Detection of radioactive 35S at Fukushima and other Japanese sites

Authors


Abstract

[1] The Fukushima nuclear power plant was severely damaged by an earthquake and concomitant tsunami during March 2011. An effect of this disaster was secondary formation of radioactive 35S via the 35Cl(n,p)35S reaction, when neutrons from the partially melted reactor cores activated the coolant sea water. Here we report the first measurements of 35S in sulfate aerosols and rain water collected at six Japanese sampling sites, Hokkaido, Tsukuba, Kashiwa, Fuchu, Yokohama, and Fukushima, during March-September 2011. The measured 35SO42- concentrations in aerosols vary significantly. The Kashiwa (AORI) site shows the highest 35SO42- concentration (6.1 × 104 ± 200 atoms/m3) on 1 April 2011, which is nearly 100 times higher than the natural background activity. Considering the percentage loss of 35SO42- resulting from dry and wet deposition and dilution of the radiation plume in the boundary layer during transport, it was determined that the surface air concentration of 35SO42- at the Fukushima would have been 2.8 × 105 atoms/m3 during the week after the earthquake, which is in agreement with the model prediction [Priyadarshi et al., 2011a]. 35SO42- activity in rain water collected during March-May 2011 at Tokyo Tech Yokohama varies from 1.1 × 105 to 9.8 × 105 atoms/liter, whereas stream water collected near Fukushima was found to have 1.2 × 105 atoms/liter during April. Even after 6 months, 35SO42- activity remains very high (9.9 × 104 ± 770 atoms/m3) in the marine boundary layer in the Fukushima region, which implies that the reactor core was producing radioactive sulfur.

Introduction

[2] On 11 March 2011, a magnitude 9.0 earthquake occurred in the western Pacific Ocean, with its epicenter approximately 72 km east of the Oshika peninsula of Tohoku Japan (see http://earthquake.usgs.gov/earthquakes/eqinthenews/2011/usc0001xgp/). The earthquake triggered a catastrophic tsunami with severe collateral damage. In addition to a significant loss of life and infrastructure destruction, the tsunami caused damage to the Fukushima Di-ichi nuclear power plant (http://www.iaea.org). Immediately after the earthquake, the reactors at the Fukushima nuclear power plant were automatically shut down and boron–carbon control rods inserted between the fuel columns to absorb neutrons and halt the nuclear chain reaction. Even after the shutdown, the reactor core required a prolonged cooling, because the uranium fuel continues to decay into radioactive byproducts and to release heat. A preliminary computer model by the U.K. National Nuclear Laboratory showed that, even after shut down of the Fukushima nuclear power plant, the radioactive byproducts of the fission reaction still generated 7 megawatts of heat [Brumfiel, 2011a]. Because both the regular and the emergency backup power supply (diesel generators) that powered the cooling system of the power plant were severely damaged by the tsunami waves, the reactor core became sufficiently hot that it initiated a melt down and released hydrogen gas that eventually ignited and caused explosions. The core damage was estimated to be 55%, 35%, and 30% for Units 1, 2, and 3, respectively (http://www.iaea.org). Even after 6 months of continuous cooling, the reactor core was too hot to allow access [Brumfiel, 2011b]. As the result of these events, a significant amount of radioactivity was released into the atmosphere [Chino et al., 2011; Morino et al., 2011]. Potentially dangerous levels of radiation were detected nearly 30-40 km away from the nuclear plant. Approximately 4 × 1011 neutrons/m2 were emitted during the first week following the earthquake, which led to the formation of radioactive 35S [Priyadarshi et al., 2011a]. During the first few weeks (13 March to 26 March 2011) after the earthquake and tsunami, several hundred tons of sea water were pumped into the partially melted reactor core as a coolant. Neutrons emitted from the reactor core were absorbed or captured by the constituents of sea water to produce a variety of radioactive isotopes, mainly 24Na by interaction of neutrons with stable sodium via an (n,γ) reaction and 35S produced by interaction of neutrons with stable chlorine via an (n,p) reaction [Dryssen and Nyman, 1955; Love and Sam, 1962]. Since 24Na has a very short half-life (15 h), the total radioactivity after a week was due nearly exclusively to 35S [Dryssen and Nyman, 1955].

