Introduction

A major earthquake followed by a tsunami occurred in Japan on 11 March 2011. This event led to the greatest nuclear accident since the Chernobyl disaster in the Fukushima Daiichi nuclear power plant, which negatively influenced the nearby marine environment. A huge amount of radioactive pollutants was released into the northwest Pacific. The accident at the Chernobyl nuclear power plant (NPP) in 1986 demonstrated transparently that radioactive pollutants from a nuclear accident can spread from region to region and across the world, resulting in major damage and long-term environmental effects1,2,3 Although the ocean has a great capacity to dilute and disperse radioactive materials due to its large volume and complex current systems, the long half-lives of radionuclides allow them to remain in marine environments for a long period of time, potentially affecting the environment or human beings via biological concentration and transfer through the marine food chain. Therefore, understanding the fate, transport and ecological consequences of the long-lived radionuclides in the oceanic environment is of paramount importance. To understand the transport of radioactive pollutants released from the FNA in the northwest Pacific and their potential impact, eight monitoring cruises were conducted by the State Oceanic Administration of China from 2011–2014 (one cruise every half year).

134Cs, 137Cs and 90Sr are the most important artificial radionuclides released from the FNA. Before the FNA, 137Cs and 90Sr could be detected in the ocean, whereas 134Cs could not be detected. Therefore, if 134Cs is detected and 137Cs and 90Sr exceed their range of background, there is strong evidence of pollutants released from the FNA. This paper reports the monitoring results of 134Cs, 137Cs and 90Sr of the third cruise, which was performed in May–June 2012. It was very expensive and difficult to perform the monitoring work. There were seldom large-scale-monitoring of the radioactive status in the northwest Pacific more than one year after the FNA. Thus, all the data in this paper are important and provide an opportunity to evaluate the radioactive impact of the nuclear accident.

Methods

Sampling stations

Water samples were collected during May–June 2012. The sampling stations are shown in Fig. 1. Black dots signify vertical sampling stations and blue dots signify surface stations.

Figure 1
figure 1

Map of sampling stations.

This figure was drawn using Surfer.

Full size image

Sampling method

At vertical stations, samples were collected at depths of 1 m, 100 m and 200 m. For the samples collected at depths of 1 m and 100 m, 125 l of seawater was collected by submersible pumps. For the samples collected at a depth of 200 m, 60 l of seawater was collected by a CTD-rosette assembly with Niskin bottle samplers (Model Sea-Bird 911 plus, Sea bird electronics, Inc., Bellevue, Washington, USA.). At the surface stations, only surface seawater was sampled. All of the seawater was stored in polyethylene barrels with acidification.

Radionuclide measurements

134Cs and 137Cs were measured using an AMP (ammonium phosphomolybdate) coprecipitation-γ spectrometry method according to the Technical Specification for Marine Radioactivity Monitoring issued by the Division of Marine Environmental Protection, State Oceanic Administration of China. (No. 10 [2011] Haihuanzi). Briefly, 30 mg of CsCl and 18 g of AMP were added to 60 l of seawater and the mixture was stirred and adjusted to obtain a pH ≤ 3. The mixture was stirred for 30 min until it was well mixed. After standing for more than 24 h, the supernatants were removed by siphoning and the residual was filtered. After elution 3 times using deionized water, the filter membrane was placed into the muffle and incandesced at 450°C for 2 h. The residue was measured using a HpGe γ spectrometer to determine the activities of 134Cs and 137Cs.

90Sr was measured by a di (2-ethylhexyl) phosphoric acid (HDEHP) extraction-β counting method according to the Technical Specification for Marine Radioactivity Monitoring issued by the Division of Marine Environmental Protection, State Oceanic Administration of China. (No. 10 [2011] Haihuanzi). A total of 2.00 ml of 100 mg/ml Sr(NO3)2, 1.00 ml of 20 mg/ml Y(NO3)2, 60 g of NH4Cl and 400 g of Na2CO3 was added to 40 l of seawater while stirring. The mixture was stirred for 30 min until well mixed. After standing for more than 24 h, the supernatants were removed by siphoning and the residual was filtered. A total of 300–400 ml of 10 mol/l HNO3 was used to dissolve the sediment. The solution was extracted twice using 50 ml of 10% di (2-ethylhexyl) phosphoric acid (HDEHP) and the organic phase was re-extracted twice using 20 ml of 10 mol/l HNO3. A total of 5 ml of C2H2O4 was added to form a saturated solution and the solution was adjusted to pH = 1.5–2.0 using a 6 mol/l NH3H2O solution and 2 mol/l HNO3. Finally, the YC2O4 sediment was produced. The YC2O4 was filtered and placed into α/β counter to determine the activity of 90Y. The activity of 90Sr was then calculated from the 90Y data according to a special formula.

