Environ. Sci. Technol. 2004, 38, 3498-3504
Sediment Core Record of Global Fallout and Bikini Close-in Fallout Pu in Sagami Bay, Western Northwest Pacific Margin JIAN ZHENG* AND MASATOSHI YAMADA Nakaminato Laboratory for Marine Radioecology, National Institute of Radiological Sciences, 3609 Isozaki-cho, Hitachinaka, Ibaraki, 311-1202 Japan
The total 239+240Pu activity and 240Pu/239Pu atom ratio in the sediments in Sagami Bay of the western Northwest Pacific margin were investigated using ICP-MS with a shield torch system. 239+240Pu inventories in the examined sediment cores were found to be much higher than those predicted from atmospheric global fallout (42 MBq/km2) at the same latitude. In addition, elevated 240Pu/239Pu atom ratios ranging from 0.22 to 0.28 were observed in the sediment samples. On the basis of the vertical profiles of 239+240Pu and characterized 240Pu/239Pu atom ratios in a sediment core collected in the center of Sagami Bay, we identified two distinct sources of fallout Pu in the bay: the global stratospheric fallout with characteristic 240Pu/239Pu ratio of 0.18 and the transported close-in fallout derived from Bikini and Enewetak surface nuclear weapon test series in the 1950s. We propose that the Pu transportation was mainly due to oceanic processes (for example, through the North Equatorial Current and the Kuroshio Current). Using a two fallout end-member model, we find that the contribution of Bikini close-in fallout Pu ranged from 44 to 59% in Sagami Bay sediments. To the best of our knowledge, this is the first report that Pu contamination, which originated from Bikini and Enewetak nuclear weapon test series in the 1950s, has extended westwards as far as the Japanese coast.
Introduction It was almost five decades ago that anthropogenic radionuclides, such as 239Pu (T1/2 ) 2.44 × 104 yr), 240Pu (T1/2 ) 6.58 × 103 yr), and 137Cs (T1/2 ) 30.0 yr) began to enter the environment from the fallout of atmospheric nuclear weapons tests. Since about 70% of the earth’s surface is covered by oceans, over two-thirds of the fallout has entered them. It was estimated that, since 1945, 10.87 PBq of 239,240Pu has been deposited on the oceans as global fallout from 543 atmospheric nuclear tests (1). The fallout from most nuclear tests has been distributed globally, but in the largest ocean, the Pacific, which is the major repository of plutonium released from atmospheric tests of nuclear weapons, the fallout from the U.S. ground-level test series conducted in the 1950s at Bikini and Enewetak Atolls comprised a substantial tropospheric close-in fallout as well (2). The distribution of Pu in the Pacific Ocean has been extensively investigated because Pu may potentially present * Corresponding author e-mail: telephone: +81-29-265-7130; fax: +81-29-265-9883;
[email protected]. 3498
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 13, 2004
a threat against human health due to its high chemical toxicity and long half-live, and it also serves as a powerful geochemical tracer to study fundamental issues in marine sciences, such as the estimation of sediment mixing rates (3) and elucidation of the scavenging and removal process in the water column (4-9). Livingston et al. (10) recently reviewed the behavior of Pu in the Pacific Ocean. Distribution changes that evolved with time have been observed. For instance, the 239,240Pu subsurface maximum in the central northwest Pacific Ocean has moved from about 500 m in 1973 down to 800 m at present, and the 239,240Pu inventory in water has decreased. In contrast, changes in the subsurface maximum in the Western Equatorial Pacific Ocean are much less, with little change in inventory, although it has been estimated that ca. 0.2 TBq/yr Pu is released from Bikini and Enewetak Atolls (11). It was pointed out that much of the Pu distribution changes resulted from physical circulation of the regional water masses. In addition, in contrast to the conservative radionuclide 137Cs, which has a very small affinity for particle association and its behavior consequently is controlled by the physics of ocean circulation and mixing, Pu is a particlereactive radionuclide with a strength of particle association; thus, its ultimate behavior is linked to these affinities, and the particle scavenging process can exert substantial controls on the distribution of Pu in the water column and sediments. This behavior difference is evident from the sediment distribution coefficient (Kd) for Pu and 137Cs, defined as the ratio of radionuclide concentration in sediment and water (12). To understand the fate of Pu in the ocean, Pu in Pacific sediments should be studied because Pacific sediments are believed to be the ultimate sink for Pu present there. Several studies on the Pu in Pacific sediments have been reported, especially shelf and slope studies on both eastern and western Pacific margins. On the eastern side, a comprehensive review of the work on the California basin sediments was