[3] Once produced, 35S was oxidized to 35SO2 gas, which was eventually further oxidized to 35SO42- aerosol in the atmosphere, similar to the natural atmospheric process [Brothers et al., 2010; Priyadarshi et al., 2011b; Tanaka and Turekian, 1991]. It was subsequently transported in the atmosphere depending on air mass trajectories and removed from the atmosphere by dry/wet deposition and radioactive decay. 35S is a unique tracer in that it provides information on the number of neutrons emitted from the reactor core and can be used to probe the condition of the reactor core as well as the containment vessel. Because of the unavailability of samples from near the Fukushima power plant, there have been no measurements of 35S until now. Based on 35S measurements of atmospheric sulfate and SO2 samples collected at La Jolla, California, and concomitant model calculations, the concentration of 35SO42- in surface air was estimated to be 2 × 105 atoms/m3 at Fukushima during March 2011 [Priyadarshi et al., 2011a]. Here we report the first measurement of radioactive 35S in sulfate aerosols and rain water collected at different sites in Japan (Figure 1) and compare the data with the model calculation of Priyadarshi et al. [2011a].

Figure 1.

Details of the sampling sites. Aerosol samples were collected at six different places in close proximity to the Fukushima nuclear power plant (37.25°N, 141.02°E): Tokyo Institute of Technology (Tokyo Tech), Yokohama (35.52°N, 139.48°E); Atmosphere and Ocean Research and Institute (AORI), University of Tokyo, Kashiwa, Chiba Prefecture (35.90°N, 139.94°E); Tokyo University of Agriculture and Technology (TUAT), Fuchu, Tokyo (35.68°N, 139.48°E); National Institute of Environmental Studies (NIES), Tsukuba, Ibaraki Prefecture (36.05°N, 140.12°E); Hokkaido Research Organization, Sapporo (43.08°N, 141.34°E); and Kawamata town, Fukushima prefecture (37.40°N, 140.36°E). These sampling sites except for Sapporo are located within a 250 km radius and south of the Fukushima nuclear plant.

Materials and Methods

[4] Aerosol samples were collected at six different places in close proximity to the Fukushima nuclear power plant (37.25°N, 141.02°E; Figure 1): Tokyo Institute of Technology (Tokyo Tech) Yokohama (35.52°N, 139.48°E); Atmosphere and Ocean Research and Institute (AORI), University of Tokyo, Kashiwa, Chiba Prefecture (35.90°N, 139.94°E); Tokyo University of Agriculture and Technology (TUAT), Fuchu, Tokyo (35.68°N, 139.48°E); National Institute of Environmental Studies (NIES), Tsukuba, Ibaraki Prefecture (36.05°N, 140.12°E); Hokkaido Research Organization, Sapporo (43.08°N, 141.34°E); and Kawamata town, Fukushima prefecture (37.40°N, 140.36°E). These sampling sites, except for Sapporo, are located within a 250 km radius and south of the Fukushima nuclear plant. The aerosol samples were collected from March through September 2011 using a high-volume air sampler. Rain water samples were collected only at Tokyo Tech during March-May 2011. In addition, a water sample from a local water stream, situated in Fukushima prefecture, nearly 50 km away from the Fukushima nuclear power plant, was collected in April 2011. The samples were analyzed at the University of California, San Diego. The details of the sample processing for analyzing 35S activity in sulfate aerosols have been described by Brothers et al. [2010]. Three or four liters of rainwater was passed through a pre-prepared anion resin column to trap sulfate on the resin surface, and 15 ml of HBr (1 M) was passed through the resin column to elute sulfate ions from the resin. The solution was then neutralized by adding Ag2O and oven dried. Sulfate was cleaned of organics as discussed by Brothers et al. [2010]. The 35S activity was counted in an ultralow-level liquid scintillation spectrometer (Wallac 1220 Quantulus) and was corrected for the background activity (which is 1.07 DPM and probably is due to the radioactivity contributed from vial material and the scintillation gel used for counting) and the decay time from the sample collection date. The natural variation in 35SO42- concentration, as reported by Priyadarshi et al. [2011b, 2012], is from 130 to 900 atoms/m3 and from 200 to 1600 atoms/m3 at Scripps Pier in southern California and in Antarctica, respectively.