For the samples collected at a depth of 200 m, only 60 l of seawater was obtained. The collected seawater was used for the measurement of 134Cs and 137Cs, but there was not enough sample to measure 90Sr.

Results

Concentrations of 134Cs, 137Cs and 90Sr

The contents of 134Cs, 137Cs and 90Sr are shown in Tab. 1. The activities of 134Cs ranged from undetected to 9.68 Bq/m3, with an average of 1.21 Bq/m3. The activities of 137Cs ranged from 0.41 to 18.10 Bq/m3, with an average of 3.39 Bq/m3. The activities of 90Sr ranged from 0.56 to 3.57 Bq/m3, with an average of 1.49 Bq/m3. The background data before the FNA are also listed in Tab. 1. According to the data from IAEA's MARiS database from 2000 to 2010, the radioactivity of 137Cs in surface seawater near Japan ranged from 0.04 to 3.4 Bq/m3, with an average of 1.7 ± 0.6 Bq/m3 (n = 961). 134Cs was usually not detectable in surface seawater because there has been no new major source over the past 20 years and the 134Cs that has been released over the last century has decayed to an extremely low level, which is below the detection limit. The activity of 90Sr in surface seawater in the northwest Pacific was 0.01 to 2.6 Bq/m3, with an average 1.2 ± 0.4 Bq/m3(n = 871). The monitoring data in this report were clearly higher than the monitoring data from before the FNA.

Table 1 Concentrations of 134Cs, 137Cs and 90Sr
Full size table

Distributions of 134Cs and 137Cs

The distributions of 134Cs and 137Cs at depths of 1 m, 100 m and 200 m are shown in Fig. 2 and Fig. 3, respectively. A total of 52% of the surface sample contained 134Cs. However, that amount increased to 70% and 69% at the 100 m layer and 200 m layer, respectively. Generally speaking, 134Cs was found as far south as 21°N (200 m) and as far east as 152°E. 137Cs was found at all sample stations. From the surface to a depth of 200 m, the activities of 137Cs and 134Cs increased with increasing depth. At a depth of 200 m, both 137Cs and 134Cs had the highest values. At the adjacent sea area of Taiwan Island, 134Cs was not found in the surface water and 100 m layer. However, at a depth of 200 m, 134Cs was found (1.99 ± 0.23 Bq/m3) at 21.50°N, 125.00°E, which is 430 km away from the southernmost point of Taiwan Island.

Figure 2
figure 2

The distributions of 134Cs at depths of 1 m, 100 m and 200 m in the North Pacific.

These figures were drawn using Surfer.

Full size image
Figure 3
figure 3

The distributions of 137Cs at depths of 1 m, 100 m and 200 m in the North Pacific.

These figures were drawn using Surfer.

Full size image

The distributions of 90Sr

The distributions of 90Sr at depths of 1 m and 100 m are shown in Fig. 4. 90Sr was found in all samples. Unlike 137Cs and 134Cs, 90Sr activities did not clearly change with depth and instead were relatively uniform. The highest 90Sr activity was only slightly higher than the upper limit of the background (Tab. 1).

Figure 4
figure 4

The distributions of 90Sr at depths of 1 m and 100 m in the North Pacific.

These figures were drawn using Surfer.

Full size image

Discussion

Variation in radioactivity level

The highest activities and the average activities of the three artificial radionuclides from June 2011 to June 2012 are shown in Fig. 5. The first monitoring cruise was conducted in June 2011. The highest activities of 137Cs, 134Cs and 90Sr were approximately 400 ~ 800 km away from Fukushima, reaching up to 800 Bq/m3. However, the activities of the three nuclides decreased with each subsequent cruise. The second monitoring cruise was conducted in December 2011. The highest activities of 137Cs and 134Cs decreased ten-fold to approximately 80 Bq/m3. In the third cruise, the activities decreased even further. Comparing the monitoring data to the previous radioactivity data before the FNA, we can conclude that the northwest Pacific was clearly influenced by the accident even after one year had passed.

Figure 5
figure 5

Variations in the highest activities (left) and the average activities (right) in the North Pacific during cruises 1–3.

This figure was drawn using Origin.

Full size image

Transport of artificial radionuclides

Marine scientists want to know how radioactive pollutants are transported in the North Pacific Ocean and many numerical simulations have been reported4,5,6,7. The radioactive pollutants from the FNA were transported eastward at a speed of approximately 200 km/month8, which was consistent with the trajectories of Argo floats launched near the Fukushima nuclear power plant. The pollutants are thus expected to reach the coastal water of the USA after approximately 4–6 years following the North Pacific Current (Fig. 6)5,7,9. The entire North Pacific Ocean will be labeled with Fukushima 137Cs 10 years after the accident7. Mesoscale eddies are expected to disperse radioactivity in complicated patterns. The radioactivity of 137Cs was rapidly diluted and dispersed in the North Pacific Ocean5. The mesoscale eddies in the Kuroshio Extension played an important role in diluting radioactive cesium4, which is why 134Cs was found as far east as 152°E and had relatively low activities more than one year after the accident. Because there were no other stations in the area east of 152°E, we were unable to determine how far east the radioactive pollutants were transported.

