Evaluating the transport of surface seawater from 1956 to 2021 using 137 Cs deposited in the global ocean as a chemical tracer

. We analyzed the spatiotemporal variations in the 137 Cs activity concentrations in global ocean surface seawater from 1956 to 2021 using the HAMGlobal2021 (Historical Artiﬁcial radioactivity database in Marine environment, Global integrated version 2021) and other published data. The global ocean was divided into 37 boxes. When observing the 0.5-year median value of 137 Cs in each box in the Paciﬁc Ocean, we noticed that the values gradually increased or had almost constant levels in the 1950s and 1960s, and then decreased exponentially in 1970–2010, immediately before the Fukushima Daiichi Nuclear Power Plant Station (F1NPS) accident. In the northern North Atlantic Ocean and its marginal sea, the 0.5-year median values of 137 Cs showed large variations in the directly discharged 137 Cs from the reprocessing plants. The 137 Cs inventory in the surface mixed layer in 1970, when 137 Cs was released into the surface seawater, was estimated to be 184 ± 26 PBq. In 1975 and 1980, the 137 Cs inventory increased to 201 ± 27 and 214 ± 11 PBq, respectively, due to direct discharge from the Sellaﬁeld and La Hague nuclear fuel reprocessing

1 Introduction 137 Cs is regarded as one of the most abundant artificial radionuclides in the ocean because of its long half-life (30.17 years) and large fission yield that originates from large-scale atmospheric weapons tests.Atmospheric nuclear weapons tests occurred from 1945to 1980. During 1945to 1963, large-scale atmospheric nuclear weapons tests were conducted by the United States.In 1963, the Partial Nuclear Test Ban Treaty was signed, and these tests in the atmosphere by the United States, former Soviet Union, and Great Britain were conducted underground.However, France continued with atmospheric tests until 1974 and China until 1980.In addition, 137 Cs has been released into the Pacific Ocean by local fallout from ground tests on Bikini Atoll in the Marshall Islands between 1946 and 1958 as a result of tests conducted by the United States (e.g., UNSCEAR, 2000;Aoyama et al., 2006;Aoyama, 2010;Inomata, 2010).Because large amount of 137 Cs released into the atmosphere, leading to fallout onto the ocean surface, the ocean is recognized as the largest receptor of 137 Cs on Earth.Furthermore, other sources, such as the accidental release from nuclear facilities (the Three Mile Island nuclear power plant in 1979), sea dumping of nuclear wastes from nuclear facilities carried out in 1986 in the north-central East China Sea/Sea of Japan by the former Soviet Union and Russian Federation, lost nuclear weapons, and the use of radioisotopes in human activities, such as industry, medicine, and science, are recognized.These contributions in the environment are minor compared to those from the dominant sources listed above (UNSCEAR, 2000;IAEA, 2005).
The 137 Cs released by the large-scale nuclear weapons tests was largely deposited in the western part of the North Pacific Ocean at approximately 30-45 • N latitude and in the western part of the North Atlantic Ocean at approximately 30-50 • N latitude from the late 1950s to the early 1960s (UNSCEAR, 2000;Aoyama et al., 2006).By using a dataset of 137 Cs measurements in rainwater, seawater, and soil, 137 Cs deposition estimated by global fallout in the Northern Hemisphere with a 10 • × 10 • grid was 765 ± 79 PBq (Aoyama et al., 2006).The deposition of 137 Cs in the Southern Hemisphere was significantly lower than that in the Northern Hemisphere.The maximum deposition of 137 Cs in the Southern Hemisphere occurred approximately 1 year after that in the Northern Hemisphere, owing to the stratosphere air mass exchange between the northern and southern stratosphere (Hirose and Aoyama, 2003a, b).
The dissolved 137 Cs discharged from nuclear reprocessing plants, namely the Sellafield plant (Irish Sea, United Kingdom) and the La Hague facility (English Channel coast, France), are also large sources.Since 1952, the discharge of several radionuclides from the Sellafield plant has occurred in the Irish Sea.The released 137 Cs amount reached a maximum in the mid to late 1970s (Gray et al., 1995;Guegueniat et al., 1997;UNSCEAR, 2000).The discharged 137 Cs from the Sellafield plant from 1951 to 2020 was estimated to be 41.4 PBq (OSPAR, 2021).The maximum discharged 137 Cs from the La Hague plant occurred in 1971, and the amount of discharged 137 Cs decreased over time.The total amount of 137 Cs released from the La Hague plant was estimated to be 1.04 PBq (OSPAR, 2021).These discharged 137 Cs components were then transported to the northern North Atlantic Ocean, its marginal seas, and the Arctic Ocean (Smith et al., 1998;Maderich et al., 2021).
The Chernobyl accident on 26 April 1986 also released 137 Cs into the environment (e.g., Molero et al., 1999;Steinhauser et al., 2014;Miyao et al., 1998), which was estimated to be 85 ± 26 PBq (Nuclear Energy Agency, 2002).Although most of the 137 Cs derived from the Chernobyl accident was deposited on land, a significant amount of 137 Cs was released into the ocean, which was estimated to be 15-20 PBq (Aarkrog, 2003).In particular, the Chernobyl fallout increased the 137 Cs activity concentrations in the Baltic Sea (Zaborska et al., 2014), resulting in it being the most radioactive, contaminated area, with 4.5 PBq of the total inventory (CEC, 1990).The Black Sea also received the Chernobyl 137 Cs fallout in 1986 (Bezhenar et al., 2019;Egorov et al., 1999), and the inventory was estimated to be 2-3 PBq (Egorov et al., 1999).The 137 Cs released into the Black Sea flowed into the Mediterranean Sea (Bezhenar et al., 2019).In the Mediterranean Sea, the total deposition of 137 Cs from the Chernobyl accident was estimated to be 2.5 PBq in 1986 (Delfanti and Papucci, 2010).Furthermore, the fallout of 137 Cs from the Chernobyl accident also occurred as a single, small pulse in the western North Pacific Ocean and Sea of Japan (Miyao et al., 1998;Inomata et al., 2009), and the Chernobyl release contributed only a few percent compared to the previous 137 Cs water column inventory in these regions (Aoyama et al., 1986).
The Fukushima Daiichi Nuclear Power Plant Station (F1NPS) accident is also recognized as being a large source of 137 Cs.On 11 March 2011, large amounts of 137 Cs were released into the western North Pacific Ocean by atmospheric deposition and the direct discharge of liquid-contaminated stagnant water from the F1NPS because of the extraordinary earthquake and the subsequent giant tsunami (IAEA, 2015;UNSCEAR, 2013).The released 137 Cs amount by the F1NPS accident and the subsequent distributions were investigated by numerous researchers and summarized in Buesseler et al. (2017).In this study, we used our estimation because these studies were considering the mass balances among atmosphere, land, and ocean.The atmospheric deposited amount of 137 Cs into the ocean from the atmosphere was estimated to be 11.7-14.8PBq (Aoyama et al., 2016c).Directly discharged liquid 137 Cs from the F1NPS was estimated to be 3.6 ± 0.7 PBq by using the observation data around the F1NPS and model simulation (Tsumune et al., 2012(Tsumune et al., , 2013)).The 137 Cs inventory into the North Pacific Ocean in the surface mixed layer was estimated to be 15.2-18.3PBq (Aoyama et al., 2016c), which are values that Earth Syst.Sci.Data, 15, 1969Data, 15, -2007Data, 15, , 2023 https://doi.org/10.5194/essd-15-1969-2023 are consistent with the values estimated by optical statistical analysis (Inomata et al., 2016) and model simulation (Tsubono et al., 2016).It is noted that the region that contains F1NPS deposition in the western North Pacific Ocean is almost the same region as that with the global fallout of 137 Cs in the 1950s and 1960s.
Other sources, such as nuclear waste dumping, discharges from nuclear power plant operation, the release of radionuclides from satellite failures, local underwater nuclear tests, nuclear weapons accidents, and the use in industry and medicine, are considered minor contributors to 137 Cs concentrations in the global ocean (Livingston and Povinec, 2000;IAEA, 2005). 137Cs exists mainly in the dissolved form, which allows it to move with seawater.After its release into the ocean, 137 Cs undergoes radioactive decay, with a half-life of 30.17 years, during transport into the ocean by undergoing oceanic physical processes, such as advection, diffusion, and downward transport.Therefore, 137 Cs has been used as a marine tracer for decades to study physical processes, such as the longrange transport of water masses (e.g., IAEA, 2005;Hirose and Aoyama, 2003a, b;Inomata et al., 2009Inomata et al., , 2012;;Nakano et al., 2010;Povinec et al., 2003;Tsumune et al., 2003Tsumune et al., , 2011)).Furthermore, 137 Cs is used to assess the radioactive doses (or radiological effects) on the human body due to the uptake of marine foods containing anthropogenic radionuclides (e.g., IAEA, 2005;UNSCEAR, 2013).
The spatiotemporal variations in 137 Cs activity concentrations in the surface seawater in the global ocean from 1956 to 2021 were investigated in this study.This study is an extension of our previous research, in which we analyzed the measured data up to 2005 (Inomata et al., 2009).The data used in this study were adopted from HAMGlobal2021 (Historical Artificial radioactivity database in the Marine environment, Global integrated version 2021; Aoyama, 2021), which contains data from the F1NPS accident.The HAM-Global2021 database contains information on several radionuclides ( 134 Cs, 137 Cs, 90 Sr, 3 H, 239,240 Pu, 241 Am, and 14 C) in the global ocean.The data were measured from 1956 to 2021.The dataset of the International Atomic Energy Agency Marine Radioactivity Information System (IAEA MARIS) was also included in HAMGlobal2021.In addition to this, the data measured in the North Atlantic Ocean and its marginal seas by IRSN (Institut de Radioprotection et de Sûreté Nucléaire; Baily du Bois et al., 2020) were also contained in this study.
In this paper, we present the following new insights.
1.The distribution of 137 Cs in the mixed layer in the global ocean includes the following: spatiotemporal variations in 137 Cs activity concentrations in the surface seawater in the global ocean from 1956 to 2021, spatiotemporal variations in the 137 Cs inventory in the surface mixed layer in the global ocean based on the reconstructed 2 min longitude/latitude 137 Cs deposition data and surface mixed-layer depths in each box, and an estimate of the apparent half-residence time (Tap) in each box in the global ocean.
2. The behavior of 137 Cs in the surface mixed layer in the global ocean includes the following: a comparison between the amount of 137 Cs deposited from the atmosphere due to global fallout and that directly released from nuclear reprocessing plants, the behavior of 137 Cs released from nuclear reprocessing plants in the Atlantic to the Arctic, and an estimate of the amount of 137 Cs released from the F1NPS in 2011 by using the inventory in each box.