Results and Discussion

[5] 35S activities measured in sulfate aerosols collected at Tokyo Tech, AORI, TUAT, NIES, Hokkaido, and Fukushima (different aerosol size fraction) are shown in Figures 2 and 3. Even though the sampling sites are relatively close to each other (southward direction from Fukushima, except for Hokkaido), a large variation in 35SO42- activities was observed during March-April 2011. 35SO42- concentrations were 0.26 × 103 − 11.78 × 103, 0.69 × 103 − 61.4 × 103, 0.51 × 103 − 1.42 × 103, 1.41 × 103 − 18.05 × 103, 0.56 × 103  −  2.63 × 103 atoms/m3 at Tokyo Tech, AORI, TUAT, NIES, and Hokkaido, respectively (Table 1). The 35S activities observed at Tokyo Tech, AORI and NIES are significantly higher than those observed at TUAT, NIES, and Hokkaido sites. At Fukushima, 35SO42- varies from 8.1 × 103 to 9.9 × 104 atoms/m3 and from 46 to 62 atoms/m3 in fine and coarse fractions, respectively (Table 2). 35SO42- concentrations at AORI (6.1 × 104 atoms/m3) and Fukushima (1.2 × 105 atoms/m3) are the highest 35S activities ever measured in any atmospheric sample and are nearly 100 times higher than the natural background. Radioactive 35S (half-life 87 days) is also produced by the interaction of cosmic rays with 40Ar in the Earth's atmosphere [Lal and Peters, 1967]. The natural background 35SO42- concentration in the atmosphere varies from 300 to 900 atoms/m3 [Brothers et al., 2010; Priyadarshi et al., 2012]. Even at Antarctica, where the production rate of 35S is maximal [Lal and Peters, 1967], 35SO42- varies from 120 to 1600 atoms/m3 [Priyadarshi et al., 2011b]. Such a high concentration of 35SO42- at AORI indicates the presence of another source of 35S in the marine boundary layer. Priyadarshi et al. [2011b] demonstrated that the partially damaged reactor core of the Fukushima nuclear power plant produced radioactive 35S via 35Cl(n,p)35S reaction during the first few weeks following the earthquake. The Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model developed by NOAA's Air Resources Laboratory (ARL) [Draxler and Rolph, 2011] was used to calculate the air mass back-trajectories to determine the origin and pathway of the air masses affecting the sampling sites involved in this study. We considered the air mass mixing and its transport within the boundary layer; the backward air mass trajectories were calculated for 72 h at three different altitudes (10, 500, and 1000 m) over each sampling station. As shown in Figure 4, surface air masses at AORI arrived from the north near Fukushima on 1 April 2011 and are thus responsible for the observed spike in 35SO42- activity (6.1 × 104 atoms/m3). The sampling site at Hokkaido lies north of Fukushima and is thus not affected by the 35S emission at Fukushima. A possible reason for lower 35SO42- activity at TUAT is its geographical location and the prevailing air mass transport vectors. TUAT is situated in Fuchu city, which has many terraces and hills across it. The Fuchu and Kokobunji hills align from west to east whereas the Sengen-Yama hill is located in the northeastern side, which may affect the air mass trajectories coming from Northeast Fukushima. Fine-scale air mass trajectory analysis is not available, but future modeling efforts could be instrumental in developing this.

Figure 2.

35SO42− activity measured at five different sampling sites near Fukushima. The maximum 35SO42− activity (6.1 × 104 atoms/m3) was observed at AORI on 1 April 2011.

Figure 3.

35SO42− measured in different aerosol size fractions (Q5, Q6, Q7, representing aerosol size fraction between 0.69 and 1.3 µm, between 0.39 and 0.69 µm, and less than 0.39 µm, respectively) collected at Fukushima prefecture. 35SO42− in the fine particles (Q7<0.39 µm) is nearly 100 times higher than the natural background and shows that, even after 6 months, of the nuclear disaster, the reactor core was active and producing 35S.

