129I and its species in the East China Sea: level, distribution, sources and tracing water masses exchange and movement

Anthropogenic 129I as a long-lived radioisotope of iodine has been considered as an ideal oceanographic tracer due to its high residence time and conservative property in the ocean. Surface water samples collected from the East China Sea (ECS) in August 2013 were analyzed for 129I, 127I and their inorganic chemical species in the first time. The measured 129I/127I ratio is 1–3 orders of magnitude higher than the pre-nuclear level, indicating its dominantly anthropogenic sources. Relatively high 129I levels were observed in the Yangtze River and its estuary, as well as in the southern Yellow Sea, and 129I level in seawater declines towards the ECS shelf. In the open sea, 129I and 127I in surface water exists mainly as iodate, while in Yangtze River estuary and some locations, iodide is dominated. The results indicate that the Fukushima nuclear accident has no detectable effects in the ECS until August 2013. The obtained results are used for investigation of interaction of various water masses and water circulation in the ECS, as well as the marine environment in this region. Meanwhile this work provides essential data for evaluation of the possible influence of the increasing NPPs along the coast of the ECS in the future.

I in the environment 2 . The large and well documented 129 I releases and the high residence time of iodine in the ocean make 129 I an ideal oceanographic tracer, a number of studies have been carried out in the North Atlantic and Arctic Oceans using 129 I and its species for water circulation and marine environment investigation [3][4][5] . A few investigation of 129 I in the Pacific Ocean were reported [6][7][8] . However, investigation of 129 I in the East China Sea (ECS) and the Yellow Sea (YS) is rare. The Fukushima nuclear accident happened in March 2011 has released a large amount of radioactive substance to the atmosphere and seas. The measurement of 129 I in the Fukushima offshore seawater has shown a significant Fukushima derived 129 I signals 9,10 . However, there was no report showing the Fukushima derived radioactive substance has transported to the ECS and YS. In the recent years, 30 nuclear power reactors have been installed in the coastal region of China. The environmental safety and impact of nuclear facilities are highly concerned. 129 I, as a fission product, can be used to monitor and evaluate the impact of the nuclear facilities; this requires a baseline value of 129 I. However, no such 129 I baseline in seawater along the coastal region of China is available.
Iodine is one of redox sensitive elements; it occurs mainly as iodide (I − ) and iodate (IO 3 − ) with minor organic iodine in seawater. Iodate is thermodynamically stable and predominant in the open sea 11,12 , but can be converted to iodide in surface photic zones as well as in oxygen-deprived basins 13 . Reduction of IO 3 − to I − in seawater is often related to biological activity such as phytoplankton 14 , microorganism 15 . The speciation analysis of the anthropogenic 129 I can provide critical information for understanding the marine environment and process.
Due to the increased discharges of pollutants and nutrients, red tide frequently occurs in the ECS and YS in the recent years. Speciation analysis of iodine isotopes can provide a new insight to investigate and evaluate the impact of the pollution. To our best knowledge, no data on the species of iodine isotopes in the ECS has been reported. There are a numbers of water masses and/or water currents in the ECS. The interaction among these water masses directly influences the environment of the ECS. Investigation of the 129 I distribution in this region will provide an insight into the exchange and interaction of these water masses.
In this work, surface seawater samples collected in the ECS and its adjacent region were analyzed for iodine isotopes and their species, aiming to provide a baseline dataset of the distribution of 129 I and 127 I in this region and to trace the exchange and circulation of water masses in this region. The chemical species of 129 I and 127 I can be used to investigate the marine circumstance and environmental process in the ECS.

Discussion
129 I concentrations ((0.73-3.99) × 10 7 atoms/L) and the 129 I/ 127 I atomic ratios ((3.1-120) × 10 −11 ) in the surface water in the ECS is 1-3 orders of magnitude higher than the pre-nuclear value (1.5 × 10 −12 for 129 I/ 127 I ratio and 0.043 × 10 7 atoms/L for 129 I concentration) 16 , indicating the dominating anthropogenic source of 129 I in the surface seawater in the ECS. The highest 129 I level ((1.2-4.0) × 10 7 atoms/L for 129 I concentrations and (10.1-119) × 10 −11 for 129 I/ 127 I atomic ratios) was observed in Yangtze River mouth and a relative high 129 I levels of (1.0-2.5) × 10 7 atoms/L for 129 I concentrations and (7.0-12) × 10 −11 for 129 I/ 127 I ratios were measured in the southern Yellow Sea, and the 129 I levels decline towards to the ECS shelf, the lowest 129 I levels ((3.1-6.4) × 10 −11 for 129 I/ 127 I ratios) in the investigated area were measured in the ECS shelf and southern ECS.
