Tracing the Atmospheric Input of Seawater-Dissolvable Pb Based on the Budget of 210Pb in the East Sea (Japan Sea)

In order to determine the atmospheric input of 210Pb and seawater-dissolvable Pb in the East Sea (Japan Sea), we measured the concentrations of total 210Pb and dissolved Pb (<0.2 μm) in seawater and 210Pb and 226Ra in sinking particles. The East Sea is deep (∼3700 m) and enclosed by surrounding continents except for the shallow sills (<150 m). Since the East Sea is located off the East Asian continent under the westerlies, the concentrations of 210Pb and dissolved Pb in this sea are significantly affected by terrestrial sources through the atmosphere. The vertical profiles of total 210Pb and dissolved Pb generally showed a surface maximum and then decreased with depth. The concentrations of dissolved Pb in the surface water were 2 and 3 times higher than those in the North Pacific and North Atlantic Oceans, respectively. Using an independent box model (upper 1000 m or 2000 m), we estimate the atmospheric input of 210Pb to be 1.46 ± 0.25 dpm cm−2 y − 1, which is within the range of published results from the land-based sites (0.44–4.40 dpm cm−2 y − 1) in South Korea, China, and Japan. Based on this flux, the residence time of total 210Pb in the East Sea is calculated to be approximately 7.1 ± 1.6 years, which is twice lower than the previous estimation. Combining the residence time of 210Pb and the inventory of dissolved Pb, the atmospheric input of seawater-dissolvable Pb is estimated to be 0.98 ± 0.28 nmol cm−2 y − 1. This flux is approximately 25% of the Pb flux through the wet deposition (acid-leachable fraction). Thus, our results suggest that the flux and fate of atmospheric Pb in the ocean can be successfully determined using an accurate mass balance model of naturally occurring 210Pb.


INTRODUCTION
The naturally occurring radionuclide 210 Pb (t 1/2 = 22.3 years), belonging to the 238 U decay series, is produced from 226 Ra (t 1/2 = 1600 years), via 222 Rn (t 1/2 = 3.8 days) and other short-lived radionuclides (t 1/2 < 30 min). In the upper ocean, most of the 210 Pb originates from atmospheric deposition, while that in the deep ocean is mainly produced by in situ decay of 226 Ra. Therefore, an excess of 210 Pb over 226 Ra ( 210 Pb ex : 210 Pb-226 Ra) is observed in the upper ocean, except for the polar regions, where the atmospheric input of 210 Pb is small (Elsässer et al., 2011;Persson and Holm, 2014;Baskaran and Krupp, 2021). 210 Pb is rapidly removed by particles in the surface water, so the residence time of 210 Pb in the surface water is short: 1-3 years in the North Pacific and North Atlantic Oceans Tsunogai, 1973, 1976;Bacon et al., 1976;Rigaud et al., 2015). The occurrence of 210 Pb ex decreases toward the continent as enhanced scavenging occurs at the ocean margin (Bacon et al., 1976;Anderson et al., 1994;Seo et al., 2021). 210 Pb has been useful in tracing the behavior of anthropogenic Pb in marine environments. For example, the atmospheric input of Pb was estimated using Pb/ 210 Pb ratios in the rain and the steady-state 210 Pb flux (Settle et al., 1982). The decreasing trend of anthropogenic Pb flux to the Sargasso Sea in response to decline in the emission of United States leaded gasoline was revealed based on the reduced Pb/ 210 Pb ratios in the surface water from 1979 to 1987 (Boyle et al., 1986;Shen and Boyle, 1988;Sherrell et al., 1992). Those studies also combined the Pb/ 210 Pb ratios with 3 H-3 He thermocline ventilation model (Jenkins, 1980) to reveal the importance of isopycnal transport on Pb distributions in that region. Sherrell et al. (1992) also suggested that dissolved Pb and particulate Pb were in equilibrium within the residence time of particulate matter based on the Pb isotope ratios ( 206 Pb/ 207 Pb) and 210 Pb in suspended particulate matter. However, these 210 Pb applications included significant uncertainties since it is difficult to constrain the actual inputs of atmospheric 210 Pb to the specific ocean region. In previous studies, the atmospheric input of 210 Pb has been estimated by sampling the aerosol and/or rain from the land (islands or coastal sites) (e.g., Turekian et al., 1983;Turekian, 1989), sampling the aerosol and/or rain from the ocean via research cruise (e.g., Rengarajan and Sarin, 2004;Niedermiller and Baskaran, 2019), and modeling (e.g., Turekian et al., 1977;Feichter et al., 1991;Balkanski et al., 1993). However, the landbased sampling methods cannot cover vast areas of the ocean and the sampling during cruise covers only a limited time period. The models also suffer from the lack of measured data (Nozaki et al., 1998), although their results with the measured data can be more representative.
