Tracing the upwelling process in the northern Benguela upwelling system (nBUS) by 129I

Tracing the upwelling process in the northern Benguela upwelling system (nBUS) by 129 I Jos e María L opez-Guti errez a , b , * , Deon C. Louw c , Martina Ro zmari c d , Victoria L erida-Toro a a Centro Nacional de Aceleradores (CNA), Universidad de Sevilla. Junta de Andalucía. Consejo Superior de Investigaciones Científicas, Parque Científico y Tecnol ogico Cartuja, Thomas Alva Edison 7, 41092, Sevilla, Spain b Universidad de Sevilla. Departamento de Física Aplicada I. Escuela Polit ecnica Superior, C/ Virgen de  Africa, N  7, 41011, Sevilla, Spain c National Marine Information and Research Centre (NatMIRC), Ministry of Fisheries and Marine Resources, 1 st Strand Street, 13000, Swakopmund, Namibia d International Atomic Energy Agency, Environment Laboratories, MC 98000, Monaco highlights graphical abstract New data on 129 I in seawater from the Southern Hemisphere, where available published data is very limited, is presented. Concentrations (10 6 -10 7 atoms/kg) fit the expected values of the global behaviour of 129 I in the marine environment. First set of 12 I/ 236 U values in the Southern Hemisphere is presented. The capacity of 129 I to trace the upwelling process in this zone is shown. article info Article history: Received 6 August 2020 Received in revised form 14 November 2020 Accepted 18 November 2020 Available online 21 November 2020 Handling Editor: Martine Leermakers Keywords: Accelerator mass spectrometry (AMS) Northern Benguela upwelling system (nBUS) Iodine-129 Uranium-236 Southern hemisphere abstract New data on the presence of 129 I in seawater in the Southern Hemisphere measured by Accelerator Mass Spectrometry (AMS) is presented. The samples were collected in 2014 along the Namibian coast during a cruise organised by the National Marine Information and Research Centre (NatMIRC), the national laboratories of the Ministry of Fisheries and Marine Resources (MFMR) in Namibia, and the IAEA Environment Laboratories (IAEA NAEL) in Monaco. The Benguela upwelling system is known as one of the most important marine upwelling regions in the world. Strong winds induce an offshore transport of surface seawater which is substituted by cool subsurface water inshore. As this water is nutrient-rich, which leads to high primary productivity, the Benguela upwelling system has a very important role as afishing production area. The 129 I concentrations in samples were between (0.66 ±0.14) 10 7 and (1.45 ±0.30) 10 7 atoms/kg. The highest 129 I concentrations were found in the offshore surface samples. Deep-sea and inshore samples contained lower 129 I concentrations, possibly as an effect of the upwelling process. A comparison with previously published studies suggests that the presence of 129 I in the northern Benguela upwelling system (nBUS), is mainly due to the impact of nuclear weapons global fallout, without any evident impact of nuclear fuel reprocessing. ©2020 Elsevier Ltd. All rights reserved. 1. Introduction The presence of 129 I(T 1/2 ¼15.7 10 6 years) in the environment *Corresponding author. Centro Nacional de Aceleradores (CNA), Universidad de Sevilla. Junta de Andalucía. Consejo Superior de Investigaciones Científicas, Parque Científico y Tecnol ogico Cartuja, Thomas Alva Edison 7, 41092, Sevilla, Spain. E-mail address: [email protected] (J.M. L opez-Guti errez). Contents lists available at ScienceDirect Chemosphere journal homepage: www.elsevier.com/locate/chemosphere https://doi.org/10.1016/j.chemosphere.2020.129065 0045-6535/©2020 Elsevier Ltd. All rights reserved. Chemosphere 265 (2021) 129065 has changed significantly since the beginning of the nuclear era. The main anthropogenic activities that have released this radionuclide into the biosphere include atmospheric nuclear weapons tests, nuclear accidents and discharges from nuclear fuel reprocessing plants. Nuclear bomb tests carried out between 1945 and the 1980s were responsible for the presence of approximately 90 kg 129 I in the environment (Hou, 2004;Raisbeck and Yiou, 1999) while nuclear fuel reprocessing plants have discharged more than 6000 kg (He et al., 2013). The impact of nuclear accidents has not been so severe, and it has been estimated that, e.g. between 1 and 6kgof 129 I were released into the atmosphere after the Chernobyl accident (Aldahan et al., 2007;Paul et al., 1987). The influence of these activities has not been homogeneous on the planet, as the Northern Hemisphere has been especially affected (Reithmeier et al., 2010). The most important nuclear fuel reprocessing plants, Sellafield (United Kingdom) and La Hague (France), release 129 I mainly in liquid form (He et al., 2013), most of which is transported by currents to the North Sea and from there to the Arctic Ocean along the Norwegian coast (Villa et al., 2015;Vivo-Vilches et al., 2018). Atmospheric releases by these reprocessing plants also contain 129 I. However, as 129 I is directly released to the troposphere, the distribution and therefore the impact on the Southern Hemisphere is low (Xing et al., 2015). The distributions of