Table 1. Measurement of radioactive 35S in sulfate aerosol collected at different sampling sites near Fukushima
Sample IDCollection Time (m/d/y)Air Volume (m3)35SO42- Atoms per m3 (a103)
Starting Date End Date
  • a

    The standard error associated with individual measurement.

Tokyo Institute of Technology (Tokyo Tech), Yokohama
13/25/11, 3/26/119913.06 ± 0.12a
23/26/11, 3/26/117350.56 ± 0.11
33/26/11, 3/27/117240.72 ± 0.12
43/27/11, 3/28/1120011.25 ± 0.04
53/28/11, 3/30/11294511.78 ± 0.07
63/30/11, 4/1/1129600.73 ± 0.03
74/1/11, 4/3/1129310.48 ± 0.03
84/3/11, 4/5/1129800.35 ± 0.03
94/5/11, 4/7/1129690.41 ± 0.03
104/7/11, 4/9/1129731.03 ± 0.07
114/9/11, 4/11/1129621.75 ± 0.08
124/11/11, 4/13/1129590.94 ± 0.07
134/13/11, 4/15/1129801.11 ± 0.07
144/15/11, 4/17/1129840.88 ± 0.07
154/17/11, 4/19/1129931.25 ± 0.08
164/19/11, 4/21/1129610.46 ± 0.07
174/21/11, 4/23/1129800.79 ± 0.07
184/23/11, 4/25/1129850.77 ± 0.07
194/25/11, 4/28/1144790.76 ± 0.05
204/28/11, 5/1/1144130.70 ± 0.05
215/1/11, 5/4/1144950.93 ± 0.07
225/4/11, 5/8/1128841.03 ± 0.09
235/8/11, 5/10/1129800.88 ± 0.09
245/10/11, 5/13/1145110.62 ± 0.06
255/13/11, 5/17/1159491.22 ± 0.05
265/17/11, 5/20/1144671.32 ± 0.06
275/20/11, 5/24/1159590.35 ± 0.04
285/24/11, 5/27/1144930.49 ± 0.05
295/27/11, 5/30/1141380.50 ± 0.05
305/30/11, 6/3/1162000.74 ± 0.04
316/3/11, 6/7/1159000.97 ± 0.04
326/7/11, 6/10/1144150.34 ± 0.04
336/10/11, 6/14/1159690.73 ± 0.04
346/14/11, 6/17/1144690.26 ± 0.04
356/17/11, 6/21/1158870.29 ± 0.03
366/21/11, 6/24/1144920.55 ± 0.04
376/24/11, 6/28/1159610.34 ± 0.03
386/28/11, 7/1/1144510.62 ± 0.04
Atmosphere and Ocean Research Institute (AORI), Kashiwa, Chiba Pref.
393/23/11, 3/25/11130317.08 ± 0.10
403/25/11, 3/28/1123530.70 ± 0.04
413/28/11, 3/30/11159713.90 ± 0.11
423/30/11, 4/1/11156861.40 ± 0.20
434/1/11, 4/3/1115530.96 ± 0.05
444/3/11, 4/5/1115861.14 ± 0.05
454/5/11, 4/7/1115841.65 ± 0.13
464/7/11, 4/9/1115411.62 ± 0.14
474/9/11, 4/11/1115843.44 ± 0.15
484/11/11, 4/13/1115742.58 ± 0.15
494/13/11, 4/15/1115581.54 ± 0.14
504/15/11, 4/17/1115391.78 ± 0.15
514/17/11, 4/19/1115664.11 ± 0.17
524/19/11, 4/21/1115792.35 ± 0.15
Tokyo University of Agriculture and Technology, Fuchu, Tokyo
533/24/11, 3/26/1128370.65 ± 0.03
543/26/11, 3/28/1128850.52 ± 0.03
553/28/11, 3/30/1128551.22 ± 0.04
563/30/11, 4/1/1128511.16 ± 0.03
574/1/11, 4/3/1129390.52 ± 0.03
584/3/11, 4/5/1127791.00 ± 0.08
594/5/11, 4/7/1128690.97 ± 0.08
604/7/11, 4/9/1128681.05 ± 0.08
614/9/11, 4/11/1128851.42 ± 0.08
624/11/11, 4/13/1128820.81 ± 0.08
634/13/11, 4/15/1127430.93 ± 0.08
644/15/11, 4/20/1128990.87 ± 0.07
654/20/11, 4/22/1129110.70 ± 0.07
664/22/11, 4/24/1128570.97 ± 0.10
674/24/11, 4/26/1128571.26 ± 0.10
684/26/11, 4/28/1128491.05 ± 0.10
694/28/11, 5/13/1128250.78 ± 0.09
National Institute of Environmental Studies (NIES), Tsukuba, Ibaraki Pref.
703/24/ 11, 3/29/1170655.02 ± 0.03
713/29/11, 4/1/11423018.05 ± 0.14
724/1/11, 4/4/1142244.39 ± 0.08
734/4/11, 4/7/1142841.41 ± 0.06
744/7/11, 4/11/1156861.94 ± 0.05
Hokkaido Research Organization, Sapporo
753/25/11, 3/28/1139531.26 ± 0.10
763/28/11, 3/30/1129331.20 ± 0.08
773/30/11, 4/1/1129330.96 ± 0.08
784/1/11, 4/3/1129180.78 ± 0.07
794/3/11, 4/5/1129190.76 ± 0.07
804/5/11, 4/7/1128570.95 ± 0.08
814/7/11, 4/9/1129191.08 ± 0.08
824/9/11, 4/11/1129231.05 ± 0.08
834/11/11, 4/13/1129220.96 ± 0.08
844/13/11, 4/15/1129241.04 ± 0.07
854/15/11, 4/17/1129292.63 ± 0.10
864/17/11, 4/19/1129300.56 ± 0.07
874/19/11, 4/21/1129320.83 ± 0.10
884/21/11, 4/23/1129310.75 ± 0.10
894/23/11, 4/25/1129370.62 ± 0.09
Table 2. 35S Measurement in Aerosol Particles of different Size Fraction Collected From Fukushima Prefecture, ~100 km From the Fukushima Nuclear Power Planta
Sample IDCollection Time (m/d/y) Start Time End TimeAir volume (m3)35SO42- atoms/m3 (×103)
  • a