presented by Sholkovitz (13). Beasley et al. (14) reported on transuranic measurements in shelf sediments off Washington and Oregon. On the western side, our laboratory has made a series of marginal studies. For example, Nakamura and Nagaya (15-18) reported on Pu in sediments in the East China and Yellow Seas and in the Seto Inland Sea. Yamada and Nagaya (6, 19) investigated 239,240Pu and 137Cs in sediments from Sagami Bay and Tokyo Bay. All these studies reached the same major conclusion that the Pacific marginal sediments underlying waters with high biological productivity and large suspended sediment loads are major sinks for Pu removal and sequestration. In addition, general excess inventories of Pu over those anticipated from global fallout were observed. The above-mentioned studies drew conclusions mainly based on the total activity of 239,240Pu and the activity of 137Cs; the source of the additional Pu input in these marginal sediments, however, remained unknown because of the lack of a distinctive 240Pu/239Pu signature. Information on Pu isotopic compositions in sediments is very useful in understanding the source of Pu present there since the 240Pu/239Pu ratio has proved to be a powerful fingerprint to identify the Pu contamination (5, 20, 21). The atomic ratio of 240Pu/230Pu in fallout may vary, depending upon the specific weapon design and test yield. The global fallout average 240Pu/239Pu atom ratio is 0.18, based upon atmospheric aerosol sampling, soil samples, and ice core data (5). It is also recognized that different test series can be characterized by either lower or higher ratios. For example, fallout from the Nevada test site has a lower 240Pu/239Pu ratio, averaging 0.035 (20), while elevated 240Pu/239Pu ratios 10.1021/es035193f CCC: $27.50
2004 American Chemical Society Published on Web 05/21/2004
FIGURE 1. Sampling map for Sagami Bay sediment samples. (0.21-0.36) in fallout from the 1950s from the Pacific Proving Grounds were reported (5). Recently, we have developed a highly sensitive method for the determination of 240Pu and 239 Pu in marine sediment samples by means of ICP-MS (inductively coupled plasma-mass spectrometry) with a shield torch system (22). In the present study, we have applied this method in order to determine 240Pu and 239Pu and their ratios in sediment core samples collected in Sagami Bay. On the basis of the 240Pu/239Pu ratio signature, we identified the additional Pu input in Sagami Bay as being from the transported close-in fallout from Bikini and Enewetak nuclear weapon tests. We demonstrated that Pu activity and isotopic ratio profiles of marine sediment cores contain records of global fallout and close-in fallout Pu, which provide chronological information on the recent sedimentation. We proposed that the transport route is mainly through the oceanic currents (e.g., the North Equatorial Current and the Kuroshio Current). Using a two fallout end-member model, we resolved the relative contribution of Pu between global fallout and close-in fallout in Sagami Bay sediments.
Materials and Methods Chemical and Reagents. All commercial chemicals were of analytical reagent grade and were used without further purification. Nitric acid, HCl, HClO4, HF, NH4I, H2O2, and NaNO2 were obtained from Kanto Chemicals (Tokyo, Japan). Plutonium-242 (CRM 130, Plutonium Spike Assay and Isotopic Standard, New Brunswick Laboratory, USA) was used to spike the samples. As an internal standard, Bi (10 ng/mL) was used to correct the sensitivity shift during the measure-
ment. The anion-exchange resin used in this study was AG 1-X8 (100-200 mesh, Bio-Rad). The certified reference material employed for the validation of the analytical procedure was IAEA-368 (ocean sediment from Mururoa Atoll). Sample Collection and Sample Preparation. The marine sediment samples were collected in Sagami Bay during two cruises (KT-90-04 and KT-91-03) of the R/V Tansei Maru, operated by the Ocean Research Institute, University of Tokyo. The sampling sites are shown in Figure 1, and more precise locations are listed in Table 1. Sediment samples were taken using a box corer, and the subcores were carefully selected from the box core. On shipboard, the subcores were cut into 1-, 2-, or 3-cm segments mostly taking sections of 1 or 2 cm from the top 10 cm. The samples were analyzed for Pu in our laboratory. The extraction of Pu from sediment samples was performed based on the procedure described by Muramatsu et al. (23). In brief, the samples were dried at 110 °C for at least 4 h. After the water content was measured, an aliquot of approximately 10 g was weighed out and spiked with 242Pu (100 pg) as yield monitor. The dried sample was placed in a Pyrex beaker, and 8 M HNO3 (about five times the sample weight) was added. The extraction was performed on a hot plate (180-200 °C) for at least 3 h. The warm supernatant (extract) was filtered through a glass-fiber filter. The residue in the beaker was extracted again with 8 M HNO3 for about 30 min and then filtered. The combined filtrates were heated on a hot plate until a thick wet paste was obtained. The wet paste was dissolved by adding concentrated HNO3 while VOL. 38, NO. 13, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3499