The East Sea (Japan Sea) is an enclosed deep marginal sea in the northwestern Pacific Ocean, with a maximum depth of 3700 m and a surface area of 1.0 × 10 6 km 2 . This sea has a deep water formation and a meridional overturning circulation similar to the global ocean (Ichiye, 1984;Gamo, 1999;Kim et al., 2001), although the turnover time of deep water (∼100 years) is shorter than the global circulation time (Watanabe et al., 1991;Tsunogai et al., 1993;Kumamoto et al., 1998). Since the East Sea does not have a large discharge from rivers and is located off the eastern part of the Asian continent, large amounts of lithogenic and anthropogenic elements enter the upper ocean through atmospheric deposition (Park et al., 2006;Jo et al., 2007;Kim et al., 2011;Yan and Kim, 2015;Seo and Kim, 2020). The deep water of the East Sea is almost disconnected from the North Pacific Ocean, except through four shallow sills (<150 m). Thus, the East Sea is an ideal site to study the atmospheric inputs of various elements and their behaviors within a closed system.
In this study, we attempt to estimate the atmospheric input of seawater-dissolvable Pb in the East Sea using the inventory of dissolved Pb in the water column and the residence time of 210 Pb. The residence time of 210 Pb is calculated using the mass balance of the input terms (atmospheric deposition; ingrowth from 226 Ra in seawater) and the output terms (decay of 210 Pb; removal by sinking particles). For this mass balance estimation, we measured the distributions of total 210 Pb in seawater, dissolved Pb in seawater, and settling fluxes of 226 Ra and 210 Pb through 1000 m and 2000 m from sediment trap samples. In addition, we compile previously published 226 Ra and 210 Pb data including the Japan side for more accurate estimation.

Sampling
For the measurements of total 210 Pb in seawater, sampling was conducted from April 6 to May 3 in 2015 on R/V Akademik M.A. Lavrentyev (Figure 1). Seawater samples (10 L, n = 29) were collected in high-density polyethylene (HDPE) bottles from Niskin samplers. The samples were acidified with 12 M HCl (pH < 2) within 1 hour to prevent 210 Pb from sorbing onto the bottles and then stored until analysis.
For the measurements of dissolved Pb in seawater, ultraclean sampling was conducted from January 26 to February 2 in 2018 on R/V Isabu and from October 26 to November 22 in 2019 on R/V Akademik Oparin, respectively (Figure 1). We used PRISTINE ultra-clean CTD (UCC) and Teflon-coated Niskin-X samplers on the R/V Isabu and R/V Akademik Oparin, respectively. All procedures, including cleaning and sampling, followed the GEOTRACES protocol (Cutter et al., 2017). Seawater samples (125 mL, n = 64) were collected in precleaned low-density polyethylene (LDPE) bottles from PRISTINE or Niskin-X samplers through pre-cleaned polyethersulfone capsule filters (0.2-µm pore size; AcroPak-200, Pall). The samples were acidified to pH ∼1.8 within 1 hour after sampling using 12 M HCl (ultra-high pure grade, ODLAB) and stored for laboratory analysis. Milli-Q water (18.2 M , n = 15) was used as a procedural blank.
