise took place off the coast of the Fukushima Prefecture. The details of sampling radiocaesium have already been provided elsewhere (Aoyama et al., 2020 submitted).
During this cruise, 11 samples were collected for tritium measurement, 4 at the estuaries of the four rivers 130 within 1 km from the river mouth, 4 at the stations within 1 km from the coast, and 3 at the stations 12-15 km from the coast 2020a).
Since June 2014, the radiocaesium activity concentrations at the Tomioka port, which is located 10 km south of FNPP1, are being measured on a regular basis. However, limited 3 H data were collected here. At Hasaki, which is located 180 km south of FNPP1, the radiocaesium activity concentrations were 135 measured on a regular basis, monthly to bi-monthly. The details of the sampling of radiocaesium are already provided elsewhere (Aoyama et al., 2012). Again, limited 3 H data were collected here. In October 2014, 11 samples of 3 H were collected, 4 at the estuaries of the four rivers within 1 km from the river mouth, 4 at the stations within 1 km from the coast, and 3 at the stations 12-15 km from the coast 2020a). 140 In addition, two samples were collected at the Tomioka river in 2015 and 2018, and one sample was collected at the Tone river in 2015. All the sampling locations are shown in Figure 1 and Aoyama et al. (2021aAoyama et al. ( , 2021bAoyama et al. ( and 2021c.

Method
One litre of seawater collected in October 2014 during the SoSo 5 rivers cruise underwent 3 H analysis 145 through electrolytic enrichment at the 3 H Laboratory (Miami, USA). The 3 H activity concentration measurement of the seawater samples collected at the Tomioka port and Hasaki were also conducted through electrolytic enrichment at Chikyu Kagaku Kenkyusho Inc. (Nagoya, Japan), and the 3 H activity concentrations of the fish filets were measured at the Kyushu Environmental Evaluation Association (Fukuoka, Japan). We also measured the 3 H activity concentration in the river waters of Tomioka and 150 Tone rivers through electrolytic enrichment, too.
Fish filets collected at the Fukushima coast in 2014.were analysed with respect of Tissue-free water tritium, TFWT, which was extracted from the fish filet samples using freeze-drying techniques, where TFWT was trapped at low temperatures and then electrolytic condensation was performed to increase the tritium activity concentration. Total organically bound tritium, TOBT, of the freeze-dried fish filet 155 samples was obtained by complete oxidation of the organic compounds at high temperatures in an oxygen-rich environment, and then, the obtained tritiated water was trapped at low temperatures and then distilled. These distilled tritiated water samples were measured using a liquid scintillation counter.

3 H activity concentration in seawater
In 2014-2015, the 3 H activity concentrations at coastal stations of Mano-1, Niida-1, Odaka-1, Uedo-1, and Tomioka port ranged from 90 Bq m -3 to 175 Bq m -3 , while the 3 H activity concentrations at Niida-5, Odaka-5, and Uedo-5 stations, which are located 11-15 km from the coastal stations, decreased and source, while the 3 H might come from a broader range of locations since related to riverine inputs from rivers located both north and south of the FNPP1 site.

3 H activity concentration in fish filet
3 H activity concentrations of tissue free water, i.e. tissue-free water tritium (TFWT), in fish filet of 195

Sebastes cheni, Hexagrammos otakii, Okamejei kenojei, Lateolabrax japonicas, Paralichthys olivaceus,
and Gadus microcephalus vary from 97 ±11 Bq m −3 to 144 ± 11 Bq m −3 (Aoyama et al., 2020b), as shown in Figure 3. These fishes were collected close to the FNPP1 site and the 3 H activity concentrations were similar to the 3 H activity concentrations in the surrounding seawater, as shown in Fig. 3. This is consistent with previous knowledge of rapid equilibrium between TFWT and the environment (Eyrolle et al.,200 2018; Eyrolle-Boyer et al., 2014). However, total organically bounded tritium (TOBT) exceeded TFWT with a ratio of TOBT to TFWT ranging from 1.54 to 2.10 in this study (Aoyama et al., 2020b), which indicates that these organisms fed on food that was contaminated earlier (Eyrolle-Boyer et al., 2014).

3 H activity concentrations in river water 205
The 3 H activity concentration at 56 North canal, 56N, of Fukushima Dai-ichi Nuclear Power, hereafter FNPP1, varied significantly and showed a decreasing trend in general during the period from 2013 to 2019 while the 3 H activity concentrations at Ukedo port and FNPP2 port were not significantly different during the period from 2013 to 2019 (Fig. 4).
As shown in Fig. 4, the 3 H activity concentrations at Ukedo port and FNPP2 port were also lower in 210 general rather than those at Ukedo river and Tomioka river before March 2015; however, recently, they became quite similar. This suggests a decreasing trend of the 3 H activity concentrations at Ukedo and Tomioka rivers and a tendency of Ukedo and FNPP2 ports to retain almost constant 3 H activity concentrations, as shown in Fig. 4. In 2014, we also observed that the 3 H activity concentrations at estuary and ports, except close to the FNPP1, are characterised by levels close to river water. Indeed, the 215 Takahata et al. ) also reported a 3 H/ 137 Cs activity ratio of 0.012 ± 0.007 for the samples studied by them, which is similar to the 3 H/ 137 Cs activity ratio in water that is released from the 235 damaged FNPP1 reactors (~0.01; (Nishihara et al., 2015)). Direct discharge of 3 H into the ocean from FNPP1 was estimated to be approximately 0.05 PBq by multiplying the total 137 Cs emission and the 3 H/ 137 Cs ratio reported by Aoyama et al. (Aoyama et al., 2016b). This estimate ( 0.05 ± 0.03 PBq) is slightly lower than that obtained by Povinec et al. (2013) using the same method ( 0.3 ± 0.2 PBq) for the samples collected in June 2011 from the south of the 240 FNPP1 site during the KOK cruises (Buesseler et al., 2012). In this study, we standardised major axis regression that considered both x-axis and y-axis errors to obtain a reliable 3 H/ 137 Cs activity ratio from the mixing between open-ocean water and source waters from river and the FNPP1 site. As shown in Table 1