    Higher 35S activity is associated with fine sulfate particles, which are produced mainly from gas-phase oxidation of 35SO2. Q1, Q2, Q3, Q4, Q5, Q6, and Q7 represent aerosol size fraction of >10.2 mm, 4.2-10.2 mm, 2.1-4.2 mm, 1.3-2.1 mm, 0.69-1.3 mm, 0.39-0.69 mm, and <0.39 mm, respectively.

Cedar forest, Kawamata-cho, Fukushima prefecture
0718-Q17/9/11, 7/18/1166160.34 ± 0.04
0718-Q2  0.65 ± 0.05
0718-Q3  0.52 ± 0.04
0718-Q4  1.41 ± 0.06
0718-Q5  1.34 ± 0.06
0718-Q6  30.41 ± 0.18
0718-Q7  57.53 ± 0.24
0725-Q17/18/11, 7/25/1157830.20 ± 0.05
0725-Q2  0.15 ± 0.05
0725-Q3  0.13 ± 0.05
0725-Q4  0.56 ± 0.06
0725-Q5  8.12 ± 0.12
0725-Q6  10.31 ± 0.14
0725-Q7  6.02 ± 0.11
0801-Q17/25/11, 8/1/1157460.27 ± 0.05
0801-Q2  0.43 ± 0.05
0801-Q3  0.46 ± 0.05
0801-Q4  2.06 ± 0.08
0801-Q5  14.56 ± 0.15
0801-Q6  17.97 ± 0.17
0801-Q7  50.38 ± 0.26
0815-Q18/8/11, 8/15/1158700.20 ± 0.05
0815-Q2  0.48 ± 0.050
0815-Q3  0.50 ± 0.05
0815-Q4  4.55 ± 0.10
0815-Q5  28.39 ± 0.20
0815-Q6  56.0 ± 0.27
0815-Q7  34.12 ± 0.22
0824-Q18/15/11, 8/24/1174440.16 ± 0.04
0824-Q2  0.19 ± 0.04
0824-Q3  0.25 ± 0.04
0824-Q4  1.64 ± 0.06
0824-Q5  10.72 ± 0.12
0824-Q6  0.52 ± 0.05
0824-Q7  28.03 ± 0.19
0830-Q18/24/11, 8/30/1149750.21 ± 0.05
0830-Q2  0.18 ± 0.05
0830-Q3  0.22 ± 0.05
0830-Q4  1.60 ± 0.08
0830-Q5  11.95 ± 0.15
0830-Q6  16.74 ± 0.17
0830-Q7  14.75 ± 0.16
0916-Q18/30/11, 9/16/1112,7620.05 ± 0.02
0916-Q2  0.04 ± 0.02
0916-Q3  0.05 ± 0.02
0916-Q4  0.43 ± 0.03
0916-Q5  4.59 ± 0.06
0916-Q6  8.58 ± 0.08
0916-Q7  8.14 ± 0.07
School ground, Kawamata-cho, Fukushima prefecture
0715-Q17/9/11, 7/13/1131050.63 ± 0.09
0715-Q2  0.48 ± 0.09
0715-Q3  0.77 ± 0.10
0715-Q4  3.96 ± 0.14
0715-Q5  7.24 ± 0.17
0715-Q6  20.74 ± 0.26
0715-Q7  62.32 ± 0.41
Agriculture field, Kawamata-cho, Fukushima prefecture
0722-Q17/9/11, 7/22/1199090.12 ± 0.03
0722-Q2  0.15 ± 0.03
0722-Q3  0.21 ± 0.03
0722-Q4  0.68 ± 0.03
0722-Q5  16.52 ± 0.13
0722-Q6  1.14 ± 0.04
0722-Q7  50.64 ± 0.20
0819-Q18/17/11, 8/19/1114600.56 ± 0.16
0819-Q2  0.44 ± 0.16
0819-Q3  0.41 ± 0.16
0819-Q4  1.35 ± 0.18
0819-Q5  7.74 ± 0.27
0819-Q6  12.39 ± 0.31
0819-Q7  8.84 ± 0.28
0908-Q18/27/11, 9/7/1114600.56 ± 0.17
0908-Q2  0.50 ± 0.17
0908-Q3  0.49 ± 0.18