For 226 Ra and 210 Pb in sinking particles, the conical type sediment traps (MARK7G-21, McLane) were deployed at 1000 m and 2000 m depths from December 1998 to December 1999 (Figure 1). All sample cups were filled with HgCl 2 solutions before deployment to prevent the samples from bacterial degradation. After the recovery of the sediment traps, samples were kept below 4 • C until they were transported to the landbased laboratory for further analysis.

Analytical Methods
The method for measuring the total phase of 210 Pb in seawater in this study was similar to that of previous studies (Kim and Kim, 2012;Seo et al., 2021). Briefly, all seawater samples for total 210 Pb were stored for more than 2 years to allow for the equilibrium between 210 Pb and 210 Po. 209 Po spike (1.5 dpm) and Fe 3+ (50 mg) carrier were added to the samples and left to equilibrate for 24 hours. Ammonium hydroxide was used to adjust the pH to ∼8 for the co-precipitation of 210 Po and Fe(OH) 3 . The supernatants were removed and then the precipitates were filtered. The precipitates were digested with a mixture solution of concentrated HNO 3 and HCl (1:1, v/v) to remove any organic matter in the samples. The mixture solution was dried down and then re-dissolved in 50 mL of 0.5 M HCl. After adding ∼0.5 g of ascorbic acid to reduce Fe 3+ , Po was plated onto a silver disk at a temperature of ∼80 • C with stirring for 3 hours. 210 Po activities on the silver disks were counted by using alpha spectrometry (Alpha Analyst, Canberra). The measured counts were corrected for the background of the alpha spectrometry, decay of 210 Po during counting, recovery of 209 Po spike, decay of 210 Pb from sampling to plating, and the reagent blank (Church et al., 2012). The reagent blank for 210 Pb was 0.0175 ± 0.004 dpm (n = 5), which is comparable with those in Roca-Martí et al. (2021). The blank accounted for 1.1-4.6% (average: 2.7 ± 1.0%, n = 35) of the total 210 Pb in this study.
The concentrations of dissolved Pb were determined using an online pre-concentration system (seaFAST SP3; Elemental Scientific) coupled to a sector-field inductively coupled plasma mass spectrometry (ICP-MS; Element 2, Thermo Fisher Scientific). Approximately 10 mL of sample was buffered to pH ∼6.2 with a 4 M ammonium acetate buffer. The sample was loaded onto the seaFAST column filled with Nobias-chelate PA1 resin (200 µL), subsequently rinsed with a mixed solution of Milli-Q water and buffer to remove the salt. Then, Pb was eluted with 1.6 M HNO 3 (ultra-high pure grade, ODLAB). During the analysis, rhodium was used for an internal standard to correct the changes in ICP-MS sensitivity for each sample. The procedural blank and detection limit of this method was 3.9 pmol kg −1 and 2.8 pmol kg −1 , respectively. The accuracy of the measurement was determined by analyzing the certified reference materials (CASS-6 and NASS-7; National Research Council of Canada) and GEOTRACES reference standards (GSC: bottle number 97 and GSP: bottle number 36).
For 210 Pb and 226 Ra analyses in sinking particles, sediment trap samples were filtered through a 1 mm nylon mesh to separate swimmers and then freeze-dried. The freeze-dried samples were packed into gamma vials and sealed to avoid the loss of 222 Rn. After more than 3 weeks for the secular equilibrium between 226 Ra and its daughter ( 214 Pb and 214 Bi), the activities of 210 Pb and 226 Ra were measured using a gamma counter, with a high-purity germanium well detector (HPGe, Canberra), for the gamma-ray energy of each isotope (46.5 keV for 210 Pb; 351.9 keV for 214 Pb; 609.3 keV for 214 Bi). All analytical results are summarized in Supplementary Tables 1-4.