The measured 129 I level in surface water in the ECS is more than one order of magnitude lower than that in contaminated seawater. In the North Sea, Baltic Sea, Norwegian Sea and Arctic, 129 I/ 127 I ratios of 10 −8 -10 −6 were reported because of huge amount of 129 I discharged from the European nuclear fuel reprocessing plants (NFRPs) at Sellafield (UK) and La Hague (France) has dispersed with sea current to these area 4,[17][18][19][20] . In the Fukushima offshore water, the 129 I/ 127 I ratios reached to 2.2 × 10 −9 (6.3 × 10 8 atoms/L for 129 I concentration) due to the releases from the Fukushima accident 10 (Supplementary Table S4).
Compared to the region with similar latitude as the ECS but no direct contamination by human nuclear activity, the 129 I level in the ECS is comparable to those observed in adjacent sea around Japan before Fukushima accident, where 129 I concentrations of (1.2-2.1) × 10 7 atoms/L (with 129 I/ 127 I ratios of 4.64-7.2 × 10 −11 ) were report 6 . But it is lower than those in the north Atlantic (31-50°N) ( 129 I concentrations of (4.0-12.7) × 10 7 atoms/L and 129 I/ 127 I ratios of (2-57) × 10 −10 ) 3 . The higher 129 I level in the North Atlantic results from the leakage and southwards dispersion of the high reprocessing derived 129 I water in the English Channel. In addition, the 129 I in the ECS is also comparable to those observed in the Indian Ocean (40-50°S), where the 129 I concentrations were (0.60-1.52) × 10 7 atoms/L ( Fig. 3) 21,22 . Based on above, it can conclude that 129 I in the ECS water fall in normal global fallout background level.
However, the 129 I in the ECS continental shelf are higher than that in the South Atlantic Ocean and the Antarctic ((0.05-0.4) × 10 7 atoms/L for 129 I concentrations and (0.28-2.9) × 10 −11 for 129 I/ 127 I ratios) (Fig. 3, Supplementary Table S4) 23,24 . This should be attributed to less fallout of NWTs derived radioactive substances in the south hemisphere. The dominantly anthropogenic 129 I in the ECS might originate from 1) releases from nuclear facilities in coastal region of the ECS; 2) fallout of the atmospheric NWTs in 1945-1980; 3) releases from NFRPs and long distance transport; 4) nuclear accidents (e.g. Chernobyl and Fukushima accidents).
Although huge amount of 129 I has been produced during the operation of the nuclear power plants (NPPs) 1 , most of it was kept in the fuel elements and only very tiny fraction might be released if leakage of fuel elements occurred. Up to 2013, total 13 nuclear power reactors in 4 NPPs including Qianshan (30.43°N, 120.92°E), Fangjiashan (30.43°N, 120.92°E), Sanmen (29.04°N, 121.62°E), and Fuqing (25.53°N, 119.52°E) in coastal region in the ECS were in operation. However, there is no report on the significant releases of fission products from any of these NPPs. An investigation on the 129 I in the seawater surround a Chinese NPP in the coastal region of the ECS has confirmed no excessive 129 I in the seawater in 10 km range of this NPP 25 . There is no any NFRPs or other nuclear facilities surrounding the ECS. Therefore, the direct contribution of nuclear facilities to the 129 I in the ECS is negligible.
It has been estimated that 50-150 kg of 129 I has been released to the atmosphere from the atmospheric NWTs in 1945-1980 26    sediment core and these tests were implemented more than 30 years ago, the direct contribution of the Chinese NWTs to the 129 I in the ECS, especially the present surface seawater, should be insignificant. However, the global fallout of 129 I from all atmospheric NWTs might be one of 129 I sources in the ECS because of the large releases. A relative high 129 I signal of the global fallout peaked in 1962-1963 has been observed in the sediment core collected in the Jiaozhou Bay 27 . Compared to the direct deposition of 129 I in the ECS in early time of atmospheric NWTs, the leaching of 129 I from the earth surface by rainwater and subsequent transport to the ECS through rivers might be one of important source and pathway of 129 I in the ECS.