The mass balance includes the following:
an estimate of the 137 Cs amount in net outflow to the downstream box and downward transport below the surface mixed layer, and finally, a summary of the 137 Cs transport time in the global ocean.by the IAEA Marine Radioactivity Information System (MARIS) was combined.Furthermore, we used the 137 Cs data reported in the IRSN (Institut de Radioprotection et de Sûreté Nucléaire) database (Baily du Bois et al., 2020).Finally, all of these data were compiled into a single, comprehensive database for this study.
This new database contains a total of 56 447 data points corresponding to 137 Cs measured in surface waters (0-20 m) in the global ocean between 1956 and 2021.The 137 Cs concentration data points used in this study were significantly larger than those employed in our previous research, which included 22 368 data points obtained until 2005 (Inomata et al., 2009).Measured 137 Cs data, however, were very limited, and it is impossible to cover the distribution of 137 Cs in the global ocean.In this study, the global ocean was divided into 37 boxes to investigate the temporal variations in 137 Cs activity concentrations in surface seawaters by using almost all available data (Inomata and Aoyama, 2022a; Fig. 1).These boxes were divided by showing the latitudinal and longitudinal distributions based on the known ocean currents (IAEA, 2005;Open University, 2004), the latitudinal distributions of global fallout, location of the reprocessing plants, and F1NPS, under the assumption that 137 Cs activity concentrations in the box are almost same (Hirose and Aoyama, 2003a;Inomata et al., 2009;IAEA, 2005).Marginal seas are based on the definition by IHO (1953)  The 37 boxes set in this study are almost identical to those in our previous study (Inomata et al., 2009).However, the box areas have been slightly modified by taking into account the ocean currents.The region in the South China Sea, which   (Tsumune et al., 2011;Tsubono et al., 2016) and assessing the radiation doses delivered to humans through the ingestion of marine food (Aarkrog et al., 1997;IAEA, 2005;UNSCEAR, 2013).The 0.5-year median values of the surface 137 Cs concentrations in each box were produced by the grid value producing the command of the block median programs (Wessel et al., 2013).The block median reads the arbitrary data (x, y, and z) and calculates the median value in a grid defined in the setting of the range.In the case of t year, the data within t year ±0.5 years were used to calculate the median values.We produced the dataset (box number, year, and 137 Cs activity concentrations) for each box.These gridded data were recalculated to continuous curvature splines with adjustable tension, and these values are regarded as the 0.5-year median value.In the case of box 2, significantly higher concentrations were observed only near the F1NPS (box 2.1-2.6)because of the direct release of 137 Cs.We used only the 137 Cs activity data in box 2.0 for the analysis of the 0.5-year median values because the significantly higher values are localized and do not reflect concentrations throughout box 2.

Estimates of apparent half-residence times in global surface seawater
The 137 Cs activity concentrations are mainly dominated by fallout into the surface seawater, radioactive decay with a 30.17-yearhalf-life, horizontal transport, and downward transport below the mixed layer.Based on the long-term measurement of the 90 Sr deposition by the global monitoring network, the cumulative 90 Sr deposition reached a maxi-mum in the late 1960s to the early 1970s (UNSCEAR, 2000).This suggests that the atmospheric deposition of 137 Cs had a minor contribution to the surface seawater 137 Cs activity concentrations after 1970 (UNSCEAR, 2000;Hirose and Aoyama, 2003a;Aoyama et al., 2006).The decrease in 137 Cs activity concentrations in the surface seawater in each box is approximated by the corresponding fitting of the exponentially decreasing curves.This decreasing rate of 137 Cs activity concentrations is controlled by radionuclide decay and physical ocean circulation because the contribution from large-scale deposition by atmospheric nuclear tests was negligible after 1970 (UNSCEAR, 2000).The regression line of the 0.5-year median value of 137 Cs for each box was determined, and the apparent half-residence time (Tap) due to the radioactive decay and ocean physical processes was estimated.
The Tap of 137 Cs was calculated using the following equations: 137 Cs = 137 Csoexp(−λ Cs,apparent ) (1) λ Cs,apparent = λ Cs,ocean + λ Cs,decay (2) where λ Cs,apparent , λ Cs,ocean , and λ Cs,decay are the decay constants for apparent decay, physical oceanographic decay, and radioactive decay, respectively.λ Cs,apparent is estimated by using the regression line of the 0.5-year median value of 137 Cs, as shown in Eq. (1).Tpo is the apparent half-residence time causing the oceanic physical processes, and λ Cs,ocean was estimated as λ Cs,apparent and λ Cs,decay in Eq. (2).Considering that the half-life of 137 Cs (T 1/2 ) is 30.17years, the Tap should be shorter than the half-life if no source of 137 Cs exists in the region of interest.A shorter Tap means that 137 Cs is removed quickly in the area and/or the 137 Cs inflow amount is small in the area compared with the 137 Cs outflow amount.
In other words, a Tap shorter than the radioactive decay time indicates that the variations in the 137 Cs activity concentrations are strongly controlled by physical ocean processes.In contrast, a longer Tap and a negative Tpo value mean that 137 Cs is preserved in the region for a longer time and/or there is an influx of a water mass with higher 137 Cs in the region compared to the 137 Cs outflow from the region.However, the exponentially decreasing trend from 1970 to 2010, before the F1NPS accident, did not estimate values for all boxes.Tap values from 1970 to 2010 were estimated for the western North Pacific Ocean, Sea of Japan, and East China Sea.For other boxes, Tap, therefore, was estimated for several periods, taking into account the source contribution as follows, where Tap1 is before 1970 (periods with nuclear weapons tests at a global scale), Tap2 is the period from 1970 to 1986-1990 (until the Chernobyl accident), Tap3 is from 1990 to 2010 (after the Chernobyl accident), and Tap4 is after 2011 (after the F1NPS accident).There were some regions in which there were no Tap estimates, such as the northern North Atlantic and surrounding waters, because the decreasing trend of 137 Cs could not be approximated by Eq. (1).
2.4 F1NPS contribution to the 0.5-year 137 Cs median values in the North Pacific Ocean After several years following the F1NPS accident, 137 Cs activity concentrations in the surface seawater gradually increased in the Sea of Japan, East China Sea, subarctic North Pacific Ocean, and Bering Sea (Aoyama et al., 2016a(Aoyama et al., , 2017;;Inomata et al., 2018;Smith et al., 2017;Kumamoto et al., 2019).To estimate the contribution of the F1NPS accident, 137 Cs derived from the F1NPS accident ([F1NPS-137 Cs]) in 2011 was estimated using the following equation.The "[Global-137 Cs]each box" in 2011 was estimated by extrapolating the exponential regression line from 1990 to 2010 under the assumption that the apparent half-residence time of the 137 Cs activity concentrations were of the same value.

Reconstruction with the 2 min latitude/longitude 137 Cs deposition amount in the global ocean
To estimate the 137 Cs inventory in each box in the global ocean, we reconstructed the 137 Cs global fallout distribution with a 2 min latitude/longitude grid based on the 10 • latitude/longitude grid data of the 137 Cs deposition, which was constructed using 137 Cs data measured in rainwater, seawater, and soil by Aoyama et al. (2006).The topographybathymetry information was based on the 2 min Gridded Global Relief Data (ETOPO2; Earth Topography; NOAA National Geophysical Data Center, 2006), taking into account the ellipticity of the Earth (Oki and Kanae, 1997;Suga et al., 2013).