TABLE 1. Sediment Samples Used in This Study
station
area
sampling date
KT-90-04-2
Tokyo Canyon
Mar 4, 1990
KT-91-03-8
Sagami Nada
Mar 3, 1991
KT-91-03-P
Sagami Nada
Mar 3, 1991
KT-91-03-6
Sagami Trough
Feb 28, 1991
KT-91-04-3
Sagami Trough
Mar 4, 1990
warming on the hot plate. The sample solution was adjusted to the acidity of 8 M HNO3 by adding deionized water. Before loading the sample solution on the anion-exchange column, NaNO2 (0.2 M) was added to the filtered extract to convert Pu(VI) to Pu(IV), which is the only retainable form in the chromatography column. The separation of Pu, U, and matrix was performed with AG 1-X8 anion-exchange column (11 × 1.0 cm i.d.). The sample solution was loaded onto the column at a flow rate of < 2 mL/min. The column was washed with 8 M HNO3 (40 mL) and with 10 M HCl (40 mL). Pu was then eluted with a mixture of 6 M HCl-0.26 M HF (50 mL). The obtained Pu fraction was further purified using a smaller AG 1-X8 column (10 × 0.7 cm i.d.). For Pu fraction, 4 mL of concentrated HNO3 and 0.2 mL of HClO4 were added, and the solution was heated on the hot plate to remove HF in the eluate. After the solution had been taken to dryness, the residue was dissolved in 8 M HNO3 (20 mL) and loaded on the second AG 1-X8 column. After washing with 8 M HNO3 (40 mL) and 10 M HCl (40 mL), NH4I (5%)-HCl (29-71) solution (50 mL) was used to elute Pu. HNO3 (4 mL), HClO4 (0.2 mL), and H2O2 (1 mL) were added to the eluate, which was heated on a hot plate to decompose any organic material and to expel the excess iodine. After the solution had been taken to dryness, the residue was dissolved in 4% HNO3 (5 mL) for ICP-MS measurement. Based upon comparisons of 242Pu signals for recovered solutions versus ion counts for the same 242Pu activity added to 5 mL of 4% HNO3 solution, the chemical yield in the employed sample preparation procedure was estimated to be in the range of 59-78%. ICP-MS Measurements. The ICP-MS instrument used in this study was a HP-4500 quadrupole ICP-MS (Yokogawa Analytical Systems, Tokyo, Japan). The sample introduction system included a Scott-type spray chamber fitted with a Babington-type nebulizer. To obtain an extremely high sensitivity for actinides (such as Pu, Am, and U), the measurement was performed using a shield torch system operated at normal plasma conditions. Details about the optimization for optimal operation conditions were described elsewhere (22). Concentrations of 239Pu and 240Pu were calculated (isotopic dilution method) from the results of isotopic ratios to the spike (242Pu). The abundance sensitivity based on 238U measurement (masses 237 and 238) was about 1 × 10-6. Using a 10-g sediment sample, detection limits of 0.0095 mBq/g for 239Pu and 0.0355 mBq/g for 240Pu were achieved. These detection limits are sufficiently low for the determination of Pu in marine sediment samples. With the ion chromatographic separation and purification, U was separated from Pu, which is evident from the high decontamination factor of U (1.2 × 104) that we obtained. Thus, the interference of UH+ on the determination of 239Pu is negligible in our analytical system. In addition, the analytical method was validated by the analysis of IAEA 368 (ocean sediment) reference material with satisfactory results for the 3500
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 13, 2004
location 35-02.1° N 139-40.1° E 35-04.8° N 139-16.2° E 35-59.1° N 139-18.4° E 35-55.6° N 139-24.8° E 35-37.9° N 140-06.3° E
water depth (m)
core length (cm)
808
14
1329
28
1424
22
1474
22
2519
20
accuracy of 239+240Pu activity and 240Pu/239Pu ratio (22). During the ICP-MS measurements, three determinations of the Pu isotopes were made for each sample solution. The relative standard deviation (RSD) for the determination of 240 Pu/239Pu ratio ranged from 0.4% to 3% in most cases for the sediment samples.
Results and Discussion Pu Activities and 240Pu/239Pu Ratio. Sagami Bay is encircled on three sides by land and is a deep-water body having its deepest depth below 1500 m; this part is known as the Sagami Trough. In the present study, five sediment cores collected from Sagami Nada (KT-91-03-08 and KT-91-03-P), Tokyo Canyon (KT-90-04-2), and Sagami Trough (KT-91-03-6 and KT-91-04-3) were analyzed for their 239+240Pu activities and 240Pu/239Pu ratios by ICP-MS. The results are summarized in Table 2. A comparison between the total 239+240Pu activities obtained by ICP-MS and by R-spectrometry (19) showed a good agreement, indicating that the data sets obtained by ICP-MS were highly reliable. It was reported previously that the surface 239+240Pu activities in the Sagami Nada cores were noticeably higher than those in the other cores from Tokyo Canyon