the released radionuclides were more homogeneous with the nuclear bomb tests, the most powerful of which were carried out during the 1950s and 1960s in the Northern Hemisphere. A significant part of the released radioactivity was injected into the stratosphere, distributed in it and slowly deposited on the whole planet surface, having a more homogeneous impact. Therefore, average 129 I concentrations in the hydrosphere increased from natural levels of ~4 10 5 to ~1.16 10 7 atoms/kg due to nuclear bomb tests and to ~10 9 atoms/kg in the Northern Hemisphere from reprocessing (Snyder et al., 2010). However, 129 I concentrations up to 10 11 e10 12 atoms/kg have been measured in certain zones of the Northern Hemisphere (Hou et al., 2007). A much lower impact on the Southern Hemisphere explains the scarcity of the available data in the literature about the presence of 129 I in this area of the planet. Nevertheless, 129 I conservative behaviour in seawater makes it a very important tracer of environmental processes, especially in the hydrosphere (Edmonds et al., 1998;Raisbeck et al., 1995). Moreover, during recent years, its tracer properties have been enhanced by its combined use with other conservative radionuclides such as 236 U(Castrillejo et al., 2017;Christl et al., 2015). Additionally, its long half-life can help reconstruct the impact of other iodine radioisotopes in the case of nuclear accidents (G omez-Guzm an et al., 2017). These properties demonstrate the requirement to increase our knowledge about the presence of 129 I and its transport mechanisms, especially in areas in which less information is available. In this work, the results on the 129 I levels in seawater samples collected in 2014 off the Namibian coast (from 20  Sto27  S) are presented. From a global point of view, the presented data is relevant as no data on seawater systems is available for these latitudes and very little for other latitudes in the Southern Hemisphere. The Namibian coast is part of the Benguela upwelling system (BUS), which is one of the most important upwelling systems in the world with high primary production (Carr and Kearns, 2003;Duncombe Rae, 2005;Shannon and Nelson, 1996). Strong equatorial wind stress induces offshore transport of surface water, which is replaced by cool, nutrient-rich subsurface water, which creates a fishery rich environment. The Benguela upwelling system is divided in the northern and southern Benguela upwelling systems (nBUS and sBUS, respectively) by the Lüderitz upwelling cell, which is the main upwelling cell. Apart from the main upwelling cell smaller coastal upwelling cells are also found close to Conception Bay, south of Walvis Bay, Cape Frio and Palgrave Point (Boyer et al., 2000)(Fig. 1). During austral spring and summer, the southeasterly winds promote upwelling in the sBUS (south of Hondeklip Bay, about 30  18 0 S) and during austral winter and spring in the nBUS, from approximately Cape Cross (21  46 0 S) to north of Cape Frio (18  26 0 S), however upwelling conditions are almost perennial at the main upwelling cell (from about 26  Sto28  S), (Shannon and Nelson, 1996). It is relevant to study the effect of the upwelling process on 129 I concentration and radioactive impact on this area. 2. Material and methods 2.1. Sampling Samples were collected in May 2014 off the Namibian coast. During this period, a sampling campaign was organised by the National Marine Information and Research Centre (NatMIRC) in Namibia, and the IAEA Environment Laboratories (IAEA NAEL) in Monaco. The aim was to establish baseline levels for a series of anthropogenic and natural radionuclides. The sampling points covered a large area from 20  Sto27  S latitudes from (Fig. 1). At each seawater sampling station, samples were taken from the surface (±2 m) by pumping water using a vessel pumping system and at different depths by using Niskin bottles attached to a rosette and a CTD instrument. Samples for 129 I analysis were sent to the Centro Nacional de Aceleradores (CNA) without pre-treatment. Fig. 1. Sampling locations along the Namibian coast. Plot made with Ocean Data View (Reiner and Schlitzer, 2020). J.M. L opez-Guti errez, D.C. Louw, M. Ro zmari c et al. Chemosphere 265 (2021) 129065 2 Results were obtained for 18 seawater samples, 14 surface samples and four deep stations. Oceanographic data (salinity, temperature, and oxygen) was measured during the survey and used for results interpretation. 