Estimate of the 137 Cs inventory in the surface mixed layer in the global ocean
We estimated the 137 Cs inventory in the surface mixed layer in each box and compared it with the 137 Cs fallout amount until 1970 because a major input of 137 Cs into the surface seawater occurred in the early 1960s and reached a maximum in 1970 (IAEA, 2005;Aoyama et al., 2006).The 137 Cs deposition in the surface seawater after 1970 was negligible compared to the activity concentrations in the surface seawater.Therefore, the behavior of 137 Cs in the surface seawater after 1970 is mainly controlled by radioactive decay and ocean physical transport processes.In this study, the 137 Cs inventory was estimated under the assumption that 137 Cs was mixed and homogeneously distributed within each box, with a 0.5-year timescale in the surface mixed layer.However, the Irish Sea (box 23) was subdivided into five regions (boxes 23.1-23.5)because significantly higher values were observed around the directly discharged area from the Sellafield plant.It is noted that there were no available data in box 23.5.In the northern North Atlantic Ocean, higher 137 Cs activity concentrations were observed in the region close to the Irish Sea or the North Sea.In this case, the northern North Earth Syst.Sci.Data, 15, 1969Data, 15, -2007Data, 15, , 2023 https://doi.org/10.5194/essd-15-1969-2023 Atlantic Ocean was also divided into two boxes (box 25.1 and box 25.2).Box 2 is also divided into seven subboxes because significantly higher concentrations were observed near the F1NPS.
The 137 Cs surface mixed layer inventory (unit is PBq) was estimated using the following equation, assuming that 137 Cs activity concentrations are almost constant in the surface mixed layer.
The sea area in each box was calculated using the basin mask assigned to each 2 min latitude/longitude square, which was created based on ETOPO2 (NOAA National Geophysical Data Center, 2006).The average mixed-layer depth in each box was calculated using the 2 • latitude/longitude gridded Mixed Layer Depth Climatology data compiled by the French Research Institute for the Exploitation of the Sea (IFREMER; Montégut, 2004;Mignot et al., 2007).These data were regridded as the 2 min latitude/longitude data to set the same scale as that in the sea area dataset.In this dataset, the mixed-layer depth was defined as the depth at which the surface temperature decreases by 0.2 • C and the density decreases by 0.03 kg m −3 .The mixed-layer depths were estimated by 780 000 profiles recorded in the World Ocean Database 09 (WOD09) National Oceanographic Data Center (NODC), using the conductivity, temperature, and depth (CTD) profile , the World Ocean Circulation Experiment (WOCE) version 3.0 Profiling Floats Data (PFL) CTD (1990-2002), and ARGO program (1995-2008).The mixed-layer depth was the monthly time interval with a seasonal variation that is deeper in winter and shallower in summer.It is recognized that seawater subducted from the ocean surface in the mode water formation region associated with winter convective mixing because of the lower buoyancy from the ocean surface (Hanawa and Talley, 2001).The flow through the winter mixed layer ventilates the seawater into the ocean interior.The maximum monthly mixed-layer depth in each box, mainly in the winter months, was used to calculate the 137 Cs inventory in the mixed layer.The mixedlayer depths used to estimate the inventory ranged from 33 to 182 m.In the western North Pacific Ocean (box 2), significantly higher 137 Cs values around the F1NPS cause the higher 0.5-year median value after 2011.Then, we used our previous estimated inventory by performing a statistical optical interpretation analysis in the subarctic North Pacific Ocean and western and eastern North Pacific Ocean (Inomata et al., 2016).
The 137 Cs density in the surface seawater (unit kBqm −2 ) was also estimated by using the following equation.In the marine environment, 137 Cs activity concentrations after 1970 were dominantly controlled by radioactive decay and physical ocean processes, except for the release by accidents and reprocessing plants.As for the physical oceanographic processes, 137 Cs in the surface seawater in each box receives inflow from the upstream box, outflow to the downstream box, and downward transport below the surface mixed layer.In fact, estimates of the 137 Cs transport amount in these processes were very difficult to make in this study.Therefore, outflowed 137 Cs by these processes were represented as net in-/outflowed 137 Cs in each box.The mass balance of 137 Cs in the surface mixed layer was considered to be as shown in the following equations.
[ 137 Cs inventory] box, t i is the value of the initial year, and [ 137 Cs inventory] box,t i +5 is the 137 Cs inventory after the t year in each box.In this study, this mass balance was estimated to be every 5 years from 1970 to 2015.In the case of 1970, the 137 Cs deposition amount until 1970 was used as the value of the initial year in each box.In the northern North Atlantic Ocean and Arctic Ocean, an extremely large inflow was estimated in 2000 due to the extremely large values included in the dataset.These data in 2000 and 2005 were removed from the figures.These results are described in Sect.4.4.

Results
Figure 2 shows the temporal variations in the 0.5-year median value of 137 Cs in each box.Table 2 lists the (2022c).In this section, temporal variations in the 0.5-year average 137 Cs values in each box are described.

North Pacific Ocean (boxes 1-3)
Figure 2a shows the temporal variations in the 0.5-year median 137 Cs values in the subarctic North Pacific Ocean (box 1), western North Pacific Ocean (box 2), and eastern North Pacific Ocean (box 3).The 137 Cs in this region largely originated from atmospheric deposition due to largescale weapons tests (e.g., Aoyama et al., 2006;Inomata et al., 2012), the Chernobyl accident in 1986 (Miyao et al., 1998), and the F1NPS accident after 2011 (e.g., Aoyama et al., 2016a, c;Inomata et al., 2016).The temporal variations in the 0.5-year median values of 137 Cs were the highest in the mid-and late 1950s.In the 1960s, the 0.5-year median values of 137 Cs increased gradually and reached a maximum in 1968.Then, the values decreased exponentially.The 0.5-year median values of 137 Cs after approximately 1970 became even smaller than those in the 1960s because the supply due to 137 Cs deposition was negligible.This implies that the variations in the 137 Cs activity concentrations after 1970 strongly depended on the physical processes in the ocean.Until the early 1960s, the 0.5-year median values of 137 Cs in the subarctic North Pacific Ocean and western North Pacific Ocean were higher than those in the eastern North Pacific Ocean.However, the 0.5-year median values of 137 Cs in the eastern North Pacific Ocean increased in the 1960s and were slightly higher than those in the western North Pacific Ocean in the 1970s.A small peak that occurred in 1986 was caused by the deposition of 137 Cs from the Chernobyl accident.The supply of 137 Cs due to Chernobyl fallout in the western and eastern North Pacific Ocean was larger than that in the subarctic North Pacific Ocean.After 2011, the 0.5year average 137 Cs values increased in this region.However, the maximum values in the eastern North Pacific Ocean were several years later than those observed in the subarctic and western North Pacific Ocean due to the basin-scale transport of 137 Cs in the North Pacific Ocean.In 1986, a small peak due to 137 Cs fallout from the Chernobyl accident was observed in the Sea of Japan and Sea of Okhotsk.In the 1980s, the 0.5-year median value of 137 Cs was also high in the South China Sea.After the 1990s, the 0.5-year median value of 137 Cs in the Sea of Okhotsk was smaller than those in the other boxes (the Sea of Japan, East China Sea, and South China Sea) in this region.An increase in the 0.5-year average 137 Cs values in 2011 occurred in the Sea of Japan, Sea of Okhotsk, and Bering Sea because of 137 Cs deposition originating from the F1NPS.In the Sea of Japan and East China Sea, the 0.5-year median values of 137 Cs gradually increased after the F1NPS accident.In addition, an increase in the 0.5-year median values of 137 Cs derived from the F1NPS occurred gradually in the Sea of Okhotsk and the Bering Sea.In the South Pacific Ocean (the western South Pacific Ocean (box 10), eastern South Pacific Ocean (box 11), eastern Southern Ocean (box 12), and middle Southern Ocean (box 19); as shown in Fig. 2d), the 0.5-year median 137 Cs activity concentrations in 1961 ranged from 1.4 to 2.5 Bq m −3 , whereas in 1967, the 0.5-year median of 137 Cs increased to 4.5-9.9Bq m −3 .Afterwards, the 0.5-year median value of 137 Cs decreased exponentially, although the available data were limited.Moreover, substantially lower 0.5-year median 137 Cs values of less than 1 Bq m −3 were observed in the middle Southern Ocean in 2013.