and Sagami Trough (19). This finding was confirmed in the present work. The activities of 137Cs were also measured for these samples (19) and were corrected for decay to the date of collection; thus, we could calculate the activity ratios of 239+240Pu/137Cs, which are listed in Table 2. The mean value of 0.75 ( 0.23 (the error represents the 95% confidence limit) obtained in the present work based on five sediment cores was consistent with the result (0.65 ( 0.17) in the previous work (19) based on 10 sediment cores. Also, as shown in Table 2, compared with the global fallout Pu (240Pu/239Pu ratio, 0.18), the 240Pu/239Pu ratios detected in the sediment cores generally showed elevated values, ranging from 0.22 to 0.28, with a trend toward higher ratios in the deeper sediment layers. The mean inventory of 239+240Pu (191 ( 120 MBq/km2) in the sediments from Sagami Bay was reported previously (19). This mean inventory is much higher than the expected atmospheric global fallout value at 30-40° N (42 MBq/km2) (24). In addition, the measured 239+240Pu/137Cs ratios with a mean of 0.75 ( 0.23 (the error represents the 95% confidence limit) are significantly higher than the ratio predicted from the atmospheric fallout of 0.021 (24). It was suggested that higher Pu inventory could be due to an additional input of Pu of land-origin: by rivers and by winds. According to an IAEA technical report (12), the sediment distribution coefficient (Kd) in pelagic ocean and the coastal sediment concentration factor (CF) for Pu are each 1 × 105 (mean value), while the Kd and CF for Cs are 2 × 103 and 3 × 103, respectively. It was seen that the Kd and CF values of Pu were almost 2 orders of magnitude larger than that of Cs; therefore, the higher 239+240Pu/137Cs ratios could be attributed to an
TABLE 2. Activities of depth in core (cm)
239+240Pu
239+240Pu
ICP-MS
(mBq/g, dry weight) and Atom Ratios of
239+240Pu r-spectrometryb
atom ratio 240Pu/239Pu
Pu/137Cs activity ratio
240Pu/239Pu
depth in core (cm)
in Sediment Cores from Sagami Baya
239+240Pu
ICP-MS
239+240Pu r-spectrometryb
atom ratio 240Pu/239Pu
Pu/137Cs activity ratio
Tokyo Canyon KT-90-04-2 0-2 2-4 4-6 6-8
2.96 ( 0.09 2.67 ( 0.21 1.65 ( 0.14 0.89 ( 0.04
2.89 ( 0.07 2.34 ( 0.07 1.62 ( 0.05 0.94 ( 0.03
0.244 ( 0.013 0.224 ( 0.004 0.241 ( 0.006 0.238 ( 0.013
1.03 ( 0.10 1.35 ( 0.18 1.33 ( 0.23 1.17 ( 0.25
KT-90-04-2 8-10 0.32 ( 0.01 10-12 12-14
0.55 ( 0.03
0.209 ( 0.004 0.78 ( 0.27
0.13 ( 0.01
Sagami Nada KT-91-03-8 0-1 4.75 ( 0.22 1-2 5.81 ( 0.11 6.11 ( 0.23 2-3 6.37 ( 0.15 5.66 ( 0.31 3-4 6.42 ( 0.06 7.28 ( 0.35 4-5 6.52 ( 0.22 5-6 7.06 ( 0.24 6-7 7.86 ( 0.10 7.94 ( 0.42 7-8 8.09 ( 0.16 7.28 ( 0.42 8-9 8.79 ( 0.18 9.64 ( 0.38 9-10 8.95 ( 0.09 8.22 ( 0.35 10-12 11.85 ( 0.29 8.95 ( 0.39 12-14 12.73 ( 0.56 11.17 ( 0.51 14-16 11.51 ( 0.10 9.71 ( 0.34 16-18 9.11 ( 0.12 6.87 ( 0.30 18-20 6.71 ( 0.12 5.09 ( 0.13 20-22 2.67 ( 0.02 2.11 ( 0.15 22-24 0.58 ( 0.02 24-26 0.26 ( 0.02 26-28 0.24 ( 0.02
0.235 ( 0.007 0.82 ( 0.09 0.234 ( 0.011 0.95 ( 0.11 0.235 ( 0.003 0.83 ( 0.07 0.241 ( 0.005 0.233 ( 0.005 0.236 ( 0.009 0.233 ( 0.007 0.237 ( 0.003 0.236 ( 0.008 0.255 ( 0.006 0.277 ( 0.004 0.272 ( 0.013 0.272 ( 0.004
0.79 ( 0.05 0.85 ( 0.06 0.89 ( 0.06 0.74 ( 0.04 0.80 ( 0.03 0.72 ( 0.04 0.90 ( 0.04 1.13 ( 0.06 1.33 ( 0.09 1.60 ( 0.23
KT-91-03-P 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 10-12 12-14 14-16 16-18 18-20 20-22
5.34 ( 0.23 6.87 ( 0.33 6.30 ( 0.07 6.63 ( 0.04 5.75 ( 0.14 5.66 ( 0.14 5.14 ( 0.05 3.02 ( 0.14 2.15 ( 0.07 1.22 ( 0.11 0.57 ( 0.07
4.87 ( 0.18 4.83 ( 0.14 7.26 ( 0.35 6.65 ( 0.27 6.61 ( 0.31 6.33 ( 0.25 5.81 ( 0.32 5.27 ( 0.16 4.50 ( 0.26 4.07 ( 0.10 3.41 ( 0.13 2.09 ( 0.06 1.37 ( 0.04 0.59 ( 0.02 0.12 ( 0.01 0.035 ( 0.003
0.247 ( 0.006 0.47 ( 0.05 0.244 ( 0.012 0.228 ( 0.006 0.259 ( 0.003 0.264 ( 0.005 0.269 ( 0.007 0.273 ( 0.001
0.60 ( 0.04 0.59 ( 0.05 0.60 ( 0.04 0.55 ( 0.04 0.65 ( 0.06 0.87 ( 0.06
0.249 ( 0.013 0.72 ( 0.09 0.239 ( 0.017 0.56 ( 0.08 0.244 ( 0.007 0.54 ( 0.08 0.52 ( 0.12
Sagami Trough KT-91-03-6 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 10-12 12-14 14-16 16-18 18-20 20-22 a
4.33 ( 0.08 3.59 ( 0.05 3.43 ( 0.18 2.68 ( 0.03 2.58 ( 0.12 1.97 ( 0.06 0.60 ( 0.06 0.13 ( 0.02
3.82 ( 0.12 2.95 ( 0.09 3.23 ( 0.07 2.93 ( 0.10 2.13 ( 0.06 1.71 ( 0.05 0.58 ( 0.03 0.34 ( 0.02 0.063 ( 0.006 0.12 ( 0.01 0.15 ( 0.02 0.023 ( 0.004 0.026 ( 0.003 0.047 ( 0.008 0.070 ( 0.009 0.13 ( 0.01
0.232 ( 0.003 0.240 ( 0.006 0.245 ( 0.005 0.269 ( 0.002 0.281 ( 0.009 0.256 ( 0.014 0.269 ( 0.024
The errors quoted are one standard deviation.
b
0.68 ( 0.04 0.65 ( 0.06 0.72 ( 0.07 0.65 ( 0.05 0.70 ( 0.24
KT-90-04-3 0-2 2-4 4-6 6-8 8-10 10-12 12-14 14-16 16-18 18-20
0.77 ( 0.05
0.91 ( 0.03 0.229 ( 0.015 0.35 ( 0.05 1.24 ( 0.05 1.17 ( 0.09 1.12 ( 0.03 0.238 ( 0.005 0.42 ( 0.05 1.44 ( 0.06 1.23 ( 0.09 0.251 ( 0.022 0.45 ( 0.07 0.58 ( 0.04 0.49 ( 0.02 0.30 ( 0.04 0.17 ( 0.01 0.022 ( 0.004 0.065 ( 0.008 0.012 ( 0.002 0.006 ( 0.002
Data cited from Yamada and Nagaya (19).