RESULTS
The vertical profiles of total 210 Pb in the East Sea showed the highest activities in the surface and decreased with depth, as observed in other major non-polar open oceans (Nozaki et al., 1980;Cochran et al., 1990;Kim, 2001;Rigaud et al., 2015;Tang et al., 2018;Horowitz et al., 2020; Figure 2A). The activities of total 210 Pb ranged from 9.3 to 16.4 dpm 100 L −1 (average: 12.8 ± 2.8 dpm 100 L −1 , n = 8) in the surface water (0-100 m) and decreased to a range from 4.2 to 6.6 dpm 100 L −1 (average: 5.0 ± 0.9 dpm 100 L −1 , n = 7) in the deep water (2500-3200 m). The activities of total 210 Pb in the surface water of the East Sea (0-100 m) were comparable with those in the North Pacific and North Atlantic Oceans, whereas the activities of total 210 Pb in the deep East Sea (2000 m) were approximately 4.8 and 2.2 times lower than those in the North Pacific and North Atlantic Oceans, respectively (Nozaki and Tsunogai, 1976;Nozaki et al., 1980;Rigaud et al., 2015).
The vertical profiles of dissolved Pb in the East Sea were similar to those of total 210 Pb, showing the surface maximum and decrease with depth, except at station E1 ( Figure 2B). This distribution pattern differed from those in other open oceans, such as the central North Pacific, North Atlantic, and South Atlantic Oceans, which displayed a sub-surface maximum (Noble et al., 2015;Zurbrick et al., 2017Zurbrick et al., , 2018Schlosser et al., 2019;Zheng et al., 2019). The concentrations of dissolved Pb ranged from 45 to 107 pmol kg −1 (average: 73 ± 15 pmol kg −1 , n = 27) in the surface water (0-100 m) and decreased to a range from 4 to 15 pmol kg −1 (average: 8.5 ± 3.7 pmol kg −1 , n = 6) in the deep water (1500-2200 m). The concentrations of dissolved Pb in the surface water of the East Sea (0-100 m) were approximately 1.8 and 3.2 times higher than those in the North Pacific and North Atlantic Oceans, respectively. On the other hand, the concentrations of dissolved Pb in the deep East Sea (2000 m) were approximately 2.1 and 3.1 times lower than those in the North Pacific and North Atlantic Oceans, respectively (Noble et al., 2015;Zheng et al., 2019;Jiang et al., 2021). At station E1 (water depth: 190 m), the concentrations of Pb were ∼56 pmol kg −1 in the surface (0-100 m) and decreased to ∼26 pmol kg −1 in the sub-surface (100-150 m), followed by an increase to 38 pmol kg −1 near the bottom sediments.

Budget of 210 Pb in the East Sea
The budget of 210 Pb in the East Sea is estimated using two different boxes of a steady-state scavenging model (0-1000 m or 0-2000 m). At steady state (∂A/∂t = 0), by neglecting advection and diffusion, the mass balance of 210 Pb can be calculated as follow: where λ, F Atm, k, and A represent the decay constant of 210 Pb (y −1 ), atmospheric depositional flux of 210 Pb (dpm cm −2 y −1 ), first-order scavenging rate constant (y −1 ), and inventory of each radionuclide (dpm cm −2 ) in the 0-1000 m and 0-2000 m, respectively. For the inventory of total 210 Pb, we compile our measured data with previously published data from the East Sea Kim and Kim, 2012;Seo et al., 2021; this study). The inventory of 226 Ra in the East Sea is from the published results (Harada and Tsunogai, 1986a;Inoue et al., 2015). The first-order scavenging flux (k Pb−210 A Pb−210 ) is measured using the 210 Figure 1). In addition, we note that sediment trap samples of this study were obtained 16 years before the seawater sampling for 210 Pb. However, the activities of 210 Pb in seawater were similar from the 1970s to the 2010s in the East Sea Kim and Kim, 2012;Seo et al., 2021;this study). Thus, we assume that there was no temporal change in 210 Pb in the East Sea over a few decades. In Eq. (1), the only unknown term is the atmospheric input of 210 Pb into the ocean (F Atm ). The atmospheric input of 210 Pb should be balanced by the in situ production from 226 Ra, in situ decay of 210 Pb, and settling flux to the deeper layer (Figure 