Subtropical and equatorial
3.5 Antarctic Ocean (boxes 13, 36, and 37) The 0.5-year median 137 Cs activity concentrations in the Antarctic Ocean were the lowest in the global ocean, although the measurement data were very limited (Fig. 2e).The 0.5-year median 137  ).The decreasing rate in the Antarctic region was larger than those in the other regions.These lowest 0.5-year median 137 Cs values and larger decreasing rates were due to the long distance from the dominant 137 Cs fallout area.The transport of 137 Cs from the Southern Ocean would be prevented by the Antarctic Circumpolar Current.The upwelling of seawater from the deeper layers in the Antarctic Ocean may also have caused the dilution of the 137 Cs activity concentrations, resulting in the lowest 137 Cs values (Kumamoto et al., 2016).
3.6 Indian Ocean (boxes 15, 16, and 17) In the 1960s, the 0.5-year median 137 Cs activity concentration was higher in the Arabian Sea (box 15; 8.9 Bq m −3 ) than in the Indian Ocean (box 16; 5.6-6.6 Bq m −3 ) and the Southern Ocean (box 17; 2.7-4.0Bq m −3 ; Fig. 2f).The median 137   In these regions, the 0.5-year median values of 137 Cs did not decrease exponentially due to the 137 Cs being discharged from the nuclear fuel reprocessing plants.In the mid-1970s, the 0.5-year median 137 Cs activity concentrations in the North Sea and northern North Atlantic Ocean increased rapidly and reached a the maximum in the years 1977 and 1980.5, respectively.Thereafter, the 0.5-year median 137 Cs activity concentrations in these regions rapidly decreased, and the decreasing rate became small after 1990, which was associated with the reduced amount of released 137 Cs.In the Arctic Ocean, the 0.5-year median 137 Cs activity concentrations increased until the mid-1970s and then decreased.The overall 0.5-year median 137 Cs activity concentrations until the 1970s in the Arctic Ocean were lower than those in the surrounding oceans.
3.9 Irish Sea and English Channel (boxes 23 and 24) The 137 Cs activity concentrations in the Irish Sea (box 23) and English Channel (box 24) were primarily affected by the discharge from the nuclear fuel reprocessing plants (Fig. 2i).Because the 0.5-year median value of 137 Cs is significantly higher than those in other regions, the scale of the y axis changes from 10 −1 to 10 6 Bq m −3 .In the Irish Sea (box 23.1), which is the discharge region, the 0.5year median 137 Cs activity concentrations increased rapidly and reached 38 533 Bq m −3 in 1976 and then decreased rapidly.The 0.5-year median 137 Cs activity concentration decreased with increasing distance from the discharge region.In box 23.2, the maximum value (2453 Bq m −3 ) was observed in 1977.The 137 Cs activity concentrations in the Irish Sea (box 23.4) were the lowest in this box.The decreasing gradient of the 0.5-year median 137 Cs values reflected the controlled discharge amount.
In the English Channel, the 0.5-year median 137 Cs activity concentrations reached a maximum in 1980.5, and these also decreased over time.The 0.5-year median 137 Cs activity concentrations in 2017 in the English Channel were 1.9 Bq m −3 ., which was at least 18 times higher than that before the Chernobyl accident (Fig. 2j).The 0.5-year median value of 137 Cs decreased rapidly to 60 Bq m −3 in 1989.The 0.5-year median 137 Cs value in 2002 was almost equal (18.3 Bq m −3 ) to that before the Chernobyl accident.The rapid decrease in surface 137 Cs could be due to the strong intrusion of surface waters to the deep layers, 137 Cs inflow into the Mediterranean Sea after passing through the Bosporus Strait, and radioactive decay (Egorov et al., 1999;Delfanti et al., 2014).However, the Black Sea continues to receive 137 Cs derived from Chernobyl by river runoff (Gulin et al., 2013).
In the Mediterranean Sea (box 27), the 0.5-year median value of 137 Cs varied from 0.8 to 12 Bq m −3 before the Chernobyl accident (Fig. 2j).After the accident, it increased to 142 and 155 Bq m −3 in 1986.5 and 1987.5, respectively.In the following years, the 0.5-year median value of 137 Cs decreased rapidly and became almost the same as that before the Chernobyl accident.This rapid decrease could have been due to the stronger intrusion of bottom water from the surface (Delfanti et al., 2000;Delfanti and Papucci, 2010).3. The negative Tpo suggests that the physical oceanographic processes were controlled by the variation in 137 Cs activity concentrations.Figure 5 shows the horizontal distributions of the 137 Cs deposition density in each box in the global ocean.These values are also listed in Table 4.The 137 Cs deposition density is high in the midlatitude region in the North Pacific Ocean (the Sea of Japan, subarctic North Pacific Ocean, Sea of Okhotsk, western North Pacific Ocean, and eastern North Pacific Ocean) and the northern North Atlantic Ocean.In the North Pacific Ocean, these regions correspond to the area in which the Kuroshio Current and Kuroshio Extension are transported.In the northern North Atlantic Ocean, the higher 137 Cs deposition area influences the Gulf Stream flow.The dominant features of these regions in the North Pacific Ocean and northern North Atlantic Ocean have received larger precipitation amounts, and stratosphere-troposphere air mass exchange has occurred (Aoyama et al., 2006).The larger air mass exchange between the stratosphere and troposphere means that the 137 Cs injected into the stratosphere by the large-scale weapons tests is transported into the troposphere and deposited on the surface by precipitation.South of 5 • N, the 137 Cs deposition density is lower than that in the northern region, and there is no significant difference between the open oceans (Pacific, Atlantic, and Indian oceans).The distribution of the 137 Cs deposition a 2 min latitude/longitude grid resolution is also well reproduced in the 10 • latitude/longitude grid deposition.
However, these estimations were almost 1.4 times larger than those in the estimation by using a model simulation (UNSCEAR, 1993), with an estimated value of 545 PBq (Aoyama, 2019).The large difference occurred in the meridional distribution in the midlatitude.This corresponds to having a larger 137 Cs fallout region that includes the Kuroshio Current and its extension areas (latitude 20-40 • N) in the Pacific Ocean and the Gulf Stream transport area (latitude 30-50 • N) in the Atlantic Ocean.It was also reported that the 137 Cs water column inventory in the North Pacific Ocean was 2-3 times larger than those in the cumulative 137 Cs fallout amount in the same latitude in the modeling results in UN-SCEAR (1993;Aoyama, 2019).Because the reconstructed 137 Cs deposition in Aoyama et al. (2006)   Figure 6 shows the horizontal distribution of the 137 Cs deposition amount as of 1 January 1970 in the surface mixed layer in the global ocean.These data are also listed in Table 4.In the Pacific Ocean, a higher 137 Cs deposition amount occurs in the subarctic North Pacific Ocean (71.6 PBq), western North Pacific Ocean (40.8 PBq), eastern North Pacific Ocean (52.4 PBq), and eastern subtropical North Pacific Ocean (47.9 PBq).In the Atlantic Ocean, a higher 137 Cs deposition amount is found in the northern North Atlantic Ocean (sum of boxes 25.1 and 25.2; 56.1 PBq) and the North Atlantic Ocean (69.8 PBq).The 137 Cs deposition amount in the Atlantic Ocean shows a significant latitudinal gradient, which is 27.3 PBq for the central Atlantic Ocean and 5.1 PBq for the South Atlantic Ocean.In the Indian Ocean, the 137 Cs deposition amount also has a north-south gradient.The 137 Cs deposition amount is the lowest in the Pacific sector (0.05 PBq),   ).In 1975, the 0.5-year median 137 Cs values in the South China Sea were higher than those in the Indonesian Archipelago (Fig. 7b).In 1980 and 1985 (Fig. 7c and d), in the Pacific Ocean, higher values were found in the eastern subtropical Pacific Ocean (9.8 Bq m −3 ).
In the North Pacific Ocean, the 0.5-year median 137 Cs values were higher in the eastern region, whereas higher values in the South Pacific Ocean were found in the western region.In 1990, the 0.5-year median 137 Cs values decreased, although the concentration distribution was similar to that of the 1980s (Fig. 7e).In particular, after 1990, the Indonesian Archipelago became the hotspot region, with relatively high 0.5-year median 137 Cs activity values in the Pacific Ocean and Indian Ocean.Higher values in this region were still observed in 2015 (Fig. 7j).higher in the North Pacific Ocean and lower in the South Pacific Ocean, became small, which lasted until 2010.In 2015, an increase in the 0.5-year median 137 Cs values was found in the western and subarctic North Pacific Ocean due to the release of 137 Cs from the F1NPS accident (Fig. 7j).The lowest value occurred in the Antarctic Ocean in the global ocean after 1970 (Fig. 7).
In the Atlantic Ocean, a latitudinal gradient that was higher in the northern North Atlantic Ocean and North Atlantic Ocean (north of 30 • N) than in the Central and South Atlantic Ocean occurs.In 2015, the values in the South Atlantic Ocean (0.7 Bq m −3 ) were almost equal to those in the Southern Ocean (0.6 Bq m −3 ; Fig. 7j).In the Atlantic Ocean, because of the discharged 137 Cs in the surface seawater from the nuclear reprocessing plants, i.e., the Sellafield plant, significantly higher 0.5-year 137 Cs median values occurred in the Irish Sea, particularly at 137 Cs discharge points (box 23.1), as shown in Figs. 8 and 9.The discharged 137 Cs was transported to the northern North Pacific Ocean (box 25.1) and North Sea from the Irish Sea (box 23.2).The 0.5-year median 137 Cs values after 1985 (Fig. 8d) decreased gradually in accordance with the discharged amount of 137 Cs (Fig. 9c).The increase in the 0.5-year median 137 Cs values in the Baltic Sea (96 Bq m −3 ) and Black Sea (56 Bq m −3 ) in 1990 was caused by the deposition of 137 Cs from the Chernobyl accident (Fig. 8e and f).Contamination due to the Chernobyl accident continued in the Baltic Sea and Black Sea until 2015 (Fig. 8j).
3.14 137 Cs inventory in the surface mixing layer in the global ocean after 1970 The horizontal distribution of the surface mixed-layer depth in the global ocean is shown in Fig. 10.The mixed-layer depth in the open ocean shows a clear latitudinal distribution that is deeper (∼ 182 m) in the higher latitudes and shallower in the lower latitudes, particularly in the equatorial Pacific Ocean (48 m for the eastern equatorial Pacific Ocean and 58 m for the western equatorial Pacific Ocean).In the coastal sea, the mixed-layer depths are shallower (33-76 m) than those in the open ocean.The mixed-layer depth in each box is also listed in Table 5.By using these mixed-layer depths, the estimated 137 Cs inventory in the surface seawater from 1970 to 2015 for every 5 years is shown in Figs.11-13.The 137 Cs inventory in the surface seawater is also listed in Table 5.In the Pacific Ocean, a higher 137 Cs inventory exists in the subarctic North Pacific Ocean (10.4 PBq), western and eastern North Pacific Ocean (6.3 and 12.2 PBq), and western and eastern subtropical North Pacific Ocean (7.1 and 11.6 PBq) in 1970 (Fig. 11a).In particular, the 137 Cs inventory in the eastern North Pacific Ocean/subtropical eastern North Pacific Ocean is larger than that in the western regions.In the South Pacific Ocean, a higher 137 Cs inventory is observed in the western subtropical South Pacific Ocean (7.2 PBq) and eastern Southern Ocean (10.7 PBq) in 1970.After 1975, the 137 Cs inventory in the eastern part is larger than that in the western part in the South Pacific Ocean.The 137 Cs inventory in the surface seawater in the Southern Hemisphere is the highest in the Pacific Ocean, followed by the Indian Ocean, and it is the lowest in the Atlantic Ocean.In the Indian Ocean, the 137 Cs inventory has a latitudinal gradient, which is higher in the Southern Ocean (15.2 PBq) and lower in the Arabian Sea in the year 1970.These latitudinal gradients of the 137 Cs inventory in the surface mixed layer continued until 2015.
In the northern North Atlantic Ocean, its marginal seas, and the Arctic Ocean (Figs. 12, 13), the 137 Cs inventory is strongly influenced by the discharged 137 Cs from fuel reprocessing plants in the Irish Sea and English Channel after 1970, in addition to the global fallout from large-scale weapons tests in the 1960s and 1960s (northern North Atlantic Ocean.2 and Arctic Ocean).In the Irish Sea, the maximum 137 Cs inventory occurred in the Irish Sea.1 (15.2 PBq) in 1975.The 137 Cs discharged into the Irish Sea.1 was transported into the Irish Sea.2 (3.3 PBq in 1980), followed by transport to the northern North Atlantic Ocean.1 (15.2 PBq in 1980), North Sea (7.5 PBq in 1980), and Barents Sea and coast of Norway (6.8 PBq in 1980).This pattern is also consistent with general pattern of seawater transport in this region (Prandle, 1984;Prandle and Beechey, 1991;Baily du Bois et al., 2020).The 137 Cs discharged from La Hague was also transported into the North Sea.
Figure 14 3, Tap2 in the eastern North Pacific Ocean, which is estimated to be 8.8 years, is shorter than that in the western North Pacific Ocean (16.9 years).This suggests that the outflowed 137 Cs amount in the eastern North Pacific Ocean was larger than the inflowed 137 Cs amount from the western North Pacific Ocean.The Tap2 value in the western subtropical North Pacific Ocean is estimated to be 34.1 years, and Tpo is estimated to be −260.7 years.This mean that the 137 Cs accumu-   (Gordon, 2005;Feng et al., 2018).According to reports, the median volume of seawater in the Indonesian  Throughflow was estimated to be 15 Sv (1 Sv = 10 6 m 3 s −1 ; e.g., Gordon et al., 2010;Feng et al., 2018).The Indonesian Throughflow consists of seawater derived from the North Pacific Ocean, South Pacific Ocean, and Antarctic Ocean.Approximately 9 Sv of seawater is transported to the Indian Ocean (Gordon, 2005).The transport of seawater from the Pacific Ocean into the Indian Ocean is revealed by the higher tritium concentrations in the seawater transported from the North Pacific Ocean through the Makassar Strait with the Mindanao Current, whereas seawater with lower tritium concentrations is transported into the Indian Ocean from the South Pacific Ocean via the Halmahera Sea (Fin et al., 1994).This suggests that most of the 137 Cs in the surface seawater inflows into the Indian Ocean from the North Pacific Ocean.
Furthermore, the South Indian Ocean is connected to the Atlantic Ocean around the Cape of Good Hope via the Agulhas Current (Sanchez-Cabeza et al., 2011), with an median seawater mass transport of 8.7 Sv, according to Stramma and England (1999).Although we used the median box value in this discussion, the signatures of 137 Cs inflow from the North Pacific Ocean to the Indian Ocean via the Indonesian Archipelago were recognized by measurements.Evidently, higher concentrations were found at approximately 100 • E in the subsurface layer, whereas lower concentrations were observed at approximately 70 • E in the surface seawater (Povinec et al., 2011).In addition, 137 Cs in the Indian Ocean is transported westwards at approximately 10-15 • S latitude (Sanchez-Cabeza et al., 2011;Povinec et al., 2011).In this section, we discuss the 137 Cs inflow from the Pacific Ocean to the Indian Ocean, followed by the Atlantic Ocean.In the Indonesian Archipelago, the 0.5-year median value of 137 Cs in 2010 was 2.7 Bq m −3 (Table 2; Fig. 7).This value is higher than those in the surrounding sea area, such as in the East China Sea (1.7 Bq m −3 ), western North Pacific Ocean (1.5 Bq m −3 ), western subtropical North Pacific Ocean (1.7 Bqm −3 ), and western equatorial Pacific Ocean (1.2 Bq m −3 ).The 0.5-year median value in the Indonesian Archipelago in 2010 decreased to approximately 53 % compared to that in 1970.These decreasing rates are smaller than those in the surrounding sea area (78 %-93 %).
Figure 16 shows the spatiotemporal variations in the 137 Cs density in the regions related to the Indonesian Throughflow (the South China Sea, Indonesian Archipelago, western subtropical South Pacific Ocean, Arabian Sea, and Indian Ocean).The 137 Cs density in each box in the global surface seawater is also listed in Table 6.The 137 Cs density in the western subtropical South Pacific Ocean was almost the same as in the Indonesian Archipelago.This result suggests that the 137 Cs derived from large-scale weapons tests in the western North Pacific Ocean flowed into the Indian Ocean by basin-scale transport.The 137 Cs densities in the Indian Ocean (the Arabian Sea, Indian Ocean, and Southern Ocean) were almost similar to or higher than those in the Indonesian Archipelago.Furthermore, the 137 Cs density in these regions decreased after the 1990s.The Tap in the Indonesian Archipelago was estimated to be 36.7 years, and Tpo was estimated to be −169.2years.It is likely that the main plume of the 137 Cs derived from the large-scale weapons tests is transported into the Indian Ocean, with a timescale of 20-30 years.
Furthermore, several studies have found that 137 Cs was transported into the South Atlantic Ocean via the Agulhas Current and then transported northwards with the Benguela Current (Sanchez-Cabeza et al., 2011;Stramma and England, 1999).The transit times from the Pacific Ocean to the Atlantic Ocean via the Indian Ocean were estimated over 4 decades via model simulations (Tsumune et al., 2011).In this study, a slight increase in the 137    atmospheric deposition of 137 Cs derived from the F1NPS accident.
After 2013, in the western North Pacific Ocean, the 0.5year median 137 Cs values decreased exponentially with seasonal variation, which was higher in summer and lower in winter.In the eastern North Pacific Ocean, the 0.5-year median 137 Cs values increased after 2014 and reached a maximum in 2018.It is clear that the 0.5-year median 137 Cs values in the eastern North Pacific Ocean were higher than those in the western region in 2018/2019.With the optimal interpretation analysis in Inomata et al. (2016), the main plume of F1NPS-137 Cs exists in the center of the North Pacific Ocean (longitude range of 165 • E-170 • W and latitude range of 30-50 • N), which corresponds to the subarctic, western, and eastern North Pacific Ocean boxes in this study in 2012.The zonal transport median speed of F1NPS-137 Cs was estimated to be approximately 8 cm s −1 from March 2011 to March 2012 (Aoyama et al., 2013).The arrival of the F1NPSderived radio caesium was detected in June 2013 in the Canadian continental shelf, and these detections continued until February 2014 (Smith et al., 2017).The 137 Cs activity concentrations reached 2 Bq m −3 , which is 2 times larger than those in the nuclear-weapons-derived 137    A slight increase in the 0.5-year median value of 137 Cs also occurred in the East China Sea and Sea of Japan after 2013 and reached maximum values in 2015/2016 (Fig. 17b).The 137 Cs activity concentrations in the East China Sea and Sea of Japan increased, following the processes elucidated in our previous studies (Inomata et al., 2018).The increase in 137 Cs activity concentrations was first observed in the subsurface layer in 2012/2013 around southern Japan in the western North Pacific Ocean.Based on the potential temperature density (σ θ ), the 137 Cs peak existed in the subtropical mode water.In the East China Sea, the increase in the 137 Cs activity concentrations also started in subsurface seawater (140 m) in 2013, and 137 Cs activity in the surface  mixed layer (0-50 m) in the East China Sea reached a maximum in 2014/2015 (Inomata et al., 2018).Increased 137 Cs in the East China Sea is caused by the following processes.The 137 Cs entrained into the subtropical-mode water is transported westwards in the subsurface seawater and has upwells along the continental shelf in the East China Sea and the Kuroshio Counter Current around the meandering Kuroshio (Ito et al., 1994).Furthermore, the 134 Cs/ 137 Cs ratios in subtropical mode water were almost the same as those in seawater in the East China Sea and Sea of Japan (Aoyama et al., 2017;Inomata et al., 2018).Then, the F1NPS-derived 137 Cs flowed into the Sea of Japan via the Tsushima Strait by the Tsushima Warm Current and reached a maximum in 2015/2016.The propagation of F1NPS-derived 137 Cs from the East China Sea to the Sea of Japan occurred over 1-2 years.