enhanced scavenging of 239+240Pu. Muramatsu et al. (25) analyzed soil samples collected from different places in Japan and found that 240Pu/239Pu ratios ranged from 0.168 to 0.181, which were comparable to the global fallout ratio. Hirose et al. (26) evaluated recent Pu fallout in Japan. They found that the 240Pu/239Pu atomic ratios in deposition samples collected in Tsukuba, Nagasaki, and Yonaguni in 2000, ranged from 0.187 to 0.206 (annual average value). Therefore, even if the excess Pu inventory was partly due to the land-origin Pu supply in Sagami Bay, it would still be from the global fallout. In other words, global fallout can be regarded as consisting of two parts: direct global fallout in Sagami Bay and the land-origin global fallout. The elevated 240Pu/239Pu ratios observed in Sagami Bay sediments indicate that there must be an additional Pu input that is isotopically distinctive from the global fallout. Sources of Pu in Sagami Bay Sediments. Based on the UNSCEAR report (1) in 2000, Nakano and Povinec (2) estimated the annual marine deposition of 239+240Pu from global and close-in (resulting from Bikini and Enewetak Atolls nuclear weapon test) fallout (Figure 2). They noted that the maximum marine deposition of 239+240Pu from global fallout was found in 1963 and that the maximum deposition of Bikini close-in fallout 239+240Pu occurred in 1954, due to the extensive surface nuclear weapon tests in the Bikini Atoll. Compared with our data on vertical profiles of 239+240Pu and 137Cs activities
FIGURE 2. Annual marine deposition of 239+240Pu from global and Bikini close-in fallout estimated on the basis of UNSCEAR 90Sr data (adopted from Nakano and Povinec (2)). and 240Pu/239Pu ratios in a sediment core from Sagami Nada (KT-91-03-8) as shown in Figure 3, there was a striking similarity between the annual marine deposition of Pu and the Sagami Bay sediment core record. In the sediment core, the activity of 239+240Pu increased with core depth and reached a peak at the layer of 12-14 cm. Coincidently, a concentration maximum of 137Cs was observed in the same layer (Figure VOL. 38, NO. 13, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3501
time while the Pu level has been continuously decreasing (Figure 3). Several possibilities could be considered for this observation: (i) the stratospheric and tropospheric Pu sources have been continuously deposited in similar proportions but with decreasing fluxes to maintain this atom ratio; (ii) Pu from these two sources was already premixed in some other reservoir (deep or surface ocean); and (iii) this result was a consequence of mixing in sediments by bioturbation during and after deposition. Considering the fact that the observed constant Pu atom ratio in Sagami Bay sediments is similar to that observed in the water column near the Bikini and Enewetak Atolls (2), it is most likely that the observed constant Pu atom ratio in Sagami Bay sediments is due to the premix of the stratospheric and tropospheric Pu in the water column. However, more work is needed to clarify this observation.
FIGURE 3. Vertical profiles of 239+240Pu, 240Pu/239Pu, and 137Cs in a sediment core (KT-91-03-8) collected from Sagami Bay, Japan. The vertical lines shown in panel A are the global fallout ratio of 0.18 and the Bikini close-in fallout ratio of 0.30. The error bars shown in panel A are one standard deviation, and in panel B they are 1σ values derived by counting statistics. 3B). The maximum atmospheric deposition of 137Cs in Japan was observed in 1963 (27). The sediment core, therefore, retained a valid record for the maximum marine deposition of 239+240Pu fallout in 1963. Buesseler (5) indicated that the Pu signal from the Bikini and Enewetak test series could be distinguished on the basis of the much higher proportion of 240Pu relative to 239Pu. In the Sagami Nada sediment core, the 240Pu/239Pu ratio was almost constant with a value around 0.24 from the surface down to the layer of 12-14 cm, while it reached a plateau with 240Pu/239Pu ratio higher than 0.27, starting from the layer of 16-18 cm. This vertical profile was similar to the observation made by Buesseler (5) in a Pacific coral sample. Kato et al. (28) estimated the sedimentation rate in Sagami Bay. By applying the average sedimentation rate of 0.4 cm/yr in Sagami Nada, the same sampling site of our sediment core, the high 240Pu/239Pu ratio plateau in the layers of 16-22 cm could be assigned to the early 1950s. Therefore, both the characteristic high 240Pu/239Pu ratios and the corresponding deposition chronological information indicate the deposition of Pu derived from the Bikini Atoll surface test. On the basis of the above discussion, we are convinced that there were two Pu sources in Sagami Bay: (i) the global fallout, which had a maximum deposition in 1963, and (ii) the close-in fallout derived from the Bikini and Enewetak test series, which had a maximum deposition in 1954. On the basis of the vertical profiles of 240Pu/239Pu ratios in sediment cores, the close-in fallout Pu was not only deposited in Sagami Bay as a pulse injection in the 1950s test period, but it has also been transported and deposited continuously since then. Buesseler (5) noted the presence of Bikini closein fallout Pu as far north as 38° N and as far west as 160° E. In this work, we find that Pu contamination originating from the Bikini and Enewetak nuclear weapon test series in the 1950s has extended westwards as far as the Japanese coast. It is of interest to note that the Pu atom ratio has hardly changed from a value of 0.24 between the global (stratospheric) fallout deposition maximum of 1963 and the present 3502
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 13, 2004