Figure 6
figure 6

A schematic view of the formation and subduction of mode waters in the North Pacific.

The yellow and yellow-shaded ellipses indicate spreading and formation areas, respectively, of STMW (25.0–25.6 δθ). The green and green-shaded areas indicate spreading and formation areas, respectively, of Central Mode Water (CMW) (26.0–26.6 δθ), which is denser than STMW. The thick broken and solid arrows show the spreading directions of STMW and CMW, respectively. The blue and red dotted lines are surface water currents of the subarctic and subtropical gyres, respectively. SAF, KEF and STF indicate the subarctic, Kuroshio Extension and subtropical fronts along the observational line, respectively. This figure was cited from the literature9.

Full size image

The highest value and the highest average of 137Cs and 134Cs were found in the 200 m layer. Kumamoto et al. (2014) determined that radioactive pollutants can penetrate to depths of 200 m through surface cooling and vertical mixing in the winter, which supports our results. We were unable to determine how deep the radioactive pollutants were transported because we did not take samples from water deeper than 200 m. However, according to Kumamoto et al. (2014)8, 134Cs penetrated to a depth of approximately 300 m between 35°N and 40°N along the 149°E meridian Jan–Feb 2012. Therefore, in May–June 2012, the radioactive pollutants were probably transported to the deeper area than 300 m.

At 21.50°N, 125.00°E, the station XTW8-3, 134Cs was found at a depth of 200 m, which is 430 km away from the southernmost point of Taiwan Island. The presence of 134Cs in this location is unusual. It was unusual that the radioactive pollutants were transported from 37°N to 21.5°N so quickly just over one year after the accident. It was also unusual that 134Cs was not detected at the surface and 100 m depth but only at a depth of 200 m. Finally, it remains to be explained why 134Cs was not detected at other stations except for the one at 21.50°N, 125.00°E. We re-checked the data and verified that they are accurate. After excluding the possibility that the data was incorrect, we determined that there was a possible explanation for the unusual findings. This was that the detected 134Cs was derived from the formation and subduction of Subtropical Mode Water (STMW)8. To the south of the Kuroshio Extension (KE) between 30°N and 35°N, STMW is formed and penetrates to a depth of approximately 400 m (25.6 δθ) in late winter. This STMW then spreads to the nearly subtropical front through advection over the Kuroshio recirculation region9,10 (Fig. 6). 134Cs in the North Pacific Ocean occurred mainly in March 2011, when STMW was just being formed. Therefore, the 134Cs that arrived to the south of the KE was most likely immediately mixed vertically to depths of 300–400 m. The 134Cs in newly formed STMW then started to spread to approximately 20°N along subsurface isopycnals (25.0–25.6 δθ)8. According to relevant STMW references11,12,13,14,15,16,17, the essential features of STMW are that the potential vorticity no greater than 2.0 × 10−10/m·s, the depth is 150 ~ 350 m, the salinity is 34.6 ~ 34.8, the temperature is 16 ~ 18°C and the potential density is 25.0–25.6 δθ. The potential vorticity was calculated using the following equation: PV = (f/ρ) (Δδθ/ΔZ)10, where f is the Coriolis parameter, ρ is the density of seawater, Δδθ is the difference between adjacent potential vorticity layers and ΔZ is the interval between adjacent potential vorticity layers. For a depth of 200 m at station XTW8-3, the potential vorticity was 2.0 × 10−10/m·s, the potential density was 25.13 δθ, the salinity was 34.72 and the temperature was 17.73°C (Tab. 2). These values agreed well with the essential features of STMW. The time required for 134Cs emitted from the Fukushima nuclear power plant to enter the formation area of STMW is approximately 1 ~ 2 months8. From there, approximately one year is needed to transport the radioactive material via STMW from the area of formation (35°N) to low-latitude areas near 23°N10. The speed required is 1°/month (12° over 12 months). Based on this speed, we calculated that 134Cs released from the Fukushima nuclear power plant and transported to 21.5°N requires approximately 14.5 ~ 15.5 months. In this report, 134Cs that was released during the FNA was transported to 21.5°N 15 months after the accident. The predicted time scale is consistent with the actual situation. Based on an analysis of our own data regarding potential vorticity, depth, salinity, temperature, potential density and time scale, we have demonstrated that only the seawater at a depth of 200 m at 21.50°N and 125.00°E was most likely from the STMW at station XTW8-3. Contour maps of temperature, salinity and potential density for section XTW8 at depths of 0–500 m are shown in Fig. 7. The essential features of STMW in section XTW8 are marked by the different colors. By overlapping the three maps, it is clear that the STMW exists between depths of 200 m and 300 m; however, observations were only collected at a depth of 200 m in this area (Fig. 7). As shown in Tab. 2, the calculated potential densities at a depth of 200 m for the other stations in section XTW8 are out of the range of typical STMW values (i.e., 25.0–25.6 δθ). Therefore, only the seawater sample from a depth of 200 m at station XTW8-3 was from STMW. This is a possible reason for why 134Cs was not detected at the surface and a depth of 100 m at station XTW8-3.