4). Then, the atmospheric input of 210 Pb is calculated to be 1.37 ± 0.22 and 1.55 ± 0.27 dpm cm −2 y −1 according to the water boxes of 0-1000 m and 0-2000 m, respectively. The results are consistent despite the different depths of both boxes, indicating no significant effect of the lateral transport of total 210 Pb. The atmospheric input of 210 Pb in the East Sea (1.46 ± 0.25 dpm cm −2 y −1 ) is similar to the average of previously reported values (1.64 ± 1.10 dpm cm −2 y −1 ) in this sea based on the land-based measurements or the numerical modeling, which showed large variations depending on the measurement sites and periods (0.44-4.40 dpm cm −2 y −1 ; Tokieda et al., 1996; FIGURE 2 | Vertical profiles of (A) total 210 Pb (dpm 100 L −1 ) and (B) dissolved Pb (pmol kg −1 ) in the East Sea.  Henderson and Maier-Reimer, 2002;Hirose et al., 2004;Yi et al., 2007;Akata et al., 2008;Du et al., 2008Du et al., , 2015Tateda and Iwao, 2008; Table 1). It is reasonable to assume that the atmospheric input of 210 Pb calculated from the 210 Pb budget in the ocean represents the actual atmospheric flux over the annual to decadal time scales. The atmospheric 210 Pb flux in the East Sea is 2-4 times higher than that in the major open oceans, such as the North Pacific (0.22-0.30 dpm cm −2 y −1 ; Turekian, 1989),  equatorial Pacific (0.11-0.51 dpm cm −2 y −1 ; Murray et al., 2005), North Atlantic (0.40-0.69 dpm cm −2 y −1 ; Turekian et al., 1983;Kim et al., 1999), and Indian Oceans (0.73 dpm cm −2 y −1 ; Sarin et al., 1999), respectively. However, the atmospheric input of 210 Pb in the East Sea approaches the upper limit of that in the corresponding latitudinal belt of global fallout curve (average: 0.96 ± 0.58 dpm cm −2 y −1 , 30-40 • N, Baskaran, 2011). This higher flux in the study region is known to be due the elevated emanation of 222 Rn from the Asian continent (Baskaran, 2011;Zhang et al., 2021).
The residence time of 210 Pb in the water column can be calculated using Eq. (2): (2) where τ is the residence time of total 210 Pb. The residence times of total 210 Pb are estimated to be 4.8 ± 1.2 years and 7.1 ± 1.6 years in the 0-1000 m and 0-2000 m, respectively. The calculated result in this study (7 years, 0-2000 m) is approximately 2.3 times lower   is associated with the different atmospheric input term.  assumed that approximately 2.0 dpm cm −2 y −1 of 210 Pb entered the East Sea from the atmosphere, and only a quarter of that (0.5 dpm cm −2 y −1 ) was transported into the deeper layer. It resulted in the net removal flux of 210 Pb at 2000 m to be 0.8 dpm cm −2 y −1 , which is 2.5 times lower than the measured flux from the sediment trap (moored depth: 2000 m) (2.0 dpm cm −2 y −1 ; this study) or sedimentation rates in this region (water depth: ∼2200 m) (1.79 to 2.70 dpm cm −2 y −1 ; Hong et al., 1997;Hong et al., 1999). Thus, our estimated residence time of 210 Pb in the East Sea, which is much shorter than the previous estimation, appears to be more reliable. The residence time of 210 Pb in the East Sea is 1.5-15 times lower than that in the major oceans, such as the North Pacific (54 years, ∼1800-4000 m, Craig et al., 1973;96 years, ∼1000-4000 m, Nozaki andTsunogai, 1976), southeastern Pacific (95 years, ∼0-3700 m, Niedermiller and Baskaran, 2019), North Atlantic (15-22 years, ∼0-3000 m, Cochran et al., 1990), and Indian Oceans (10-15 years, ∼500-4000 m, Obata et al., 2004). The shorter residence time of 210 Pb in the water column of this sea seems to result in the lower concentrations of 210 Pb and Pb in the deep ocean, relative to other major oceans, although their atmospheric input fluxes were higher (e.g., Akata et al., 2008;Sakata and Asakura, 2009;Du et al., 2015). This was also evidenced by the fractionations of rare earth elements (Seo and Kim, 2020). The higher removal rates have been attributed to higher fluxes of sinking particles, which mainly consist of lithogenic materials and opal (>80%) .