2.2. Sample preparation The radiochemical method which was used is based on (Muramatsu et al., 2008). Approximately 0.5 L of filtered (Sartorius™0.45 m m) seawater was used for 129 I determination. At first, 2 mg of stable iodine carrier was added. As only isotopic ratios 129 I/I are measured in Accelerator Mass Spectrometry (AMS), the 129 Iin the samples was obtained by multiplying the isotopic ratio times the amount of stable iodine added, as the original stable iodine was negligible. Decomposition of organic matter and oxidation of iodine species to iodate (IO 3  ) was carried out by the addition of NaOCl. The sample was kept in an automatic shaker for 30 min at 150e200 rpm and it was stored in darkness for at least 12 h. Then, sodium bisulfite (Na 2 S 2 O 5 ) and hydroxylamine hydrochloride (NH 2 OH$HCl) were added for reduction of iodate (IO 3  ) to iodide (I  ) so that all iodine was in the same chemical form. After that, solvent extraction procedure was performed. Sodium nitrite (NaNO 2 ) 0.45 g/mL and nitric acid were used to oxidize iodide (I  ) to molecular iodine (I 2 ), as this chemical form is more soluble in the organic dissolvent (CHCl 3 ) which was used for extraction. Back extraction was done in an aqueous solution of 30 mL H 2 O-MQ and 0.5 mL of hydrazine (NH 2 eNH 2 ) 80%. Hydrazine acts as a powerful reduction agent of iodine into iodide. After these two consecutive extractions, the iodine was purified. We acidified the iodine solution with nitric acid and AgNO 3 was added in excess in order to precipitate silver iodide (AgI). Finally, AgI was dried and mixed with Nb powder (1:1) to increase its thermal and electrical conductivity. 2.3. Accelerator Mass Spectrometry measurements 129 I measurement was carried out at the CNA in Seville (Spain) by Accelerator Mass Spectrometry (AMS). Details on the facility and the AMS measurement can be found in (Calvo et al., 2015), although some processes have changed with respect to 129 I measurement. Iodine is extracted from the Cs sputtering ion source as I  and injected into the accelerator through an injector magnet. Typical I  are around 1e2 m A in the low energy side. Terminal voltage is set to 700 kV approximately (instead of a maximum of 1 MV) in order to choose charge state 2þafter stripping. Otherwise, the magnetic rigidity of the 129 I 2þ ions in the high energy side would be excessive for the magnet. The stripper gas used in this facility is He, which has been shown to give high transmissions for heavy ions in AMS (Vockenhuber et al., 2015). Typical transmission at the accelerator is around 50% for 129 I 2þ in these conditions. After stripping, a magnetic analyser selects 129 I 2þ ions and an electrostatic deflector carries the ions to a gas ionization chamber where the energy of the ions is measured. All samples, standards and blanks were measured for four different cycles in each measurement. One standard is measured for every four or five samples and approximately one blank for every twelve samples. The average raw value of measured blanks is calculated and subtracted to the isotopic ratios obtained for samples and standards. The uncertainty associated to the average blank isotopic ratio was estimated as a maximum between statistical errors and standard deviations of the mean. Then, standard correction is carried out for all sample isotopic ratios. The standard we use at CNA is named Q2 and has a nominal isotopic ratio of 129 I/ 127 I¼(1.1169 ±0.0046) 10 11 . It was obtained from SRM 3231 Iodine-129 Isotopic Standard developed by NIST (NIST Office of Reference Materials, n.d.) and measured at the ETH Laboratory of Ion Beam Physics (C. Vockenhuber, personal communication). Finally, an average value of the corrected isotopic ratios is calculated. The final uncertainty associated to the isotopic ratio is generally the standard deviation of the mean, as measurement times are chosen so that statistical errors are low enough. Due to the low amount of 129 I in the Namibian samples, measured isotopic ratios were also low, typically in the order of 129 I/ 127 I~10 13 -10 12 . In order to assure the quality of the data, both instrumental and chemical blanks were prepared. Instrumental blank isotopic ratios ( 129 I/ 127 Iy110 13 ) were subtracted from sample measured ratios. A total of five chemical blanks were prepared using the same amount of distilled water as used for the Namibian samples and following the same procedure. The average isotopic ratio after correction was (1.83 ±0.78) 10 13 . As this value is very similar to the one obtained for the instrumental blanks, the samples were not corrected for hypothetical contamination during sample preparation. The obtained estimated uncertainties of the results were high due to relatively high amount of carrier added to the original sample to obtain good currents in the AMS measurement. This fact reduces the final isotopic ratio in samples which reach values which are not very different to the instrumental background so the number of counts in the detector is lower and background correction needs to be done. Currently we are carrying out a study on the behaviour of low-current samples in order to reduce background effects. We are also working on the optimization of radiochemical methods to maximise the efficiency of iodine extraction. 3. Results and discussion A compilation of the results of our research is presented in Table 1. The range of 129 I concentrations in seawater samples was between (0.66 ±0.14) 10 7 and (1.45 ±0.30) 10 7 atoms/kg. For four stations (2, 4, 6 and 9), surface and deep samples were measured. Results were always lower in deep seawater samples than in surface samples, which we expected because the 129 I source is the atmosphere. 