0908-Q4  1.21 ± 0.20
0908-Q5  17.02 ± 0.38
0908-Q6  1.17 ± 0.19
0908-Q7  99.98 ± 0.77
Figure 4.

HYSPLIT-3 days air masses back-trajectories were calculated over the AORI sampling site at three different altitudes (10, 50, and 1000 m). The surface air masses arrive from northwestern regions of Japan near Fukushima and are mainly responsible for the spike in 35SO42 activity observed on 1 April 2011.

[6] Based on 35S activity measurements in California and a moving box model, the surface air concentration of 35SO42- at Fukushima was predicted to be 2 × 105 atoms/m3 in the week following the earthquake [Priyadarshi et al., 2011a]. The present data show a maximum activity of 6.1 × 104 atoms/m3 at AORI on 1 April 2011, which is about three times lower than the model estimated value. Because the AORI site is nearly 250 km away from the Fukushima nuclear plant, a dilution of the radiation plume in the boundary layer and loss of 35SO42- aerosol particles by dry and wet deposition during the transport are expected to decrease the 35SO42- concentration significantly before reaching the AORI site. Morino et al. [2011] used a 3D chemical transport model to simulate the distribution of radioactive 131I and 137Cs over Japan (both inland and the surrounding nearby oceanic surface) during 10-30 March 2011. The model considered the emission of radionuclides (131I and 137Cs) at the Fukushima nuclear power plant, horizontal and vertical advection, diffusion affecting the radiation plume during transport, dry and wet deposition of gas and aerosols, and radioactive decay to determine the percentage loss of radioactive 131I and 137Cs over Fukushima and nearby prefectures. Since 137Cs behaves as an aerosol particle due to its attachment to fine aerosols [Sportisse, 2007], we compare its distribution with 35SO42- aerosols. According to Morino et al. [2011], nearly 15% of 137Cs emitted from the Fukushima nuclear power plant were deposited over the Fukushima prefecture, whereas on average, 22% of emitted 137Cs was deposited over land in Japan during March. The percentage loss was greater (8-41%) during a transient cyclone that passed over Japan during 15-17 March and 19-23 March [Morino et al., 2011].