Table 2 The temperature, salinity, potential density, potential vorticity of section XTW8
Full size table
Figure 7
figure 7

Maps of temperature, salinity and potential density of the section XTW8.

These figures were drawn using Surfer.

Full size image

Radioactive levels in the coastal water of China

There were 9 stations in the northeast of the South China Sea. A comparison of the monitoring data of these stations with the background data reveals that 137Cs and 90Sr were within the range of the background data and 134Cs could not be detected, which suggests that the coastal water of China has not been impacted by the radioactive pollutants released from the FNA.

Radiological risk assessment

In recent decades, the importance of radiation assessment for non-human species has been recognized18. There are various radiation assessment tools, of which the ERICA tool is the best. ERICA (Environmental Risk from Ionising Contaminants: Assessment and Management) provides an integrated approach to the scientific, managerial and societal issues surrounding the environmental effects of contaminants emitting ionizing radiation, with an emphasis on biota and ecosystems19. The ERICA Assessment Tool (June 2011 version) was used to evaluate radiological risk in the marine environment of the monitoring area. The Tier 2 assessment was completed. Isotopes of 137Cs, 134Cs, 90Sr and a dose screening value of 10 μGy/h were selected in the assessment. The highest activity concentrations of 137Cs, 134Cs and 90Sr in the seawater monitoring results were input into the assessment. The environmental concentration, parameters of distribution coefficient (Kd), concentration ratio (CR), dose conversion coefficients of radiation, occupancy factors and an uncertainty factor were all set as the default value in the ERICA Tool. The weighting factors of low beta, beta/gamma and alpha were set to 3, 1 and 20, respectively. The output results of the Tier 2 assessment are listed in Tab. 3, where all of the total dose rates per organism were lower than the screening rate by approximately 3–6 orders of magnitude. These findings suggest that the radiation increments that come from the 137Cs, 134Cs and 90Sr released from the FNA are only one-thousandth to one-millionth of the screening rate (10 μGy/h).

Table 3 Total dose rate results in the ERICA tool Tier 2 assessment
Full size table

Conclusions

The activities of the artificial radionuclides 137Cs, 134Cs and 90Sr in the seawater of the northwest Pacific decreased with time. In May–June 2012, on the third monitoring cruise of the west Pacific, 134Cs was found in 63.7% of the samples. A total of 32.9% of the 137Cs data were higher than the highest value measured before the FNA (3.4 Bq/m3). The west Pacific was clearly affected by the accident even though more than one year had passed.

134Cs was found as far south as 21°N (200 m) and as far east as 152°E. From the surface water to the deep water, the activities of 137Cs and 134Cs increased with increasing depth. At a depth of 200 m, both 137Cs and 134Cs had the highest values. At 21.50°N, 125.00°E, 134Cs was found at a depth of 200 m, which is 430 km away from the southernmost point of Taiwan Island. The formation and subduction of Subtropical Mode Water (STMW) seems a reasonable explanation for this unusual situation. The formation and subduction of Subtropical Mode Water (STMW) seems a reasonable explanation for this unusual situation. The STMW is formed and penetrates to a depth of approximately 400 m (25.6 δθ) south of the KE between 30°N and 35°N in late winter before spreading to the nearly subtropical front through advection over the Kuroshio recirculation region. The radioactive pollutants are brought to the deep water (300–400 m) and transported to approximately 20°N along subsurface isopycnals (25.0–25.6 δθ). Based on an analysis of our own data regarding potential vorticity, depth, salinity, temperature, potential density and time scale, we have demonstrated that only the seawater at a depth of 200 m at 21.50°N and 125.00°E is most likely from the STMW, which successfully explains why 134Cs was not detected at the surface and a doeth of 100 m depth at station XTW8-3.

A comparison of the monitoring data with the background data reveals that the coastal water of China has not been impacted by the radioactive pollutants released from the FNA.

The radiation increments from the 137Cs, 134Cs and 90Sr released from the FNA are only one-thousandth to one-millionth of the screening rate (10 μGy/h).

All the data reported could be used as actual measured data to correct the parameters of interest for numerically modeling the radionuclide transport in the northwest Pacific.