3.1. Origin of 129 I off the Namibian coast Most of the available data on 129 I in the marine environment is related to measurements carried out in the Northern Hemisphere. As commented in the introduction, the strong impact of nuclear fuel reprocessing plants, especially at Sellafield and La Hague, has increased the natural levels of 129 I enormously. Measurements in the North Sea and in the English Channel by Hou et al. (2007), showed concentrations in the order of 10 11 atoms/kg as well as 129 I/ 127 I isotopic ratios between 10 8 and 10 6, . This indicates up to six orders of magnitude higher concentrations than the natural level (Snyder et al., 2010). Recently, Castrillejo et al. (2018) measured a transect along the North Atlantic Ocean, and found concentrations between 7.2 10 7 atoms/kg near the Portuguese coast and 77 10 7 atoms/kg in the Labrador Sea for surface seawater. In the Mediterranean Sea, Castrillejo et al. (2017) measured concentrations in surface seawater between 4.0 10 7 and 13.8 10 7 atoms/kg, indicating in this case the influence of the Marcoule reprocessing plant. The limited data available for the Southern Hemisphere shows much lower values of 129 I. Povinec et al. (2011) measured maximum concentrations of 1.15 10 7129 I atoms/kg in the Indian Ocean, which was decreasing to 0.33 10 7 atoms/kg in deeper layers. Negri et al. (2013) also measured 129 I in three seawater samples from the Atlantic Ocean near the Argentinian coast, indicating concentrations between 0.39 10 7 and 0.42 10 7 atoms/L. A more recent study by Xing et al. (2015) reported concentrations between J.M. L opez-Guti errez, D.C. Louw, M. Ro zmari c et al. Chemosphere 265 (2021) 129065 3 0.141 10 7 and 0.25 10 7 atoms/L in the Antarctic Ocean. Measured isotopic ratios were between 0.61 10 11 and 1.3 10 11 . In a follow up study in the Antarctic Ocean by the same author 129 I concentrations were in the range of (0.115e0.315) 10 7 atoms/L and isotopic ratios in the range of (0.61e1.98) 10 11 (Xing et al., 2017), so both studies had similar results. No data has been found in the literature for the South Atlantic eastern coast; however, some data in fresh water from Southern Africa has been published (Snyder and Fehn, 2004). These measurements of 129 I concentrations were typically in the order of 10 6 atoms/L in several locations between 1  S and 25  S with an exceptional value of 6.67 10 7 atoms/L at Water Hole (Botswana). Isotopic ratios were between 0.322 10 11 and 106.3 10 11 . Furthermore, Reithmeier (2005) showed measurements of 129 Iin Lake Victoria (United Republic of Tanzania, 1.3  S) and Lake Mweru (Democratic Republic of Congo, 9.8  S). He reported concentrations between 0.27 10 7 and 0.58 10 7 atoms/L and isotopic ratios of 6.060 10 11 and 61.12 10 11 , respectively. As can be seen in Table 1, the 129 I concentrations along the Namibian coast were approximately 1 10 7 atoms/kg or lower for surface samples, which is significantly lower than samples presently measured in the Northern Hemisphere. If an average iodine concentration in the sea of 60 mg/kg is estimated, isotopic ratios would range between 2.3 10 11 and 4.6 10 11 . These ratios show that the influence of discharges from nuclear fuel reprocessing plants is very low. In Fig. 2 a comparison of some results obtained from scientific literature shows 129 I in shallow seawater as a function of latitude. Results presented in this paper are in line with global tendencies and contribute to fill an existing gap regarding studied latitudes. As can also be seen in Fig. 2, the average 129 I concentration in shallow ocean regions calculated by Snyder et al. (2010) is also in good agreement with values measured in this paper. It is likely that the main contribution of the presence of 129 I in the nBUS is mainly due to the effect of global fallout from nuclear weapons tests. Although 129 I concentrations in lakes from Africa are typically in the order of 10 6 atoms/L, which is up to one order of magnitude lower than the levels in seawater from Namibia, they cannot be directly compared. Snyder et al. (2010) showed that 129 I concentrations in lakes and rivers tend to be one order of magnitude lower than adjacent marine shallow waters but isotopic ratios are much more similar to lakes and rivers. If the expected 129 I/ 127 I isotopic ratios which were estimated for Namibian samples, (2.3e4.6) 10 11 , are compared with ratios measured in African lakes (Reithmeier, 2005;Snyder and Fehn, 2004), both ratios are similar, as these are typically in the order of 10 11 . These ratios also coincide with the calculated average isotopic ratio due to nuclear weapons tests calculated by Snyder et al. (2010), which is about 410 11 . It is also possible to estimate the inventory of 129 I in the water column corresponding to the four sampling stations for which deep-sea values were measured (stations 2, 4, 6 and 9). The calculated values are (4.32 ±0.29), (3.35 ±0.58), (3.22 ±0.17) and (3.41 ±0.31) 10 12 atoms/m 2 , respectively. These values which were obtained are in line with each other and with the expected average deposition of 129 I due to global fallout, which is in the order of 1 10 12 atoms/m 2 (Snyder et al., 2010). Table 1 129 I concentration, 129 I/ 236 U ratios and basic oceanographic parameters for seawater samples collected along the Namibian coast. Station code Station name Sample Lat. Long. Distance to the shore Depth Salinity Temperature Oxygen 129 I concentration Error (1 s ) 129 I/ 236 U Error (1 s )  N  E km m PSU  C mL/L 10 7 atoms/kg 10 7 atoms/kg Atom/atom Atom/atom - 1 20002 1s 20.0 13.00 3.5 1 35.5 15.2 1.9 1.09 0.10 2.51 0.24 2 20070 2b 20.0 11.79 122.6 432 34.7 8.5 1.0 0.782 0.074 ee 2s 1 35.6 18.0 5.1 1.22 0.11 2.41 0.24 3 23002 3s 23.0 14.37 3.4 1 35.3 13.9 1.8 0.87 0.14 1.74 0.43 4 23070 4b 23.0 13.14 129.7 342 34.9 9.8 0.9 0.684 0.028 ee 4s 1 35.3 16.5 4.9 1.28 0.34 2.70 0.73 5 26002 5s 26.0 14.92 3.6 1 35.1 11.8 1.7 0.699 0.088 1.61 0.34 6 26050 6b 26.0 14.03 89.2 297 34.9 10.0 1.1 0.880 0.050 ee 6s 1 35.2 15.1 5.8 1.29 0.11 2.28 0.23 7 26090 7s 26.0 13.28 166.8 1 35.2 15.6 5.8 1.45 0.30 2.66 0.58 8 27005 8s 27.0 15.14 9.7 1 35.2 11.8 2.5 0.95 0.14 2.16 0.33 9 27050 9b 27.0 14.30 92.8 363 34.7 8.6 2.4 0.66 0.14 ee 9s 1 35.1 15.0 5.8 1.217 0.090 2.41 0.21 10 WBHC 10s 22.9 14.49 0.1 1 ee e 1.005 0.049 2.47 0.20 11 ZZ1 11s 22.4 14.30 13.6 1 35.3 13.8 0.0 1.16 0.14 2.37 0.33 12 ZZ4 12s 22.1 13.20 34.3 1 35.2 12.2 0.2 1.02 0.26 2.28 0.77 13 ZZ7 13s 21.3 13.49 18.4 1 35.4 14.4 1.0 1.218 0.063 2.90 0.22 16 Lagoon 16s 26.6 15.14 5.5 1 ee e 0.749 0.074 1.90 0.21 Fig. 2. 129 I concentrations as a function of latitude in surface seawater samples from Namibia (this work) and other data available from literature. J.M. L opez-Guti errez, D.C. Louw, M. Ro zmari c et al. Chemosphere 265 (2021) 129065 4 3.2. 129 I/ 236 U ratios The use of the dual tracer 129 I/ 236 U is very relevant as it has been demonstrated to give additional information in the tracing of environmental processes in comparison to the use of a single radionuclide. 236 U concentrations were previously measured in some of the samples presented in this work (L opez-Lora et al., 2020). The calculated 129 I/ 236 U ratios ranged between (1.90 ±0.21) and (2.90 ±0.22) (Table 1). These values can be compared to the samples previously measured in the Arctic Ocean (Casacuberta et al., 2018). A wide range of 129 I/ 236 U ratios was found in this area: from (1 ±4) in the bottom depths of the Makarov Basin to (800 ±50) in surface samples from the Barents Sea shelf. This difference shows the strong impact of nuclear fuel reprocessing plants again from Sellafield and La Hague. In Casacuberta et al. (2016), the reference 129 I/ 236 U value for global fallout is estimated to be 0.5, while it increases to 350 for seawater affected by discharges from reprocessing plants. In the case of pre-nuclear levels, the average would also have been higher, 129 I/ 236 U¼366. The results for the Namibian samples clearly indicate that global fallout is the main component of 129 I presence. 3.3. Tracing of water masses in the nBUS using 129 I In Table 1 and Fig. 3 the highest 129 I concentrations are typically observed in surface samples collected offshore. Actually, even for samples collected as close as 13.6 km from the coast (sample 11s), the 129 I concentration is already over 1 10 7 atoms/kg. On the other hand, 129 I concentrations in deep seawater samples are slightly lower and similar to values found in surface samples taken inshore. Lower concentrations for deep samples are typically expected for regions in which the main input is atmospheric. In the same graph, different colours have been chosen for the four different latitudes for which several samples are available. Inshore samples systematically show lower concentrations of 129 I than surface offshore samples, even if the relatively large errors for some of the measurements are taken into consideration. This could be interpreted as a result of the upwelling processes present in this area. Surface inshore samples could originate from central water samples upwelled near the coast. Then, as this water moves offshore along the surface of the sea, this water mixes with oceanic surface water and increases its 129 I concentration. If physicochemical characteristics are analysed for the different samples used in our research (Fig. 4), the distribution is in line with the findings by Duncombe Rae (2005) and Mohrholz et al. (2008). Three water masses can be found: Oceanic Surface Water (OSW), Central Water (CW) and Modified Upwelled Water (MUW). Sample 2s is categorized as OSW, with similar characteristics to the surface water that can be found in the Southern Atlantic outside the tropics. This sample was taken at 20  S and about 120 km offshore. Samples 4s, 6s, 7s, 9s and 12s present typical values which these authors described as MUW. They correspond to surface samples taken offshore and present the highest 129 I concentrations of all the samples. Finally, from their salinity and temperature, the rest of the samples can be considered as CW. This group includes a mixture of deep offshore water and surface inshore water. These results could support the idea that inshore surface samples originate from central waters that emerge close to the shore and would