Estimation of the net in-or outflow amount of 137 Cs
The 137 Cs deposited into the ocean surface is transported via advection and diffusion in the surface seawater, in addition to being transported to deep water below the surface mixed-layer depth, where it undergoes radioactive decay (T 1/2 = 30.17years).On 1 January 1970, 137 Cs existing in the surface seawater in the global ocean was estimated to be 187 ± 26 PBq.This value corresponds to 32 % of the deposited 137 Cs until 1970, although the 137 Cs released from reprocessing plants is included.The remaining approximately 68 % of deposited 137 Cs, which is estimated to be 577 ± 60 PBq, would be transported downwards below the surface mixed layer in the global ocean on a decadal timescale.According to the estimation by using a model simulation (Kamidaira et al., 2018), the amount of F1NPS-137 Cs transported below the surface mixed layer within 4 months is almost half.
By using the mass balance equations described in Sect.2.7, the net in-or outflowed 137 Cs amounts in each box are shown in Figs.18-21.Figure 18 shows the horizontal distributions of the net in-/outflowed 137 Cs amount in the surface mixed layer in 1970 against the 137 Cs deposition amount until 1 January 1970 (Fig. 5).Positive values (red) indicate that the net inflowed 137 Cs amount is larger, whereas negative values (blue) indicate that the net outflowed 137 Cs amount is larger in each box.The decrease in 137 Cs occurred in the North Pacific Ocean.The largest decrease, which was estimated to be 61.2PBq, occurred in the subarctic North Pacific Ocean.On the other hand, an increase in the 137 Cs amount occurred in the western subtropical South Pacific Ocean (2.8 PBq), eastern South Pacific Ocean (9.6 PBq), middle Southern Ocean (2.6 PBq), Southern Ocean (12.4 PBq), and Antarctic Ocean (1.2-4.0PBq).This suggests that some of the 137 Cs deposited into the Pacific Ocean was transported to the South Pacific Ocean (south of 40 • N), followed by movement to the Indian Ocean within 10-20 years.The outflow of 137 Cs in the northern North At-lantic Ocean and North Atlantic Ocean was also large (41.4 and 57.1 PBq, respectively).The increase in 137 Cs in the Irish Sea (Irish Sea.1; 2.3 PBq), North Sea (0.7 PBq), and northern North Atlantic Ocean (0.5 PBq) was due to the discharged 137 Cs from the Sellafield and La Hague plants (Fig. 18).
The temporal variation in the net in-/outflowed pattern of the 137 Cs amount is shown in Figs.19-21.Figure 19 shows the net in-/outflowed 137 Cs amount in each area in the global ocean from 1975 to 2015 at 5-year intervals.The net in-or outflowed 137 Cs amount corresponds to the sum of the 137 Cs amount for the previous 5 years.In 1975In , 1980In , and 1985, the values in the subarctic, western, and eastern North Pacific Ocean were negative (−0.5-2.7,−0.4-1.8, and −0.03-1.2PBq, respectively), suggesting that the net outflowed 137 Cs amount is larger than the net inflowed 137 Cs amount.On the other hand, the net in-/outflowed 137   19d).This suggests that the main 137 Cs plume derived from the large-scale nuclear weapons tests would pass through until 1990.It is also noted that a small amount of 137 Cs inflowed into the South Atlantic Ocean in 1975 (0.1 PBq) and 1980 (0.01 PBq; Fig. 19a and  b).A small 137 Cs increase also occurred in the central Atlantic Ocean after 1995 (Fig. 19c).In 2015, increased 137 Cs in the subarctic, western, and eastern North Pacific Ocean (0.02-1.2 PBq) would be caused by the 137 Cs released from the F1NPS (Fig. 19i).
In the northern North Atlantic Ocean and its marginal seas (Fig. 20), the increase in the 137 Cs amount due to the discharged 137   ents Sea and coast of Norway and transported to the Arctic Ocean (Fig. 20a-d).The contribution of the discharged 137 Cs from reprocessing plants decreased after 1985.The contribution from the Chernobyl accident was still found in the Baltic Sea, Black Sea, and Mediterranean Sea in 1990 (Fig. 20d).
In the Irish Sea and English Channel, the contributions of 137 Cs released from reprocessing plants were large in 1975, and it appears that this 137 Cs was transported to the northern North Atlantic Ocean and North Sea until 1980.In 1980, the 137 Cs amount around the source region, the Irish Sea.1, showed negative values.After 1985, the contribution from the reprocessing plants was negatively associated with the decreased discharged 137 Cs amount (Fig. 21).and transported to the North Sea, Barents Sea and coast of Norway, and the Arctic Ocean over approximately a 1-2 decadal timescale.