105 NWTs have been conducted in the Pacific Proving Grounds (PPG) in the Marshall Islands (11°N) in 1946-1962, where is about 5000 km southeast of the ECS. Besides those injected to the stratosphere and contributed to the global fallout, large amount of radioactive substances from these NWTs were deposited in the southern Pacific Ocean surrounding the Marshall Islands (close-in fallout). A number of investigations have reported that the close-in fallout of plutonium in the PPG has being carried by the north equatorial current, which is followed by the Kuroshio Current (KC) and continuously transported to the South China Sea and ECS 28,29 . The constantly significant PPG derived Pu signal observed in the ECS was attributed to the continuous re-suspension of the highly deposited plutonium in the sediment in the PPG region 28,29 . Compared to plutonium, iodine is water soluble and conservative in ocean, the close-in fallout of 129 I at the PPG can be also transported to the ECS in early time of the NWTs at PPG. The 129 I in the sediment core collected in Jiaozhou Bay in the YS has recorded a small but obvious PPG derived signal of 129 I in 1957 27 . Meanwhile, a similar 129 I peak occurred in 1957-1958 in coral samples collected in the coastal region in Philippine has also been reported 30 , which suggests that PPG derived 129 I might have a contribution to 129 I in the ECS. However, the distribution of 129 I in the ECS shows the lowest 129 I level in the east edge of the ECS, i.e. the KC belt, indicating no significant amount of 129 I in the ECS was transported from the PPG through the KC. This might be attributed that the NWTs at PPG has terminated in 1962, and transit time from PPG to the ECS is less than 2 years, the high water soluble 129 I released from these NWTs did not significantly deposited and remained in the sediment in the PPG area, instead it was quickly diluted and dispersed to a large area with sea currents. Therefore, unlike plutonium, the contribution of the PPG to the 129 I in the present surface water in the ECS is insignificant.
Releases from the NFRPs are so far the largest source of 129 I in the environment [1][2][3][4]  Besides atmospheric releases, the two NFRPs at Sellafield (UK) and La Hague (France) have discharged more than 6500 kg of 129 I to the seas, which was dispersed to a large area in northern North Atlantic Ocean by sea currents 26 , and an annual discharge of 129 I still remains at a high level of about 250 kg/y. Re-emission of the reprocessing derived 129 I from the contaminated seawater to the atmosphere has become a major source of 129 I in the air in recent years 32 . The 129 I record in the sediment core from the YS has demonstrated that the atmospheric releases of the European NFRPs and re-emission of the reprocessing derived 129 I in the sea is a major source of 129 I in the present sediment in the YC 27 . 129 I in the Japan Sea was also attributed to the European NFRPs 6 . All these suggest that the European NFRPs might be the major source of 129 I in the ECS besides the global fallout of the NWTs. The significantly higher 129 I level observed in Yangtze River mouth compared to the stations in the ECS indicates that riverine input might be an important source of 129 I in the ECS. Yangtze River is the longest river (6300 km length) in China with a catchment area of 1,808,500 km 2 located in southern China (25-34°N). The fallout of 129 I on the soil of the catchment has being continuously leached by rain and entered to river water, causing a high 129 I level in the Yangtze River. The high level 129 I in the southern YS indicate a higher deposited and riverine input of 129 I from the Yellow River. The YS and the catchment area of the Yellow River are located in north China (32.5-40°N), and therefore received higher fallout of 129 I compared to the ECS, because most of NWT sites and NFRPs are located in the middle and higher latitude in the Hemisphere 16,33 . 129 I levels in soil and seaweed in the north China and the YS is higher than that in the south China and the ECS 34,35 .
Besides the Yangtze River, there are more than 40 rivers flowing into the ECS, among them the relative big ones include Qiantang River (55,600 km 2 catchment area and 688 km length), Min River (61,000 km 2 catchment area and 577 km length) and Ou River (18,000 km 2 catchment and 388 km length). Considering total surface area of 1,940,000 km 2 , the riverine input from the terrestrial deposition is one of major contribution of 129 I in the ECS. This is also supported by investigation on the distributions of another important fallout radionuclide 137 Cs in the sediment in the Yangtze River estuary, which showed that the major part of 137 Cs in the estuary originated from riverine transport and input 36,37 .