present lower 129 I concentrations than surface samples offshore. Although 129 I concentration does not present any dependence with latitude when all our samples are included, a certain relationship can be observed if the range is limited to CW samples. In Fig. 5, the 129 I concentration is plotted against latitude. Different colours are used for shallow water samples (<50 m) and deep water samples (>50 m). Despite the large error bars in some cases, shallow water samples present characteristically slightly higher 129 I concentrations in the north than in the south. This effect is not observed in deep water samples. It has been shown that the nBUS is affected by a biannual intrusion of tropical waters, with maxima in October and February (Junker et al., 2017;Mohrholz et al., 2008; Monteiro et al., 2008). The results of our work regarding 129 I distribution could reflect the relative contribution of central waters from the north and south in the region of Namibia. Although the oxygen concentration variability in this region may be modulated by other complex factors (Monteiro et al., 2008), a higher 129 I concentration may be related to a characteristic hypoxia in tropical origin water, caused by the decomposition of sinking organic matter in the Angola gyre (Mohrholz et al., 2008). As iodine plays an important role in organic processes, this phenomenon could enhance the presence of dissolved 129 I in seawater. If the 129 I/ 236 U ratio is plotted against the 129 I concentration for the samples belonging to the CW group (Fig. 6), a clear relationship can be observed. The isotopic ratio increases as the 129 I concentration does, so the variability in the 129 I concentrations is not found in 236 U. We have found that both radionuclides have the same origin. So the dependence of 129 I/ 236 U ratios on 129 I concentration, demonstrates that there is an unknown aspect which affects 129 I and 236 U in a different way to its behaviour in these waters. If the hypothesis given in the previous paragraph about the decomposition of organic matter due to the intrusion of hypoxic water is correct, this means that it would not affect to the presence of dissolved 236 U in the way it affects 129 I. 4. Summary and conclusions A new set of data on the presence of 129 I in the northern Benguela upwelling system presented in this paper increases the scarce amount of data available in literature for the Southern Hemisphere. The 129 I concentrations measured in seawater samples from the Namibian shelf are typically around 1 10 7 atom/kg for surface water and lower for deep-sea water. These values suggest that global fallout from the nuclear tests carried out since the 1950’sis the main source of 129 I in the northern Benguela ecosystem. This is supported by the calculation of 129 I/ 236 U ratios, in the order of 1, which are much lower than samples affected by discharges from nuclear fuel reprocessing plants. Fig. 3. Evolution of 129 I concentration in surface water with distance offshore. Different symbols have been chosen for deep water and surface water samples. In addition, different colours have been assigned to the four latitudes for which more than one sample was taken so that the increase of 129 I concentration with the distance offshore can be seen. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) J.M. L opez-Guti errez, D.C. Louw, M. Ro zmari c et al. Chemosphere 265 (2021) 129065 5 The role of 129 I as a marine tracer is demonstrated in the case of the Namibian shelf which is part of the nBUS, where coastal upwelling occurs regularly and water is transported offshore due to Ekman transport. The 129 I concentrations in surface water tends to increase with distance from coastlines, which reflects this upwelling effect. Furthermore, the 129 I concentration distribution with latitude in Central Waters might be related to the intrusion of hypoxic water of tropical origin. Credit author statement J.M. L opez-Guti errez: Conceptualization, Validation, Formal analysis, Investigation, Writing eoriginal draft, Writing ereview & editing, Visualization, Supervision, Project administration, Funding acquisition. Martina Ro zmari c: Conceptualization, Validation, Investigation, Resources, Writing ereview &editing, Project administration, Funding acquisition. Deon C. Louw: Conceptualization, Validation, Investigation, Resources, Writing ereview & editing, Project administration, Funding acquisition. V. L erida-Toro: Investigation, Resources, Writing ereview &editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements This research has been partially financed by the project PGC2018-094546-B-I00 by the Spanish Ministerio de Ciencia, Innovaci on y Universidades. The authors thank Prof. X. Hou for his collaboration in sending the data compilation included in Fig. 2.We thank the Namibian Ministry of Fisheries and Marine Resources (MFMR) for its funded support for regular surveys, as well as all colleagues that were involved in the data collection, particularly the staff of the Environment Subdivision. Sampling was done on the Namibian RV Mirabilis and captain and crew members from the MFMR are also thanked for their support. The authors want to thank to Rafael García-Tenorio and Iolanda Osvath for their guidance on sampling and analytical strategies, The IAEA is grateful to the Government of the Principality of Monaco for the support provided to its Environment Laboratories. References Aldahan, A., Alfimov, V., Possnert, G., 2007. 