Conclusions
In this study, we analyzed the 137 Cs activity concentrations in the surface seawater in the global ocean by using almost all of the available historical data in the global ocean.The surface seawater was divided into 37 boxes, and the temporal variations in the 0.5-year median 137 Cs values in each box were investigated to determine the 137 Cs distribution and transport in the global surface seawater.
The 137 Cs deposition as of 1 January 1970, with 2 × 2 min resolution, is estimated to be 874 ± 90 PBq.In 1970, due to the minor contribution of atmospheric deposition, the 137 Cs inventory in the surface mixed layer in the global ocean was estimated to be 184 ± 26 PBq.This suggests that 68 % of the 137 Cs deposited into the surface seawater in the global ocean (577 ± 60 PBq) had already been transported below the surface mixed layer on a decadal timescale.Author contributions.YI (the corresponding author) conducted the data analysis and prepared the paper.MA developed the database of radioactivity.Both authors discussed the results of the data analysis.
Competing interests.The contact author has declared that neither of the authors has any competing interests.
Disclaimer.Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements.The authors thank Pascal Bailly du Bois, who provided the 137 Cs data measured in the northern North Atlantic Ocean and its marginal seas.
Review statement.This paper was edited by Dagmar Hainbucher and reviewed by Fabio Conte and one anonymous referee.
. The subarctic North Pacific Ocean (box 1; 40 • N) is the highest atmospheric deposition of 137 Cs that occurred in the 1960s in the Pacific Ocean, western North Pacific Ocean, and eastern North Pacific Ocean (boxes 2 and 3), which are located at 25-40 • N and are upstream and downstream of the Kuroshio Exten-sion.These three regions are influenced by the 137 Cs contamination derived from the F1NPS accident.The subtropical western and eastern North Pacific Ocean (boxes 4 and 5) are downstream and upstream of the North Equatorial Current associated with the subtropical gyre.The subtropical western and eastern North Pacific Ocean includes the California gyre (5-25 • N).These boxes include the contamination of local fallout, such as the Bikini Atoll.The western and eastern equatorial Pacific Ocean (boxes 6 and 7) are downstream and upstream of the South Equatorial Current.And an upwelling of seawater occurs in the eastern Southern Pacific Ocean.The western and eastern subtropical North Pacific Ocean (boxes 8 and 9) currents are downstream and upstream of the weak South Equatorial Current.The eastern subtropical South Pacific Ocean includes the French nuclear weapons test sites.The western Southern Pacific Ocean (25-40 • S) is the Tasman Sea (box 10).The eastern South Pacific Ocean (25-40 • S) is the midlatitude region of the South Pacific Ocean and includes the South Pacific current (box 11).The eastern Southern Ocean (40-60 • S) is affected by the Antarctic Circumpolar Current (box 12).The Antarctic Ocean (below 60 • S) is divided into three, namely the Antarctic sector for the Pacific (box 13), Indian (box 36), and Atlantic (box 37) oceans, and locates the polar front and continental water boundary.In the middle Southern Ocean (box 19), the Leeuwin Current, which flows southward along the continental shelf off the western Australia coast (22 to 35 • S) in the Southern Ocean (box 17), transports the current to the east.The Southern Ocean is characterized as a subtropical gyre and is connected with the Antarctic Circumpolar Current from the Atlantic Ocean side in the upstream and the Pacific Ocean side in the downstream.The Antarctic Circumpolar Current is a structure banded with several currents and forms a narrow jet with a sharp front.The Indian Ocean (box 16) is connected to the Indonesian Archipelago (box 35) by the Indonesian Throughflow from the Pacific Ocean.The Arabian Sea (box 15; 10 • S-30 • N) is affected by the equatorial current system, and the circulation in the northern Indian Ocean is associated with the monsoon system.The marginal seas of the North Pacific Ocean are classified as the South China Sea (box 33), East China Sea (box 32), Sea of Japan (box 14), and Sea of Okhotsk (box 31).The East China Sea has an influence on the bifurcation of the Kuroshio Current and is downstream of the western North Pacific Ocean and connected to the Sea of Japan via the Tsushima Warm Current.The northward-transported seawater in the Sea of Japan is connected to the Sea of Okhotsk.The Bering Sea (box 34) is downstream of the subarctic North Pacific Ocean and upstream of the Arctic Ocean.The Atlantic Ocean was divided into three, namely the South Atlantic Ocean (box 28; 60-30 • S), central Atlantic Ocean (box 29; 30 • S-15 • N), and North Atlantic Ocean (box 30; 15-45 • N).The South Atlantic Ocean is connected with the Southern Ocean via the Agulhas Current.The Irish Sea (box 23) and English Channel (box 24) are considered to be the 137 Cs direct discharge

Figure 1 .
Figure 1.Boxes dividing the global ocean.(a) Global ocean, (b) North Atlantic Ocean and its marginal sea, and (c) the Irish Sea (box 23) and English Channel (box 22).
was in box 33 in Inomata et al. (2009), is divided into two boxes, namely the South China Sea (box 33) and Indonesian Archipelago (box 35), because the seawater in the Pacific Ocean is transported into the Indian Ocean via the Indonesian Archipelago.The region in Antarctica, which was in box 13 in Inomata et al. (2009), is divided into three boxes, including the Pacific sector of the Antarctic Ocean (box 13), the Atlantic sector of the Antarctic Ocean (box 36), and the Indian sector of the Antarctic Ocean (box 37).The region in the Southern Ocean (box 17) is divided into two boxes, with the Southern Ocean (box 17) and the middle Southern Ocean (box 19).The boxes corresponding to the source region, such as the Irish Sea (box 23; boxes 23.1-23.5)for the Sellafield plant and the northern North Atlantic Ocean https://doi.org/10.