3 H activity concentrations in river water and precipitation
Matsumoto et al. (Matsumoto et al., 2013)  notice that 3 H content in precipitation during the typhoon period was lower than that in precipitation during the non-typhoon period in the same month (Yamada et al., 2015). This observation might suggest that the source of precipitation during typhoon should be of oceanic origin characterised by 3 H levels around 70 Bq m −3 as indicated previously.

The Fukushima Prefecture government also monitors the 3 H activity concentration in many rivers in 275
Fukushima. In 2013-2014, the 3 H activity concentrations at Ukedo river and Tomioka river ranged from < 650 Bq m −3 to 1100 Bq m −3 (Aoyama et al., 2020c), which were slightly higher than that in precipitations (Tagomori et al., 2015) therefore, underground water of which the 3 H activity concentration

3 H activity concentrations in open-ocean water and coastal waters off Fukushima
The background 3 H level for the western North Pacific Ocean has been estimated from the 295 GLOMARD/MARIS database to reach up to 50 ± 12 Bq m −3 (decay corrected to June 2011) in an area under the influence of the Kuroshio current (Povinec et al., 2017). Moreover, these levels could be 24 Bq m −3 in an area under the influence of the Oyashio current (Watanabe et al., 1991). The 3 H activity concentrations in surface water in subtropical gyre obtained by the WOCE/GOSHIP hydrographic program mainly along P2 and P3 lines in 1986, 1993, 1994, 2004, and 2013 are in a database WOD2018 300    Table 4.
In contrast, the 3 H activity concentration at coastal stations of FNPP1 obtained through monitoring by the Japanese government and Fukushima Prefecture ( Fukushima Prefecture 1979, andits update until 2018) was approximately 1000 Bq m −3 in 1986, as shown in Fig. 8 Table 2, and the results of flux calculation are listed in Table 3. Regarding the 3 H fluxes, the largest source is from the open-water inflow from the north of FNPP1, and it 330 reaches 52 GBq day -1 while the rivers north of FNPP1 show 3-6 GBq day -1 fluxes. From the port of FNPP1, we use Kanda's method (Kanda 2012), which considers the 3 H activity concentration at Monoageba in the port and λ = 0.44 day -1 for the exchange rate between the port and open water.
Considering the volume of water in the port and λ, 9.6 m 3 s -1 of water flux is estimated at the mouth of the port, that led to the 3 H fluxes in the range of 1.9-4.5 GBq day -1 in three cases in 2014 and 2019, which is 335 comparable with the 3 H fluxes from the rivers located north of FNPP1. In contrast, considering Tsumune's method to estimate the flux from the FNPP1 site to the open-water by using the activity concertation at 56N of FNPP1 (Tsumune et al., 2012), the 3 H flux from the FNPP1 site was found to be 28 to 86 GBq day -1 as shown in Table 3, which is one order of magnitude larger than those estimated of 1.9-4.5 GBq day -1 from the port of FNPP1. One of the reasons could be the very high variability in the 3 H results at 340 56N of FNPP1, which indicated a variable 3 H/ 137 Cs activity ratio at 56N and the port of FNPP1.

Conclusion
In 2014-2015, the 3 H activity concentrations at coastal stations Mano 1, Niida1, Odaka 1, Uedo 1, and Tomioka ranged from 90 Bq m-3 to 175 Bq m-3, and decreased between 67 Bq m-3 to 83 Bq m-3 at the stations located 12-16 km from the coast. The 3 H activity concentration at 56N of the FNPP1 site was 345 significantly high compared to the surrounding waters both north and south of the FNPP1 site. It should also be noted that the 3 H activity concentrations were similar at the stations located both north and south of the FNPP1 site, while the 137Cs activity concentrations were lower at the stations north of the FNPP1 site and higher at the stations south of the FNPP1 site. This indicated that major sources of 137Cs could be the FNPP1 site as the point source while the source of 3 H should be more diffuse and linked to riverine inputs 350 located north and south of the FNPP1 site. The 3 H activity concentration of TFWT in the fish filets of Hexagrammos otakii, Sebastes cheni, Okamejei kenojei, Lateolabrax japonicas, Paralichthys olivaceus, and Gadus microcephalus collected close to the FNPP1 site ranged from 97 ±11 Bq m-3 to 144 ± 11 Bq m-3, which were similar to the 3 H activity concentrations in the surrounding seawater, in agreement with the knowledge that the bioconcentration factor of 3 H is approximately 1. In contrast, higher values were found 355 in TOBT, which can be linked to life-history traits. The 3 H/137Cs activity ratios derived from the land side were 1.2-2.2, which is significantly high compared to that of the released radionuclides derived from the  Blue solid square: TFWT in fish filet (this study).
Red solid square: 3 H activity concentration in seawater samples at 56N of FNPP1 (large) and close to the FNPP1 site (small) obtained by TEPCO and NRC monitoring, respectively. Several data of the 3 H activity concentration at 56N of FNPP1 were 25 below the detection limit (ca. 1600 Bq m −3 ) and did not appear in this figure.

Figure 5
Relationship between 3 H and 137 Cs activity concentrations in the samples collected from Tomioka port.