129I anthropogenic budget: major sources and sinks. Appl. Geochem. 22, 606e618. https://doi.org/10.1016/ j.apgeochem.2006.12.006. Boyer, D., Cole, J., Bartholomae, C., 2000. Southwestern Africa: northern Benguela current region. Mar. Pollut. Bull. 41, 123e140 https://doi.org/https://doi.org/ 10.1016/S0025-326X(00)00106-5. Calvo, E.C., Santos, F.J., L opez-Guti errez, J.M., Padilla, S., García-Le on, M., Fig. 4. Physico-chemical parameters of the samples in this work and association to the different water masses present in this region described by Duncombe Rae (2005). Plot made with Ocean Data View (Reiner and Schlitzer, 2020). Samples 10s, 11s and 16s are not included in the plot as their salinity and temperature data were not available. Fig. 5. 129 I concentrations in seawater samples from Namibia classified as Central Waters (CW) as a function of latitude. Fig. 6. 129 I/ 236 U ratios as a function of 129 I concentration for samples classified as Central Waters (CW). J.M. L opez-Guti errez, D.C. Louw, M. Ro zmari c et al. Chemosphere 265 (2021) 129065 6 Heinemeier, J., Schnabel, C., Scognamiglio, G., 2015. Status report of the 1 MV AMS facility at the Centro Nacional de Aceleradores. Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 361, 13e19. https://doi.org/ 10.1016/j.nimb.2015.02.022. Carr, M.-E., Kearns, E.J., 2003. Production regimes in four eastern boundary current systems. Deep Sea Res. Part II Top. Stud. Oceanogr. 50, 3199e3221 https:// doi.org/https://doi.org/10.1016/j.dsr2.2003.07.015. Casacuberta, N., Christl, M., Vockenhuber, C., Wefing, A.M., Wacker, L., Masqu e, P., Synal, H.A., Rutgers van der Loeff, M., 2018. Tracing the three Atlantic branches entering the arctic Ocean with 129I and 236U. J. Geophys. Res. Ocean. 123, 6909e6921. https://doi.org/10.1029/2018JC014168. Casacuberta, N., Masqu e, P., Henderson, G., Rutgers van-der-Loeff, M., Bauch, D., Vockenhuber, C., Daraoui, A., Walther, C., Synal, H.-A., Christl, M., 2016. First 236U data from the Arctic Ocean and use of 236U/238U and 129I/236U as a new dual tracer. Earth Planet Sci. Lett. 440, 127e134. https://doi.org/10.1016/ J.EPSL.2016.02.020. Castrillejo, M., Casacuberta, N., Christl, M., Garcia-Orellana, J., Vockenhuber, C., Synal, H.A., Masqu e, P., 2017. Anthropogenic 236U and 129I in the Mediterranean Sea: first comprehensive distribution and constrain of their sources. Sci. Total Environ. 593 (594), 745e759. https://doi.org/10.1016/ j.scitotenv.2017.03.201. Castrillejo, M., Casacuberta, N., Christl, M., Vockenhuber, C., Synal, H.A., García- Ib a~ nez, M.I., Lherminier, P., Sarthou, G., Garcia-Orellana, J., Masqu e, P., 2018. Tracing water masses with 129I and 236U in the subpolar North Atlantic along the GEOTRACES GA01 section. Biogeosci. Discuss. 1e28 https://doi.org/10.5194/ bg-2018-228. Christl, M., Casacuberta, N., Lachner, J., Maxeiner, S., Vockenhuber, C., Synal, H.A., Goroncy, I., Herrmann, J., Daraoui, A., Walther, C., Michel, R., 2015. Status of 236U analyses at ETH Zurich and the distribution of 236U and 129I in the north sea in 2009. Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 361, 510e516. https://doi.org/10.1016/j.nimb.2015.01.005. Duncombe Rae, C., 2005. A demonstration of the hydrographic partition of the Benguela upwelling ecosystem at 26  40’S. Afr. J. Mar. Sci. 27, 617e628. https:// doi.org/10.2989/18142320509504122. Edmonds, H.N., Smith, J.N., Livingston, H.D., Kilius, L.R., Edmond, J.M., 1998. 129I in archived seawater samples. Deep. Res. Part I Oceanogr. Res. Pap. 45, 1111e1125. https://doi.org/10.1016/S0967-0637(98)00007-7. G omez-Guzm an, J.M., L opez-Guti errez, J.M., García-Tenorio, R., Agull o, L., Peruchena, J.I., Manj on, G., García-Le on, M., 2017. Estimating the impact from fukushima in southern Spain by 131I and accelerator mass Spectrometry detection of 129I. J. Environ. Radioact. 166, 36e44. https://doi.org/10.1016/ J.JENVRAD.2016.03.008. He, P., Aldahan, A., Possnert, G., Hou, X.L., 2013. A summary of global 129I in marine waters. Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 294, 537e541. https://doi.org/10.1016/J.NIMB.2012.08.036. Hou, X., 2004. Application of 129I as an environmental tracer. J. Radioanal. Nucl. Chem. 262, 67e75. https://doi.org/10.1023/B:JRNC.0000040855.91732.9f. Hou, X., Aldahan, A.P., Nielsen, S., Possnert, G., Nies, H., Hedfors, J., 2007. Speciation of 129I and 127I in seawater and implications for sources and transport pathways in the North Sea. Environ. Sci. Technol. 41, 5993e5999. https://doi.org/ 10.1021/es070575x. Junker, T., Mohrholz, V., Siegfried, L., van der Plas, A., 2017. Seasonal to interannual variability of water mass characteristics and currents on the Namibian shelf. J. Mar. Syst. 165, 36e46 https://doi.org/https://doi.org/10.1016/ j.jmarsys.2016.09.003. L opez-Lora, M., Chamizo, E., Ro zmari c, M., Louw, D.C., 2020. Presence of 236U and 237Np in a marine ecosystem: the northern Benguela Upwelling System, a case study. Sci. Total Environ. 