[
137 Csdensity in the surface seawater]box = [ 137 Cs inventory]box/[sea area]box (7) 2.7 Mass balance -net in-/outflow of 137 Cs in each box Figure2shows the temporal variations in the 0.5-year median value of 137 Cs in each box.Table 2 lists the 0.5-year median values of the surface 137 Cs in the global ocean from 1960 to 2020 every 5 years.The 0.5-year median values of

Figure 2 .
Figure 2. The 0.5-year median 137 Cs values of each box for (a) boxes 1-3 (subarctic North Pacific Ocean, western North Pacific Ocean, and eastern North Pacific Ocean), (b) boxes 14 and 31-34 (Sea of Japan, Sea of Okhotsk, East China Sea, South China Sea, and Bering Sea), (c) boxes 4-9 (western subtropical North Pacific Ocean, eastern subtropical North Pacific Ocean, western equatorial Pacific Ocean, eastern equatorial Pacific Ocean, eastern subtropical South Pacific Ocean, and western subtropical South Pacific Ocean), (d) boxes 10-12 and 19 (western subtropical South Pacific Ocean, eastern subtropical South Pacific Ocean, eastern Southern Ocean, and middle Southern Ocean), (e) boxes 13, 36, and 37 (Pacific sector of the Antarctic Ocean, Atlantic sector of the Antarctic Ocean, and Indian sector of the Antarctic Ocean), (f) boxes 15-17 (Arabian Sea, Indian Ocean, and Southern Ocean), (g) boxes 28-30 (North Atlantic Ocean, central Atlantic Ocean, and South Atlantic Ocean), (h) boxes 18, 20, 22, and 25 (Arctic Ocean, Barents Sea and coast of Norway, North Sea, and northern North Atlantic Ocean), (i) boxes 23 and 24 (Irish Sea and English Channel), and (j) boxes 21, 26, and 27 (Baltic Sea, Black Sea, and Mediterranean Sea).

3. 2
Marginal seas of the western North Pacific Ocean (boxes 14 and 31-34) The temporal variations in the 0.5-year 137 Cs values in the marginal seas of the North Pacific Ocean (the Sea of Japan (box 14), Sea of Okhotsk (box 31), East China Sea (box 32), South China Sea (box 33), and Bering Sea (box 34)) are displayed in Fig. 2b.The 0.5-year 137 Cs values in the marginal seas of the North Pacific Ocean also decreased exponentially.
Pacific Ocean and Indonesian Archipelago (boxes 4-9 and 35) The 0.5-year median values of 137 Cs in the western subtropical North Pacific Ocean (box 4) in the late 1950s were significantly high (62-73 Bq m −3 ) due to local fallout (Fig. 2c).In the 1970s and 1980s, the 0.5-year median values of 137 Cs almost constantly varied in the eastern subtropical North Pacific Ocean (box 9) and were higher than those in other regions.The values in the western subtropical North Pacific Ocean and western equatorial Pacific Ocean were almost constant in the 1970s and 1980s.In the eastern equatorial Pacific Ocean (box 7) and eastern subtropical South Pacific Ocean (box 9), the values increased gradually.After the mid-1980s or in the 1990s, the 0.5-year median 137 Cs values showed an exponential decrease until 2011.The 0.5-year median values of 137 Cs in the Indonesian Archipelago (box 35) showed almost the same range as those in the western subtropical and equatorial Pacific Ocean.3.4 South Pacific Ocean (boxes 10-12 and 19) Cs activity concentrations from the 1960s to the 2010s decreased from 3.3 to 0.01 Bq m −3 in the Pacific sector of the Antarctic Ocean (box 13) and from 1.6 to 0.4 Bq m −3 in the Atlantic sector of the Antarctic Ocean (box 36).In the Indian sector of the Antarctic Ocean, the 0.5-Earth Syst.Sci.Data, 15, 1969Data, 15,  -2007Data, 15,  , 2023     https://doi.org/10.5194/essd-15-1969-2023year median 137 Cs activity concentrations decreased from 5.0 to 1.1 Bq m −3 during the period from 1961.5 to 1978 (box 37 Cs activity concentrations in the 1960s showed a latitudinal gradient, with higher values in the northern areas and lower values in the southern areas.The 0.5-year median 137 Cs activity concentrations in the Arabian Sea and Southern Ocean were almost constant in the 1970s, whereas those in the Indian Ocean increased slightly (3.4-5.6 Bq m −3 ) in the 1970s.These values in the three boxes decreased to approximately 1.3-2.3Bq m −3 in the late 1990s and the early 2000s, although there were no available data in the 1980s.The 0.5-year median 137 Cs activity concentrations in the Indian Ocean were higher than those in the Arabian Sea and the Southern Ocean.Note that the lowest values (0.15 Bq m −3 ) were observed in 2012 in the Arabian Sea.3.7 Atlantic Ocean (boxes 28-30)The Atlantic Ocean (the North Atlantic Ocean in box 28, the central Atlantic Ocean in box 29, and the South Atlantic Ocean in box 30) had a north-south gradient of the 0.5-year median 137 Cs values, with higher values in the North Atlantic Ocean and lower values in the South Atlantic Ocean (Fig.2g).In the North Atlantic Ocean, relatively high values were observed in the 1970s but then rapidly decreased after 1980.After 2000, an exponentially decreasing trend was not observed in the North Atlantic Ocean.The values increased slightly and reached the maximum value (2.4 Bq m −3 ) in 2005, after which they gradually decreased.In the central Atlantic Ocean, the temporal variations in the 0.5-year median values of 137 Cs exponentially decreased, although the data were very limited.In the South Atlantic Ocean, the 0.5-year median values also decreased exponentially after 1970.Notably, the 0.5-year median values in 2003 in the central and South Atlantic oceans increased slightly compared to those in 2002.

3. 8
Figure2hshows the temporal variations in the 0.5-year median values of137 Cs in the Arctic Ocean (box 18), Barents Sea and coast of Norway (box 20), North Sea (box 22), and northern North Atlantic Ocean (boxes 25.1 and 25.2).The dominant sources of 137 Cs in this area are the global-scale atmospheric deposition in the 1960s by large-scale nuclear weapons tests and two nuclear fuel reprocessing plants after the 1970s.In these regions, the 0.5-year median values of137 Cs did not decrease exponentially due to the 137 Cs being discharged from the nuclear fuel reprocessing plants.In the mid-1970s, the 0.5-year median137 Cs activity concentrations in the North Sea and northern North Atlantic Ocean increased rapidly and reached a the maximum in the years 1977 and 1980.5, respectively.Thereafter, the 0.5-year median 137 Cs activity concentrations in these regions rapidly decreased, and the decreasing rate became small after 1990, which was associated with the reduced amount of released 137 Cs.In the Arctic Ocean, the 0.5-year median 137 Cs activity concentrations increased until the mid-1970s and then decreased.The overall 0.5-year median 137 Cs activity concentrations until the 1970s in the Arctic Ocean were lower than those in the surrounding oceans.

Figure 3 .
Figure 3.Comparison with 0.5-year median 137 Cs values in the Pacific Ocean, Indian Ocean, and Atlantic Ocean.

3. 11
Figure3shows a comparison of the 0.5-year median137 Cs values in the Pacific Ocean, Indian Ocean, and Atlantic Ocean.A significant feature is that the highest values were observed in the western North Pacific Ocean, and the values decreased exponentially over time until 2011.In contrast, the values in the western equatorial Pacific Ocean, western subtropical South Pacific Ocean, and Indian Ocean increased gradually in the 1970s and 1980s, followed by a decrease after the 1990s.The difference in the 0.5-year median137 Cs values in the Pacific Ocean and Indian Ocean became very small after 1980.Although the data are very limited, the 0.5year median 137 Cs values in the South Atlantic Ocean were lower than those in the Pacific Ocean and the Indian Ocean.In the 2000s, the 0.5-year median 137 Cs values in the South Atlantic Ocean was increased and the values were close to those in the Pacific Ocean and Indian Ocean.
Figure4shows the temporal variation in the 0.5-year median137 Cs values in the western North Pacific Ocean and the Sea of Japan as being a typical case because of the sequential time series are available in these boxes.The estimated Tap for the western North Pacific Ocean was 15.0 years, whereas the Tap in the Sea of Japan was estimated to be 16.4 years.Although we did not show that in the figures, the Tap in the East China Sea was estimated to be 17.7 years.The longer Tap in the Sea of Japan and East China Sea compared with those in the western North Pacific Ocean suggests that137 Cs was provided into the Sea of Japan and the East China Sea from the western North Pacific Ocean.The Tap estimated in all boxes until 2010 is listed in Table 3. Tap1, which is the longest at 52.0 years, is estimated in the western equatorial Pacific Ocean, and Tpo in this box is negative.The shorter Tap1, which is approximately 4 years, is in the Sea of Japan.In the case of Tap2, the longer Tap values are in the Indonesian Archipelago (36.7 years), with negative Tpo (−169.2years), and western subtropical North Pacific Ocean (34.1 years), with negative Tpo (−260.7 years).In Tap3, longer Tap occurs over 38.0 years for the central Atlantic Ocean, 15.4 years for the South Atlantic Ocean, 25.2 years for western subtropical North Pacific Ocean, and 15.6 years for western equatorial Pacific Ocean.In the many boxes with the longer Tap, the Tpo is estimated to be negative, as shown in Table3.The negative Tpo suggests that the physical oceanographic processes were controlled by the variation in 137 Cs activity concentrations.

Figure 4 .
Figure 4. Temporal variation in the 0.5-year median 137 Cs values.(a) Western North Pacific Ocean.(b) Sea of Japan.The lines represent the exponential decay of the 0.5-year 137 Cs median value between 1970 and 2010.
Figure7shows the spatial variations in the 0.5-year median 137 Cs values in the global ocean every 5 years.In 1970, the 0.5-year median 137 Cs values were higher in the North Pacific Ocean and lower in the South Pacific Ocean.In particular, the highest value (14.8 Bq m −3 ) was observed in the eastern North Pacific Ocean.In the South Pacific Ocean, relatively high values occurred in the western and western sub-

Figure 5 .
Figure 5. Horizontal distributions of 137 Cs deposition density (kBq m −2 ) as of the 1 January 1970.(a) Global ocean.(b) Northern North Pacific Ocean and its marginal seas.Black circles are locations of the F1NPS, Sellafield, La Hague, and Chernobyl power plants.