As shown in Figure 3, the concurrent detonation in Bikini Atoll and the arrival of Pu in Sagami Bay suggest a rapid connection between the equatorial Pacific and the subtropical western Pacific regarding the transportation of Pu. The direct fallout and oceanic currents would be both plausible pathways of the close-in fallout derived Pu in Sagami Bay. However, taking into account the fact that Japanese soils (including the soil vertical profile from 0 to 30 cm) have typical global fallout 240Pu/239Pu ratios (0.168-0.194) (25, 29) and that the Bikini close-in fallout Pu is being transferred to Sagami Bay continuously, it is more likely that transportation of Bikini close-in fallout Pu to Sagami Bay is mainly through an oceanic process, especially after 1958. We hypothesize that an oceanic process is responsible for the presence of Bikini close-in fallout Pu in Sagami Bay and that, in particular, the transportation route is through the North Equatorial Current (NEC) and the Kuroshio Current (see Figure 4). This hypothesis is based on the following considerations. First, a “Pu pool” exists in the water column near the Bikini and Enewetak Atolls. In the vicinity of the Marshall Islands, a high 239+240Pu concentration was observed in 1997 (30), indicating that Pu is continuously supplied from the Bikini and Enewetak Atolls even at the present, probably from both the steep slopes of the atolls and the lagoons. Hamilton et al. (11) estimated that ca. 0.2 TBq of Pu was annually released from the atolls. Also, the high 240Pu/239Pu ratio signature in the water column near the Enewetak Atoll has been characterized, ranging from 0.2 to 0.28 (2), which is consistent with the 240Pu/239Pu ratios we observed in Sagami Bay surface sediments. Second, the redistribution of Pu in the central North Pacific Ocean by the Western North Pacific gyral circulation and the Equatorial Current and Counter-Current systems has been reported (2, 10). Third, it is well-known that the westward-flowing NEC carries waters of very low biological productivity; therefore, less particle scavenging could result in a long-term residence for the largest part of the delivered Pu in the water column, which would ensure sufficient Pu supply to the Kuroshio Current. Finally, as the biggest western boundary surface current in the western Pacific, the Kuroshio Current plays an important role in the meridional transport of heat, mass, momentum, and moisture from the western Pacific warm pool to high latitudes in the North Pacific (31) due to its high speed (2.7-3.6 km/h), great thickness (1.0 km) and width (150-200 km), and high temperature (28-29 °C in summer and 22-25 °C in winter). For example, Nozaki and Zhang (32) have reported a similar rare earth element pattern between Sagami Bay water and Kuroshio water. Therefore, it is plausible that Pu delivered from NEC was further transported to Sagami Bay and rapidly removed from the water column due to enhanced scavenging of Pu from the large population of biogenic and inorganic particles present in Sagami Bay. This hypothesis can be verified by the analysis of seawater, settling particle, and surface sediment samples in the Pacific west of 160° E, from NEC to Kuroshio Current. This verification will be addressed
FIGURE 4. Transport of close-in fallout Pu from Bikini and Enewetak Atolls through North Equatorial Current and Kuroshio Current. in a future study. In fact, additional evidence supporting our hypothesis has been seen in recent literature. Kishimoto et al. (33) determined Pu in squid collected from Japanese inshore water and found that the atomic ratio of 240Pu/239Pu in all samples was higher than the value of global fallout. Kim et al. (34) have observed elevated 240Pu/239Pu atom ratios in the bottom layer sediment of the northwest Pacific Ocean, in particular, in the East Sea (Sea of Japan) near the Korean Peninsula. They also detected Pu isotopes in seas around the Korean Peninsula (35). The 240Pu/239Pu atom ratios of the 2000 m deep entire water column in the southwestern part of the East Sea (Japan Sea) were comparable to those observed in waters near the Bikini Atoll. The transportation of Pu by the branch of the Kuroshio Current has been suggested. Resolving Global Fallout and Bikini Close-in Fallout. Since two isotopically distinctive Pu sources, global fallout and Bikini close-in fallout, were identified in Sagami Bay sediments, it is possible to evaluate the individual relative contribution from global and Bikini close-in fallout Pu based on 240Pu/239Pu ratios. First, we calculated the fraction of the total Pu activity in Sagami Bay sediments from the Bikini and Enewetak tests source alone. We use a simple two endmember mixing model, which is similar to the one described by Krey et al. (36):
Y)
(Pu)B (Pu)G
)
(RG - RS)(1 + 3.66RB) (RS - RB)(1 + 3.66RG)
(1)
where (Pu) is the activity of 239+240Pu; R is the 240Pu/239Pu atom ratio; and subscripts B, G, and S refer to Bikini fallout, global fallout, and Sagami Bay sediment samples, respectively. This equation converts 240Pu/239Pu atom ratio data to activity ratios of 239+240Pu from two sources. The constant 3.66 is the ratio of the specific activities of 240Pu to 239Pu. We define
(Pu)T ) (Pu)B + (Pu)G
(2)
where (Pu)T is the total activity of 239+240Pu in the sample, and (Pu)G consists of direct global fallout and the land-origin
TABLE 3. Percentage of Bikini Fallout in Sagami Bay Sedimentsa Bikini/total (%) core ID
240Pu/239Pub
RB ) 0.3
RB ) 0.36
KT-90-04-2 KT-91-03-8 KT-91-03-P KT-91-03-6 KT-90-04-3 mean
0.231 ( 0.011 0.246 ( 0.009 0.252 ( 0.008 0.256 ( 0.012 0.237 ( 0.010
48 60 65 68 55 59 ( 8
36 44 48 51 41 44 ( 6
a The temporal variation is summarized in the Supporting Information. b Mean of 240Pu/239Pu sediment data from Table 1; the errors quoted represent the 95% confidence limit.