708 https://doi.org/10.1016/j.scitotenv.2019.135222. Mohrholz, V., Bartholomae, C.H., van der Plas, A.K., Lass, H.U., 2008. The seasonal variability of the northern Benguela undercurrent and its relation to the oxygen budget on the shelf. Continent. Shelf Res. 28, 424e441. https://doi.org/10.1016/ j.csr.2007.10.001. Monteiro, P.M.S., van der Plas, A.K., M elice, J.L., Florenchie, P., 2008. Interannual hypoxia variability in a coastal upwelling system: ocean-shelf exchange, climate and ecosystem-state implications. Deep. Res. Part I Oceanogr. Res. Pap. 55, 435e450. https://doi.org/10.1016/j.dsr.2007.12.010. Muramatsu, Y., Takada, Y., Matsuzaki, H., Yoshida, S., 2008. AMS analysis of 129I in Japanese soil samples collected from background areas far from nuclear facilities. Quat. Geochronol. 3, 291e297 https://doi.org/https://doi.org/10.1016/ j.quageo.2007.08.002. Negri, A.E., Fern andez Niello, J.O., Wallner, A., Arazi, A., Fifield, L.K., Tims, S.G., 2013. 129I dispersion in Argentina: concentrations in fresh and marine water and deposition fluences in Patagonia. Environ. Sci. Technol. 47, 9693e9698. https:// doi.org/10.1021/es400610h. NIST Office of Reference Materials, n.d. SRM 3231 - Iodine-129 Isotopic Standard (High Level) [WWW Document]. URL. https://www-s.nist.gov/srmors/view_ detail.cfm?srm¼3231. Paul, M., Fink), D., Hollos, G., Kaufman, A., Kutschera, W., Magaritz, M., 1987. Measurement of 129I concentrations in the environment after the Chernobyl reactor accident. Nucl. Instrum. Methods Phys. Res. B 29, 341e345. https:// doi.org/10.1016/0168-583X(87)90262-X. Povinec, P.P., Breier, R., Coppola, L., Groening, M., Jeandel, C., Jull, A.J.T., Kieser, W.E., Lee, S.H., Liong Wee Kwong, L., Morgenstern, U., Park, Y.H., Top, Z., 2011. Tracing of water masses using a multi isotope approach in the southern Indian Ocean. Earth Planet Sci. Lett. 302, 14e26. https://doi.org/10.1016/j.epsl.2010.11.026. Raisbeck, G.M., Yiou, F., 1999. 129I in the oceans: origins and applications. Sci. Total Environ. 237e238, 31e41. https://doi.org/10.1016/S0048-9697(99)00122-9. Raisbeck, G.M., Yiou, F., Zhou, Z.Q., Kilius, L.R., 1995. 129I from nuclear fuel reprocessing facilities at Sellafield (U.K.) and La Hague (France); potential as an oceanographie tracer. J. Mar. Syst. 6, 561e570. https://doi.org/10.1016/0924- 7963(95)00024-J. Reiner and Schlitzer, 2020. Ocean Data View. Reithmeier, H., 2005. Umweltproben als tracer für die atomsph€ arishen 131I-Frei- sentzungen un Majak. Tech. Univ. München, p. 129I. Reithmeier, H., Lazarev, V., Rühm, W., Nolte, E., 2010. Anthropogenic 129I in the atmosphere: overview over major sources, transport processes and deposition pattern. Sci. Total Environ. 408, 5052e5064. https://doi.org/10.1016/ j.scitotenv.2010.07.015. Shannon, L.V., Nelson, G., 1996. The Benguela: large scale features and processes and system variability. In: Wefer, G., Berger, W.H., Siedler, G., Webb, D.J. (Eds.), The South Atlantic. Springer International Publishing, Berlin, pp. 163e210. https://doi.org/10.1007/978-3-642-80353-6_9. Snyder, G., Aldahan, A., Possnert, G., 2010. Global distribution and long-term fate of anthropogenic129i in marine and surface water reservoirs. G-cubed 11. https:// doi.org/10.1029/2009GC002910. Snyder, G., Fehn, U., 2004. Global distribution of 129I in rivers and lakes: implications for iodine cycling in surface reservoirs. Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms 223e224, 579e586. https:// doi.org/10.1016/j.nimb.2004.04.107. Villa, M., L opez-Guti errez, J.M., Suh, K.-S., Min, B.-I., Peri a~ nez, R., 2015. The behaviour of 129I released from nuclear fuel reprocessing factories in the North Atlantic Ocean and transport to the Arctic assessed from numerical modelling. Mar. Pollut. Bull. 90, 15e24. https://doi.org/10.1016/j.marpolbul.2014.11.039. Vivo-Vilches, C., L opez-Guti errez, J.M., Peri a~ nez, R., Marcinko, C., Le Moigne, F., McGinnity, P., Peruchena, J.I., Villa-Alfageme, M., 2018. Recent evolution of 129I levels in the Nordic seas and the north Atlantic Ocean. Sci. Total Environ. 621, 376e386. https://doi.org/10.1016/J.SCITOTENV.2017.11.268. Vockenhuber, C., Casacuberta, N., Christl, M., Synal, H.-A., 2015. Accelerator mass Spectrometry of 129I towards its lower limits. Nucl. Instrum. Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 361, 445e449. https://doi.org/10.1016/ J.NIMB.2015.01.061. Xing, S., Hou, X., Aldahan, A., Possnert, G., Shi, K., Yi, P., Zhou, W., 2017. Water circulation and marine environment in the antarctic traced by speciation of 129I and 127I. Sci. Rep. 7, 1e9. https://doi.org/10.1038/s41598-017-07765-w. Xing, S., Hou, X., Aldahan, A., Possnert, G., Shi, K., Yi, P., Zhou, W., 2015. Iodine-129 in snow and seawater in the antarctic: level and source. Environ. Sci. Technol. 49, 6691e6700. https://doi.org/10.1021/acs.est.5b01234. J.M. L opez-Guti errez, D.C. Louw, M. Ro zmari c et al. Chemosphere 265 (2021) 129065 7