Atmospheric Input of Seawater-Dissolvable Pb
In order to calculate the atmospheric input of seawaterdissolvable Pb in the East Sea, we apply the residence time of dissolved 210 Pb in this study. Assuming that dissolved 210 Pb is about 80-90% of total 210 Pb in the East Sea (Kim and Kim, 2012), the residence times of dissolved 210 Pb in this sea are estimated to be 4.0 ± 1.0 years and 5.8 ± 1.3 years in the 0-1000 m and 0-2000 m, respectively. The average annual atmospheric depositional flux of seawater-dissolvable Pb can be obtained by dividing the inventory of dissolved Pb by the residence time of dissolved 210 Pb. The Pb data of station E1 is excluded from this calculation because of the distinctly low concentrations in the 100-150 m layer, which might be due to boundary scavenging. The atmospheric input of seawater-dissolvable Pb is calculated to be 0.98 ± 0.28 nmol cm −2 y −1 . Although Pb can be introduced into the East Sea from the adjacent continental shelf, including the East China Sea and the Yellow Sea, we exclude this source since the concentrations of dissolved Pb (0-100 m) in the southern East Sea (stations S1, S4, E5, and E8; ∼69 pmol kg −1 ) were lower than those in the northern East Sea (stations M9, 40, 41, 134-3, 134-9, 144, and 12; ∼78 pmol kg −1 ). The atmospheric input of seawater-dissolvable Pb in the East Sea is distinctively higher than wet deposition of Pb in the remote oceans, including the North Pacific (0.05-0.08 nmol cm −2 y −1 ; Settle et al., 1982;Duce et al., 1991), North Atlantic (0.03-0.46 nmol cm −2 y −1 ; Duce et al., 1991;Helmers and Schrems, 1995;Kim et al., 1999), and North Indian Oceans (0.01 nmol cm −2 y −1 ; Duce et al., 1991). Our calculated Pb flux is approximately 25% of the previously published fluxes around this region, which used the leaching method with nitric acid for precipitation samples collected on land (Sakata et al., 2006(Sakata et al., , 2008Sakata and Asakura, 2009; Table 2). Diluted nitric or hydrochloric acid has been widely used to desorb Pb from the particles in precipitation. We believe that the leaching method could overestimate the atmospheric depositional flux of seawaterdissolvable Pb since the much lower pH in this process than the actual pH of seawater can affect the solubility of Pb (Chester et al., 2000;Martín-Torre et al., 2015). Our results suggest that the application of 210 Pb provides a useful tool to estimate the flux of actual seawater-dissolvable Pb in the ocean. However, the estimated flux of seawater-dissolvable Pb in this study cannot distinguish the relative contribution of different origins of Pb (leaded gasoline, coal burning, and dust). Thus, future studies are necessary to determine the solubility of atmospheric Pb in the ocean according to its origins.

CONCLUSION
The budget of 210 Pb in the East Sea is determined by measuring the activities of 210 Pb in seawater and sinking particles. Based on the different depths (1000 m or 2000 m) of the scavenging box model, the atmospheric input of 210 Pb is estimated to be 1.46 ± 0.25 dpm cm −2 y −1 . Based on this atmospheric input of 210 Pb, the residence time of 210 Pb in the East Sea (0-2000 m) is calculated to be 7.1 ± 1.6 years, which is an order of magnitude lower than that in the major open oceans due to the efficient Pb removal in the East Sea. Combining this residence time and the concentrations of dissolved Pb, the atmospheric input of seawater-dissolvable Pb is calculated to be 0.98 ± 0.28 nmol cm −2 y −1 , which is ∼25% lower than the previous wet deposition results in this region. Thus, our results suggest that our approach, measuring the flux of seawater dissolvable Pb using the 210 Pb budget in the ocean, can be successfully used for other major oceans.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

AUTHOR CONTRIBUTIONS
GK conceptualized the study. HS, Y-IK, and IK performed the field sampling and analyses. HS and GK interpreted the data and wrote the manuscript. All authors contributed to the final version of the manuscript.