global fallout, which was introduced into Sagami Bay by rivers and winds (19). Hence, the percentage of Bikini fallout in Sagami Bay sediment samples can be obtained from eq 3:
(Pu)B (Pu)T
)
Y 1+Y
(3)
Krey et al. (36) reported the mass ratio of 240Pu/239Pu in global fallout to be 0.18 ( 0.02 based on a worldwide program of sampling conducted at 21 sites in 1970 and 1971 between 30° N and 60° N. The close-in fallout 240Pu/239Pu ratio estimated from the analysis of soil samples in the Marshall Islands was reported to be around 0.30 (37). Also, Buesseler (5) observed oceanic 240Pu/239Pu ratios up to 0.36 near the Bikini and Enewetak Atolls. Therefore, the percentage of Bikini fallout can be calculated in two scenarios, by using a RB value of 0.30 and 0.36, respectively. As shown in Table 3, we find that, in scenario 1 (RB ) 0.30), the contribution of Bikini fallout ranges from 48% to 68% in the sediment cores we examined, with a mean of 59%; in scenario 2 (RB ) 0.36), it ranges from 36% to 51%, with a mean of 44%. These results clearly indicate that the influence of Bikini fallout is quite significant even for the Japanese coast. VOL. 38, NO. 13, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3503
In the previous work (19), the total 239+240Pu inventory in Sagami Bay sediments was estimated to range from 36 to 474 MBq/km2 with a mean of 191 MBq/km2 based on the data obtained from 10 sediment cores. Therefore, using the obtained Bikini fallout activity percentage (59% and 44% in two scenarios), we can calculate the total inventory of Bikini fallout Pu transported into Sagami Bay sediments. Between 84 and 113 MBq/km2 of 239+240Pu from the Bikini and Enewetak test site was transported and deposited in Sagami Bay sediments, while the global fallout Pu inventory, which consisted of the direct global fallout and the land-origin global fallout, ranged from 78 to 107 MBq/km2. Considering the large inventory of 234+240Pu from Bikini fallout as compared with the direct and land-origin global fallout Pu observed in Sagami Bay, the contamination of Pu derived from the Bikini and Enewetak nuclear test series must be regarded as a longterm regional effect, instead of a pulse, local contamination. In this study, we investigated the vertical profiles of 239+240Pu and 240Pu/239Pu atom ratios in sediment cores collected from Sagami Bay of the western Northwest Pacific. We concluded that the excess Pu inventory observed in the sediment cores was due to the presence of transported closein fallout Pu from the Bikini and Enewetak surface nuclear weapon test series in the 1950s, besides the increase of Pu inventory due to land-origin global fallout Pu. The sediment core collected in the center of Sagami Bay clearly recorded a maximum Pu deposition in 1963 and the transported Bikini close-in Pu with high 240Pu/239Pu atom ratio, which provided a time marker for the study on recent sedimentation in this region. By using a two fallout end-member model, we found that the transported close-in fallout was a major source of Pu in Sagami Bay, and its contribution ranged from 44% to 59%. The identification of Bikini close-in fallout Pu in Sagami Bay will allow for model development and insights that are of general significance for better understanding of oceanic transport of weapons testing related Pu. Further study will focus on the Pu scavenging process by biogenic and inorganic particles using the sediment trap sampling technique to elucidate the details of the oceanic Pu transportation process.
Acknowledgments We thank the researchers and the crew of the R/V Tansei Maru during the KT-90-04 and KT-91-03 cruises for their help in the sampling. We also thank Dr. M. Kusakabe, Dr. T. Aono, and Dr. Z. L. Wang for their many valuable suggestions and fruitful discussions during this study. We thank referees for their useful and constructive comments on the earlier version of the manuscript.
(11)
(12) (13) (14) (15) (16) (17) (18)
(19) (20) (21) (22) (23) (24)
(25)
(26) (27) (28) (29) (30)
(31)
Supporting Information Available The temporal variation (down core distribution) of Bikini close-in fallout Pu contribution in Sagami Bay sediments. This material is available free of charge via the Internet at http://pubs.acs.org.
(32)
Literature Cited
(33)