Figure 6 .
Figure 6.Horizontal distributions of the 137 Cs deposition amount (PBq) in each box as of 1 January 1970.(a) Global ocean.(b) Northern North Pacific Ocean and its marginal seas.Black circles are locations of the F1NPS, Sellafield, La Hague, and Chernobyl power plants.
shows the time variation in 137 Cs inventories in the surface mixed layer during 1970-2015.The 137 Cs inventory in the surface mixed layer in the global ocean in 1970 was estimated to be 184 ± 26 PBq.This result indicates that 32 % of the deposited 137 Cs remained in the surface mixed layer; in other words, 68 % of the deposited 137 Cs was transported below the surface mixed layer on a decadal scale in 1970.In 1970, the 137 Cs inventory was the largest in the North Pacific Ocean, followed by the North Atlantic Ocean and the marginal sea, Atlantic Ocean, and South Pacific Ocean.The 137 Cs inventories increase until 1980, and the inventory is estimated to be 201 ± 27 PBq in 1975 and 214 ± 11 PBq in 1980 due to the discharge of 137 Cs from the Sellafield and La Hague reprocessing plants.According to the estimation by OSPAR (2021), approximately 41.4 PBq (32 PBq until 1980; Aarkrog, 2003) and 1.04 PBq (0.70 PBq until 1980; Aarkrog, 2003) of 137 Cs were discharged from the Sellafield and La Hague plants from 1970 to 1998, respectively.The contribution from the nuclear fuel reprocessing plants and large-scale nuclear weapons tests (39 %) observed in the Arctic Ocean and the northern North Atlantic Ocean and its marginal sea (the Arctic Ocean, BarentsSea and coast of Norway, Baltic Sea, North Sea, northern North Atlantic Ocean, Irish Sea, and the English Channel) resulted in a large 137 Cs inventory in 1980.After 1980, the 137 Cs inventory decreased gradually and was estimated to be 37.2 ± 3.6 PBq in 2010, immediately before the F1NPS accident.Although the 137 Cs inventory decreased over time after 1980 until 2010, immediately before the F1NPS accident, the relative contributions of the 137 Cs inventory in the South Pacific Ocean, Indian Ocean, and Atlantic Ocean gradually increased and were estimated to be 20 %, 15 %, and 20 % in 2010, respectively.After the F1NPS accident, the 137 Cs inventory increased and was estimated to be 50.7 ± 7.3 PBq.

Figure 10 .
Figure 10.Mixed-layer depth in each box in the global ocean.(a) Global ocean, (b) North Atlantic Ocean and its marginal sea, and (c) Irish Sea (unit is m).The symbol "-" means that there are no mixed-layer depth data.

4. 2
Transport of 137 Cs from the Pacific Ocean to the Indian Ocean via the Indonesian Sea throughflow As described in the previous sections, seawater with a relatively large 137 Cs density ( 137 Cs activity concentrations) is transported westwards in the equatorial Pacific Ocean and the western subtropical South Pacific Ocean.It is known that the warm seawater in the equatorial Pacific Ocean is transported into the Indian Ocean by the wind-forcing circulation through the Indonesian Archipelago, namely the Indonesian Throughflow

Figure 14 .
Figure 14.Temporal variations in the 137 Cs inventory for every 5 years in the global ocean surface seawater.

Figure 15 .
Figure 15. 137Cs density in the surface mixed layer in the North Pacific Ocean, equatorial Pacific Ocean, and South Pacific Ocean.

Figure 16 .
Figure16.137Cs density in the Indonesian Archipelago and surrounding sea in the surface mixed layer.

Figure 17 .
Figure 17.Temporal variations in the 0.5-year median values of surface 137 Cs in the northern/equatorial Pacific Ocean and its marginal seas after 2011.(a) Western and eastern North Pacific Ocean.(b) Sea of Japan and East China Sea.
Cs in the subtropical North Pacific Ocean and equatorial Pacific Ocean showed positive values (0.1-1.3, 0.08-1.3,and 0.07-1.3PBq, respectively).The net inflowed 137 Cs also occurred in the eastern subtropical South Pacific Ocean (5.2 and 1.1 PBq) and Southern Ocean (1.7 and 1.7 PBq) in 1975 and 1980 and the eastern South Pacific Ocean (0.8 PBq) in 1980.The distribution of negative and positive values reflect the 137 Cs transport pattern, where the 137 Cs deposited in the surface mixed layer in the western North Pacific Ocean was transported eastwards and accumulated into the eastern subtropical Pacific Ocean.Then, these were transported southwards, with the subsidence associated with California Current and westwards in the equatorial Pacific Ocean.The 137 Cs moved southwards due to subduction in the eastern subtropical North Pacific Ocean and upwelled in the western/eastern equatorial Pacific Ocean.The negative values (−0.9 to −1.7 PBq for 1975; −0.4 to −1.1 PBq for 1980) in the western subtropical South Pacific Ocean, western South Pacific Ocean, and eastern Southern Ocean and positive values (0.3-2.5 PBq for 1975; 0.3-1.7 PBq for 1980) in the Arabian Sea, Indian Ocean, and Southern Ocean would result in the transport of 137 Cs from the Pacific Ocean into the Indian Ocean through the Indonesian Archipelago in 1975 and 1980.After 1990, the positive 137 Cs values in the eastern/western subtropical North Pacific Ocean and equatorial Pacific Ocean became negative (Fig.

Figure 18 .
Figure 18.The horizontal distribution of the 137 Cs net in-or outflow amount in each box against the deposition amount in 1970, based on the 0.5-year 137 Cs activity concentration data.The amount of 137 Cs net outflow includes the downward transport portion below the surface mixed layer and the horizontal transport in the surface mixed layer to the downstream boxes.A positive value (red) indicates the net inflow amount, and negative values (blue) indicate the net outflow amount.(a) Global ocean, (b) northern North Pacific Ocean and its marginal seas, and (c) Irish Sea and English Channel (unit is PBq).

Figure 19 .
Figure 19.Mass balance of 137 Cs in the surface seawater in each box in the global ocean.A positive value (red) indicates net inflowed 137 Cs from the upstream boxes, and a negative value (blue) indicates the net outflow 137 Cs to the downstream boxes or below the surface mixed layer compared to the previous 5 years (unit is PBq).(a) 1975, (b) 1980, (c) 1985, (d) 1990, (e) 1995, (f) 2000, (g) 2005, (h) 2010, and (i) 2015.

Figure 20 .
Figure 20.Mass balance of 137 Cs in the surface seawater in each box in the northern North Atlantic Ocean and its marginal seas.A positive value (red) indicates the net inflowed 137 Cs from the upstream boxes, and a negative value (blue) indicates net outflowed 137 Cs to the downstream boxes or below the surface mixed layer compared to the previous 5 years (unit is PBq).(a) 1975, (b) 1980, (c) 1985, (d) 1990, (e) 1995, (f) 2000, (g) 2005, (h) 2010, and (i) 2015.

Figure 21 .
Figure 21.Mass balance of 137 Cs in the surface seawater in each box in the Irish Sea and English Channel.A positive value (red) indicates the net inflowed 137 Cs amount from the upstream boxes, and a negative value (blue) indicates the net outflowed 137 Cs amount to the downstream boxes or below the surface mixed layer compared to the previous 5 years (unit is PBq).(a) 1975, (b) 1980, (c) 1985, (d) 1990, (e) 1995, (f) 2000, (g) 2005, (h) 2010, and (i) 2015.

Figure 22 .
Figure 22.Distribution of 137 Cs deposition on 1 January 1970 and 137 Cs transport route in the surface seawater of the global ocean deduced.The circles are locations of the nuclear reprocessing plants, Chernobyl, and F1NPS.
Inomata and Aoyama (2022b).2) and western North Pacific Ocean (box 2; boxes 2.0-2.6) for the F1NPS accident, were divided into several subregions because significantly large values around the discharged region cause larger values in the estimate of the 137 Cs inventory.The box numbers, their corresponding longitudes and latitudes, the box name abbreviations, and the data number are listed in Table1.The sampling points in each box are displayed in the dataset (https://doi.org/10.34355/Kinet.KANAZAWA-U.00149;InomataandAoyama,2022a).Temporal variations in 137 Cs activity concentrations and 0.5year median values in each box are displayed in the figures in the dataset (https://doi.org/10.34355/Ki-net.KANAZAWA-U.00150)inInomataandAoyama (2022b).2.2The 0.5-year median values of 137 Cs the surface 5194/essd-15-1969-2023  Earth Syst.Sci.Data, 15, 1969-2007, 2023  137Cs activity concentrations in each boxThe 0.5-year median values of the surface 137 Cs activity concentrations in each box were calculated and are shown in the dataset (Inomata and Aoyama, 2022c, https://doi.org/10.34355/Ki-net.KANAZAWA-U.00151).These 0.5-year median values of the surface 137 Cs activity concentrations are useful for verifying the general ocean circulation models

Table 1 .
Detail of each box in the global ocean.

Table 2 .
The 0.5-year 137 Cs median value in each box.

Table 3 .
Tap and Tpo in the global ocean.

Table 4 .
137Cs deposition density and inventory as on 1 January 1970.

Table 5 .
137Cs inventory in each box in the global ocean.* * 137 Cs inventory was estimated by using the 137 Cs median value.NA means that there are no available data.MLD denotes the mixed-layer depth.