It has been estimated that 1.3-6 kg 129 I has been released from the Chernobyl accident. 131 I signal in the atmospheric samples collected at northern China in May 1986 immediately after the Chernobyl accident has been observed, and a small signal of Chernobyl accident derived 129 I has been also observed in the sediment collected in the YS 27 . However, no 131 I in the atmosphere in southern China was observed in 1986 after the Chernobyl accident. Considering the small releases, far distance, and happened more than 25 years ago, Chernobyl accident contributed insignificant amount of 129 I in the surface water in the ECS.
Fukushima nuclear accident happened in March 2011 has released 1.2 kg 129 I including about 0.35 kg 129 I discharged directly to the sea, which caused an elevated 129 I level in the Fukushima offshore seawater with the highest 129 I/ 127 I atomic ratio of 2.2 × 10 −9 in the surface seawater near the Fukushima Dai-ichi NPP 10 . Due to the dominant westerly wind after the accident, the deposition of Fukushima derived radioactive substances were relative low in China, especially in the south China. Consequently, the direct deposition of Fukushima derived 129 I in the ECS and the catchment of the rivers flowing into the ECS is insignificant. Marine 129 I from the Fukushima accident was mainly transport eastwards in the North Pacific Ocean with the KC. An elevated 129 I in the surface seawater has been observed in the North Pacific Ocean in 2013 38 . It is therefore concluded that the Fukushima accident contributed insignificant amount of 129 I to the ECS.
Based on the marine dispersion model, it was estimated that the Fukushima derived radioactive substances will arrive in the ECS in 2013 and reach to peak value in 2018 39 , however, the relative low 129 I level in the center and eastern ECS compared to the northern and western ECS indicates that Fukushima accident derived 129 I has not yet arrived to the ECS, at least not measurable until summer 2013.
The increasing gradient of 127 I concentrations and declined 129 I level in the surface water from the Yangtze River estuary towards the ECS shelf (Fig. 1) shows an interaction of Yangtze River water with seawater from the KC and TWC, and the mixing model of these water masses (Fig. 1). The high 129 I but low 127 I level in the riverine water and high 127 I but low 129 I level in seawater from the TWC and KC makes 129 I/ 127 I atomic ratio a perfect indicator to trace the water masses interaction in this region. From the distribution of 127 I and 129 I concentrations and 129 I/ 127 I ratios in the ECS (Fig. 1), it can be observed that the Yangtze River water splits into three branches when it injects into the ECS: one branch moves southwards along the continental shelf of the ECS, extends as far as down to 26°N, showed by rapidly decreased 129 I concentrations and 129 I/ 127 I ratios along this transport pathway; the second branch moves southeastwards, from the Yangtze River estuary to the continental shelf of the ECS, showed by the decreased 129 I/ 127 I ratios from 1.17 × 10 −9 to 4.14 × 10 −11 at 126°E; in the third branch, Yangtze River water moves northeastwards, encountered the cold water mass from Yellow Sea, the cyclone circulation produced by the cold water mass turns this branch to the east and extend to the Jeju Island. This observation in 2013 is supported by an investigation of depth profiles of conductivity, temperature, and drifter trajectory in the ECS water in the summers of 1997 and 1998, showing the Yangtze River water plume moved eastwards to the offshore and further to Jeju Island 40 . A significantly declined 129 I level from the southern YS to the ECS was observed for both coast water and offshore water. This trend indicates that the high 129 I Yellow Sea water moves from north to south, and meet the high 129 I level Yangtze River water from the Yangtze River estuary in the border of the Yellow Sea and ECS, and then turn to northeast at 32°N. 129 I concentrations in the surface seawater at the border of the ECS and YC are 1-3 times lower than that in the surrounding region. Meanwhile the 127 I concentrations in this region (sampling stations 47 and 48) are relative low (30-33 μ g/L). 129 I absent ground seepage water input along the shore might occur here. A hypoxic condition observed at sampling station 48 (Supplementary Table S1) supports this assumption. Low salinity and low temperature observed at sampling stations 47-49 also support this assumption. An upwelling zone in the region of 122°00′ -123°30′ E, 31°-32°30′ N (stations 47-48) has also been observed in summer 1985 41 . Therefore, the low 129 I and 127 I level in this region might results from the groundwater seepage and upwelling of seawater in this region.