(1) UNSCEAR. Sources and Effects of Ionizing Radiation; United Nations Scientific Committee on the Effects of Atomic Radiation Exposures to the Public from Man-made Sources of Radiation; United Nations: New York, 2000. (2) Nakano, M.; Povinec, P. P. J. Environ. Radioact. 2003, 69, 85106. (3) Cochran, J. K. Geochim. Cosmochim. Acta 1985, 49, 1195-1210. (4) Livingston, H. D.; Anderson, R. F. Nature 1983, 303, 228-231. (5) Buesseler, K. O. J. Environ. Radioact. 1997, 36, 69-83. (6) Yamada, M.; Nagaya, Y. J. Radioanal. Nucl. Chem. 2000, 245, 273-279. (7) Yamada, M.; Aono, T. Sci. Total Environ. 2002, 287, 97-105. (8) Bowen, V. T.; Noshkin, V. E.; Livingston, H. D.; Volchok, H. L. Earth Planet. Sci. Lett. 1980, 49, 411-434. (9) Fowler, S. W.; Ballestra, S.; LaRosa, J.; Fukai, R. Deep-Sea Res. 1983, 30, 1221-1233. (10) Livingston, H. D.; Povinec, P. P.; Ito, T.; Togawa, O. The behavior of plutonium in the Pacific Ocean. In Plutonium in the
(34)
3504
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 13, 2004
(35) (36)
(37)
Environment; Kudo, A., Ed.; Elsevier Science Ltd.: Amsterdam, 2001; pp 267-292. Hamilton, T. F.; Millies-Lacroix, J. C.; Hong, G. H. 137Cs (90Sr) and Pu isotopes in the Pacific Ocean: sources and trends. In Radionuclides in the Oceans (RADOC 96-97) Inputs and Inventories; Gueguenia, P., Germain, P., Metivier, H., Coordinators; Les Editions de Physique: Les Ulis, France, 1996; pp 29-58. IAEA. Sediment Kds and Concentration Factors for Radionuclides in the Marine Environment; Technical Reports Series 247; IAEA: Vienna, 1985; pp 2-20. Sholkovitz, E. R. Earth Sci. Rev. 1983, 19, 95-161. Beasley, T. M.; Carpenter, R.; Jennings, C. D. Geochim. Cosmochim. Acta 1982, 46, 1931-1946. Nagaya, Y.; Nakamura, K. J. Oceanogr. Soc. Jpn. 1987, 43, 345355. Nagaya, Y.; Nakamura, K. J. Oceanogr. 1992, 48, 23-35. Nakamura, K. J.; Nagaya, Y. J. Radioanal. Nucl. Chem. 1990, 138, 153-164. Nakamura, K. J.; Nagaya, Y. Accumulation of 137Cs and 239,240Pu in sediments of the coastal sea and the North Pacific. In Marine and Estuarine Geochemistry; Sigleo, C., Hattori, H., Eds.; Lewis Publishers: Chelsea, MI, 1985; pp 171-180. Yamada, M.; Nagaya, Y. J. Radioanal. Nucl. Chem. 2000, 246, 369-378. Buesseler, K. O.; Sholkovitz. E. R. Geochim. Cosmochim. Acta 1987, 51, 2623-2637. Ketterer, M. E.; Watson, B. R.; Matisoff, G.; Wilson, C. G. Environ. Sci. Technol. 2002, 36, 1307-1311. Zheng, J.; Yamada, M.; Wang, Z. L.; Aono, T.; Kusakabe, M. Anal. Bioanal. Chem. April 30, 2004; DOI: 10.1007/s00216-0042626-x. Muramatsu, Y.; Uchida, S.; Tagami, K.; Yoshida, S.; Fujikawa, T. J. Anal. At. Spectrom. 1999, 14, 859-865. UNSCEAR. Ionizing Radiation: Sources and Effects; United Nations Scientific Committee on the Effects of Atomic Radiation Exposures to the Public from Man-made Sources of Radiation; United Nations: New York, 1982; p 211. Muramatsu, Y.; Yoshida, S.; Tagami, K.; Uchida, S.; Ruhm, W. ICP-MS analysis of environmental plutonium. In Plutonium in the Environment; Kudo, A., Ed.; Elsevier Science Ltd.: Amsterdam, 2001; pp 63-77. Hirose, K.; Igarashi, Y.; Aoyama, M.; Kim, C. K.; Kim, C. S.; Chang, B. W. J. Environ. Monit. 2003, 5, 302-307. Hirose, K.; Aoyama, M.; Katsuragi, Y.; Sugimura, Y. J. Meteorol. Soc. Jpn. 1987, 65, 259-277. Kato, Y.; Kitazato, H.; Shimanaga, M.; Nakatsuka, T.; Shirayama, Y.; Masuzawa, T. Prog. Oceanogr. 2003, 57, 77-95. Muramatsu, Y.; Yoshida, S.; Tanaka, A. J. Radioanal. Nucl. Chem. 2003, 255, 477-480. Povinec, P. P.; Livingston, H. D.; Shima, S.; Aoyama, M.; Gastaud, J.; Goroncy, I.; Hirose, K.; HuynhNgoc, L.; Ikeuchi, Y.; Ito, T.; La Rosa, J.; Liong Wee Kwong, L.; Lee, S. H.; Moriya, H.; Mulsow, S.; Oregioni, B.; Pettersson, H.; Togawa, O. Deep Sea Res. II 2003, 50, 2607-2637. Qiu, B. Kuroshio and Oyashio currents. In Encyclopedia of Ocean Sciences; Steele, J. H., Ed.; Academic Press: San Diego, 2001; pp 1413-1425. Nozaki, Y.; Zhang, J. The rare earth elements and yttrium in the coastal/offshore mixing zone of Tokyo Bay waters and the Kuroshio. In: Biogeochemical Process and Ocean Flux in the Western Pacific; Sakai, H., Nozaki, Y., Eds.; Terra Scientific Publishing Company: Tokyo, 1995; pp 171-184. Kishimoto, T.; Sanada, T.; Sato, K.; Higuchi, H. J. Radioanal. Nucl. Chem. 2002, 252, 395-398. Kim, C. K.; Kim, C. S.; Chang, B. U.; Choi, S. W.; Chung, C. S.; Hong, G. H.; Hirose, K.; Pettersson, H. B. L. J. Radioanal. Nucl. Chem. 2003, 258, 265-268. Kim, C. K.; Kim, C. S.; Chang, B. U.; Choi, S. W.; Chung, C. S.; Hong, G. H.; Hirose, K.; Igarashi, Y. Sci. Total Environ. 2004, 318, 197-209. Krey, P. W.; Hardy, E. P.; Pachucki, C.; Rourke, F.; Coluzza, J.; Benson, W. K. Mass isotopic composition of global fallout plutonium in soil. In Transuranium Nuclides in the Environment; IAEA-SM-199/39; IAEA: Vienna, 1976; pp 671-678. Muramatsu, Y.; Hamilton, T.; Uchida, S.; Tagami, K.; Yoshida, S.; Robison, W. Sci. Total Environ. 2001, 278, 151-159.
Received for review October 26, 2003. Revised manuscript received April 6, 2004. Accepted April 19, 2004. ES035193F