[7] Based on the observed 35SO42- concentration peak at AORI on 1 April 2011 (Figure 2) and the model estimation of Morino et al. [2011], the surface air concentration of 35SO42- at Fukushima was estimated to be between 1.1 × 105 and 2.8 × 105 atoms/m3 during March, whereas the predicted concentration was 2 × 105 atoms/m3 [Priyadarshi et al., 2011a]. This difference may be due partially to uncertainties related to the calculation of depositional rates of radionuclides, because the collection efficiency of dry-deposited aerosols is difficult to quantify precisely [Morino et al., 2011]. In addition, the model used by Priyadarshi et al. [2011a] to calculate the concentration of 35SO42- (2 × 105 atoms/m3) at Fukushima contains unavoidable uncertainty. The model is particularly sensitive to the dilution rate of the radiation plume during the long-range transport from Fukushima to La Jolla, California. A 10% change in dilution rate in the model would change the model output (surface air concentration of 35SO42-, i.e., 2 × 105 atoms/m3) by 20%.

[8] At Fukushima, 35SO42- activity measured in fine sulfate aerosols are higher than in the coarse fraction during August through September 2011 (Figure 3 and Table 3). This is because 35S produced at the reactor was oxidized to 35SO2 and subsequently oxidized to 35SO42- in the fine fraction by gas-phase oxidation of 35SO2. We note that, even after 6 months, 35S activity was very high in the marine boundary layer in the Fukushima region, which implies that the reactor core was still active and releasing neutrons. However, the presence of a viable chlorine source is not known. The neutrons might be reacting either with residual evaporated salt deposits or with sea water coming in and out across a crack developed in the containment vessel. The reason for the higher 35S activity observed during September compared with July-August is not yet clear and warrants extended future sampling.

Table 3. 35S measured in rain water samples collected in tokyo and Fukushima during March-May 2011 shows that 35S/liter of rain water was 10 times higher than that in background rain water, as observed at La Jolla, (California) raina
Rain SamplesSampling dateAverage Rainfall Rate (mm/h)35SO42- (Atoms/Liter)Cab (Atoms/m3)K (s-1)
  • a

    The rain-scavenging coefficient (k) was calculated based on the 35SO42− concentration measured in rainwater and air.

  • b

    The concentration of 35SO42- in air collected at Tokyo Tech, Yokohama.

  • c

    Water sample collected from a stream situated in Fukushima prefecture, 50 km away from Fukushima nuclear power plant.

Tokyo Tech #13/24/111.49.8E + 05----
#24/25/1122.1E + 057707.6 b 10−3
#35/11/1121.9E + 056168.7 b 10−3
#45/13/112.61.1E + 056166.5 b 10−3
#55/24/111.72.2E + 053441.5 b 10−2
Fukushima stream waterc4/28/11 1.2E + 05  
La Jolla rain1/26/10 6.2E + 04  