Low 129 I region observed at the continental slope of the ECS (stations 28, 36 and 41) might indicate a branch of the KC in the ECS in this region. The KC formed from the subtropical mode water of the central North Pacific 42 contains low 129 I (Fig. 1). The distribution of 129 I level indicates that the major KC flows northeastwards along the out edge of the ECS continental shelf, and a small branch moves northwards when it enters the ECS along a transection at stations 28, 36 and 41. The low 129 I level at stations 11 and 12 indicates that another branch of the KC moves northwestwards and mixed with the Yangtze River water and the Taiwan warm current (TWC) at about 31°N in the northern ECS.
Low 129 I level was also observed along the south coast of the ECS at stations 24, 25 and 32. This might indicate that the high 129 I level Yangtze River water does not move southerly along the coast. Considering 127 I concentrations are also lower in this region and there is not big river flowing into the ECS in this region, it can be assumed that the groundwater seepage in this region might also occur. 127 I mainly exists as iodate with iodide/iodate molar ratios of 0.04-0.60 in most sampling stations in the ECS, this is similar to those observed in the North Sea (0.11-0.48) 4,43 . The lowest 127 I − / 127 IO 3 − molar ratio of 0.04 was observed at sampling station 41, close to the Mien Hwa Canyon (MHC) with a northwestwards KC intrusion branch. This might be attributed to the Kuroshio subsurface water (KSW) upwelling in this region 44,45 , because iodate is normally dominated in deep seawater due to less biological activity. The low iodide/iodate ratios in the continental shelf edge of the ECS might also result from some vortex upwelling, when the KC moves northeastwards along the edge of continental shelf 45 . Besides the upwelling water, the low iodide fraction (or oxidation condition) in the KC water might be another reason. No significant correlation between 127 I − / 127 IO 3 − and 129 I − / 129 IO 3 − ratios (R 2 = 0.007) was observed for all seawater samples investigated. Meanwhile 129 I − / 129 IO 3 − ratios are normally higher than 127 I − / 127 IO 3 − ratios. This is attributed to the different sources of 127 I and 129 I in these waters. 129 I in the ECS mainly originates from the atmospheric fallout and riverine input, whereas 127 I originates from the sea currents which flow into the ECS.
In the Yangtze River mouth, both of 129 I and 127 I originated from the rainwater leaching of soil in the catchment, the similar high iodide/iodate ratios of 12 for both 127 I and 129 I were measured at station 54. This indicates iodide is the dominant species of iodine isotopes. This is similar as that observed in the Elbe River in the North Sea 4 , indicating that the Yangtze River water, at least in the mouth area, is reductive/anoxic water. Declined 127 I − / 127 IO 3 − ratios of 0.6-1.5 were measured in the downstream water at stations 52 and 53. This might be attributed to the tidal action and sea water intrusion showed by an elevated salinity (0.5) at sampling station 52. The similar declined 129 I − / 129 IO 3 − ratio (1.5) was measured at station 52, whereas a much higher value up to 20 at sampling station 53. Compared to 127 I, 129 I − / 129 IO 3 − ratios might reflect the water circumstance in this region, because the contribution from the seawater intrusion is minor. The high 129 I − / 129 IO 3 − ratio at sampling station 53 indicates the reductive circumstance of the Yangtze River water remained at this station. The declined 129 I − / 129 IO 3 − ratio at station 52 close to the Yangtze River estuary indicates the effluence of the seawater, and the reductive circumstance is changed, causing oxidation of 129 I − to 129 IO 3 − .
High 127 I − / 127 IO 3 − ratios  were measured at sampling stations 1 and 51 in the ECS, which are even higher than that in the Yangtze River mouth. It has been reported that a red tide (phytoplankton blooming) occurred in the estuary of the Yangtze River in August 2013, and the measured concentrations of chlorophyll were 5.6-37.7 mg/m 3 46 . The red tide phenomenon was also observed at sampling stations 1, 2 and 51 during the sampling expedition in August 2013. The dominant iodide specie in this region might result from the reductive condition due to the phytoplankton blooming in this region 14,47 .
High iodide/iodate ratios were also observed at sampling station 4 (1.3 for 127 I and 2.9 for 129 I), which are comparable or even higher than that in the Yangtze River mouth (stations 52). An oxygen depleted condition was measured in the water at sampling station 3, 4 and 48 during sampling campaign in August 2013. The dominant iodide specie in this region (sampling sate 3-4) might results from the anoxic condition in this region, which reduce iodate to iodide.
Since the total 129 I concentrations in the water at the sampling stations 1-4 are similar to those in the southern region at sampling stations 6-9, the elevated 129 I − / 129 IO 3 − ratios in sampling station 3-4 should be attributed to the reduction of 129 IO 3 − in the sampling region (station 3-4) which was transported from the Yangtze River estuary northwards through the sampling station 6-9. This also indicates that the reduction of iodate in the anoxic region or biologically active region, especially phytoplankton blooming region, is a relative fast process.

Materials and Methods
Investigation area. The ECS is the largest marginal sea in the northwestern Pacific, bounded by China, Japan, Ryukyu archipelago, and Korea (Fig. 4). It receives fresh water input from rivers, among them Yangtze River contributes 90% of fresh water input. The KC flows into the ECS through the Channel between Taiwan and Yonaguni-jima Island, and moves northward along the edge of continental shelf. The KC follows the North Equatorial Current in the Pacific Ocean which passes through the PPG in Marshall Islands where 105 NWTs were conducted (11°N) in 1946-1962. The TWC formed by the mixing of the KC surface water and the Taiwan Strait Water, enter the ECS from the Taiwan Strait. In the northern part of the ECS shelf, the Yellow Sea Surface Current flows southwards and participates in the structure of the ECS water masses.
Collection of the seawater samples. Surface seawater samples (< 2 m depth) were collected during the R/V "Dongfanghong 2" cruise in August 2013 using a built-in seawater sampling system in the laboratory of the vessel at 54 sampling stations in the ECS and its adjacent region (Fig. 4, Supplementary Table S1). The temperature and salinity of the seawater samples were measured on-line during sampling (Supplementary Table S-1). Samples were filtered through Φ 0.45 μ m filter membrane to remove suspended particles in situ and stored in 2.5 L polyethylene plastic containers in dark under normal laboratory conditions until analysis.
Separation of 127 I and 129 I species from seawater. A modified procedure from Hou et al. 4 which is based on anion exchange chromatography was applied to separate iodide and iodate from seawater for measurement of 127 I − and 127 IO 3 − . The detailed procedure is presented in the Supporting information. A modified co-precipitation method from Luo et al. 48 was used for separation of iodide and total inorganic iodine from seawater for measurement of 129 I. For separation of iodide, after addition of 0.20 mg of 127 I − carrier and 200 Bq 125 I − tracer to 1200 ml seawater, NaHSO 3 solution was added to final concentration of 0.3 mmol/L, and then HCl was added to adjust pH to 5.1, 50 mg Ag + (as 0.10 mmol/L AgNO 3 ) was slowly added to the sample under stirring to coprecipitate iodide as AgI-AgCl. For total inorganic iodine, 500 ml seawater was taken to a beaker, 0.20 mg of 127 I − carrier, 200 Bq 125 I − (NaI) and 2.0 mmol NaHSO 3 solution were added. HCl was then added to adjust pH < 2, iodate was reduced to iodide in this condition. 150 mg Ag + was slowly added to the solution under stirring to coprecipitate the formed iodide as AgI-AgCl. The precipitate was washing with 1-7.5% ammonium to obtain a final precipitate of 1-3 mg for AMS measurement of 129 I.
Procedure blanks for iodide and total inorganic iodine were prepared using the same procedure as for the samples. In this case, deionized water was used instead of the samples.

Measurement of 127 I using ICP-MS and 129 I by AMS. Iodine in the separated iodate and iodide fractions
and original seawater was measured used ICP-MS (Thermo Scientific, X series II, USA) after 10 times dilution using 1% ammonium solution. The detection limit of 0.02 ng/mL for 127 I was obtained.
The separated AgI-AgCl coprecipitate was dried in an oven at 60-70 °C, the dried precipitate was ground to fine powder and mixed with niobium powder (325-mesh, Alfa Aesar, Ward Hill, MA) in a mass ratio of 1:5, which was finally pressed into copper holder. The 129 I/ 127 I atomic ratios in the prepared targets were measured by AMS using 3MV Tandem AMS system (HVEE) in the Xi'an AMS center. 127 I 5+ was measured as charges (current) using a Faraday cup and 129 I 5+ was measured using a gas ionization detector. All samples were measured for 6 cycles and 5 min per sample in each cycle. The procedure background of 129 I/ 127 I ratio was measured to be 1.0 × 10 −13 . For the samples with 129 I/ 127 I ratio of 10 −10 , the measurement uncertainty is less than 1.75% 49 . A detailed description of AMS system and measurement of 129 I has been reported elsewhere 50 .