[9] The 35SO42- concentration in rain water collected during March-May 2011 at Tokyo Tech Yokohama varies from 1.1 × 105 to 9.8 × 105 atoms/liter, whereas stream water collected near Fukushima was found to have 1.2 × 105 atoms/liter during April (Table 3). The concentration in rain water is nearly 10 times higher than the 35SO42- concentration (6.2 × 104 atoms/liter) contained in rain water collected at La Jolla, California, during January 2010. Atmospheric processes such as rainout or washout cleanse the atmosphere by removing aerosol particles and gases. Model simulation dealing with long-range transport of dust and aerosols and deposition on the surface/ocean depends sensitively on the adapted aerosol-scavenging coefficient. We utilized 35S measurements in aerosols and rainwater to calculate the aerosol-scavenging coefficient. The scavenging coefficient (k) is defined as [Okita et al., 1996]:

display math

[10] where Cw and Ca are the 35S concentration in rain water (mg/liter) and air (µg/m3), respectively; P is the precipitation rate (cm/s); and h is the height of the cloud top (cm). We do not have measurements of the rate of precipitation of each individual event. The average precipitation rate (varying between 1.4 and 2.6 mm/h) was taken from the Japan Meteorological Agency and is given in Table 3. The height of the cloud top was assumed to be 2 km based on the fact that rainfall rate is more significant below the melting height (2-3 km), where the main wet removal occurs in cloud [Mittermaier and Illingworth, 2003]. After this height, the precipitation rate decreases and becomes less intense [Scott, 1982]. A recent measurement also shows that the mean cloud top height in the Hawaiian region is 2.1 km [Zhang et al., 2012]. For our sample, the value of k varies from 6.5 × 10-3/s to 1.5 × 10-2/s, which agrees with other measurements [Andronache, 2004; Chate et al., 2011; Laakso et al., 2003; Maria and Russell, 2005; Okita et al., 1996; Schumann, 1989]. The calculated values from Andronache [2004] vary from 2.1 × 10-3/s to 4.3 × 10-4/s, for corresponding precipitation rates of 1.5 mm/h and 4.7 mm/h.

[11] It has been demonstrated that 35S is a unique tracer in understanding air mass mixing and quantifying aerosol dry and wet deposition [Cho et al., 2011; Priyadarshi et al., 2011b, 2012; Tanaka and Turekian, 1991; Turekian and Tanaka, 1992]. Approximately 2 × 1011 atoms of 35S were produced from Fukushima within the first 6 months following the earthquake, which is lower by a factor of 1011 compared with the total global production of 2 × 1022 atoms of 35SO42- by cosmic rays, of which 5 × 1020 atoms of 35S are contained in boundary layer itself. However, because Fukushima is a very localized point source, the released 35SO42- was detected several thousand miles away from the source.

[12] The presence of excess 35SO42- (two orders of magnitude higher than the natural background) provides a unique opportunity to understand the chemical transformation, the air mass transport, and potentially the fate of 35SO42- in soil on short time scales. A yearlong sampling of aerosol, soil, and rain water collected at several sampling sites in Japan for 35S analysis along with a regional transport model will further resolve and quantify the distribution rate of regional radiogenic (and by proxy stable) sulfur in the natural environment.

Conclusions

[13] We report the first measurement of radioactive 35S in sulfate aerosol collected at six different sampling sites close to the Fukushima nuclear power plant during March-September 2011. A very high 35SO42- activity was observed at Fukushima, AORI, NIES, Tsukuba, and Tokyo Tech Yokohama that is nearly 100 times higher than the natural background 35S. Based on 35SO42- concentrations measured at AORI in April, the surface air concentration of 35S was estimated to be 2.8 × 105 atoms/m3 at Fukushima during March. Even after 6 months, 35S activity was very high in the marine boundary layer in the Fukushima region, which implies that the reactor core was still active. The reason for higher 35SO42- during September compared with July-August is probably the periodic use of seawater as a coolant. 35SO42- concentration in rainwater collected at Tokyo Tech Yokohama varies from 1.1 × 105 to 9.8 × 105 atoms/liter. The 35SO42- concentration measured in air and rainwater was employed to determine the rain-scavenging coefficient, which varies from 6.5 × 10-3 to 1.5 × 10-2 s-1.

Acknowledgments

[14] We thank the reviewers of the manuscript for their insightful comments that improved this manuscript. We also thank K. Kita for aerosol sampling at Kawamata town. This work was supported by KAKENHI (23224013 and 24110003) of the Ministry of Education, Culture, Sports, Science and Technology, Japan.

Ancillary

Advertisement