Seasonal distributions and fluxes of 210 Pb and 210 Po in the Northern South China Sea

Introduction Conclusions References


Introduction
Within the context of carbon sequestration and elemental cycling, material conveyance through particle settling from the surface layer to the deep ocean is an important pathway of elemental removal in the ocean.In view of the important role played by particulate matter, information obtained from the temporal and spatial distributions of particlereactive radionuclides can provide insight into particle dynamics in the ocean (Cochran and Masque, 2003) The distributions of 210 Pb (t 1/2 = 22.3 yr) and 210 Po (t 1/2 = 138 d) have been used extensively by marine chemists to determine the removal rates of particles and associated elements from the ocean (Bacon et al., 1976;Nozaki et al., 1990).Although both 210 Pb and 210 Po are particle-reactive, there exist subtle differences in the geochemical behaviour of the two radionuclides in the ocean.For example, 210  tends to be scavenged by inorganic particles, whereas polonium has a higher affinity to biogenic particles (Cherry et al., 1975;Fisher et al., 1988), which results in a shorter residence time of 210 Po compared to 210 Pb in the surface ocean (Bacon et al., 1976).
In addition, atmospheric input via aerosol deposition is the dominant source for 210 Pb in the upper layer of the ocean, whereas an atmospheric source is almost negligible for 210 Po (Turekian et al., 1977).
Known as a semienclosed marginal sea, the South China Sea is an oligotrophic system under the influence of monsoonal modulation (Tseng et al., 2005).Besides sparse data in the surface water (Nozaki et al., 1998;Wei et al., 2015;Yang et al., 2006), vertical profiles of 210 Pb and 210 Po in the South China Sea have been reported by Obata et al. (2004) and Chung and Wu (2005).Obata et al. (2004) measured 210 Pb and 210 Po activities in unfiltered seawater, whereas Chung and Wu (2005) measured dissolved and particulate 210 Pb and 210 Po.When a comparison is made, it is surprising to note that a large difference exists between the two data sets.Not only systematic discrepancies of 210 Pb and 210 Po concentrations, but also large differences of the degrees of disequilibria of 210 Pb/ 226 Ra and 210 Po/ 210 Pb were found between the two studies.Obata et al. (2004) found the deviation of 210 Po from secular equilibrium in the whole water column to be small.On the contrary, a large deficiency of 210 Po relative to 210 Pb was reported by Chung and Wu (2005) and led to their conclusion that 210 Po is removed in an unorthodox way quickly by picoplankton uptake then transferred to the higher trophic level in the northern South China Sea.In order to determine the correct geochemical processes controlling the behavior of 210 Pb and 210 Po, it is necessary to verify the distributions of 210 Pb and 210 Po in the South China Sea. of 210 Pb and 210 Po between the two studies by Obata et al. (2004) and Chung and Wu (2005), (2) to compare the geochemical behavior of 210 Pb and 210 Po, (3) to investigate seasonal variations of the scavenging process, and (4) to compare the trap-measured fluxes of 210 Pb and 210 Po estimated from 210 Pb/ 226 Ra and 210 Po/ 210 Pb disequilibria, respectively, in the water column of the northern South China Sea.
2 Materials and methods

Seawater
Seawater samples were collected from four cruises to SEATS (Fig. 1 subsamples of 210 Pb were stored for at least 1 year to let 210 Po grow in from 210 Pb and the same procedures for dissolved and particulate 210 Po were followed.

Sinking particles
The sinking particles were collected by floating traps in the upper water column, and sinking particles at depths of 700, 1000, 3000, and 3500 m were collected by moored traps (Technicap PPS 5/2; 1 m 2 collecting area).Selected samples were analyzed for 210 Po content by the same procedures as the particulate samples described in the previous section.Separate sets of filters loaded with sinking particles were stored for at least one year to let 210 Po grow in from 210 Pb and the same procedures for particulate 210 Po were followed.
The silver discs were counted by alpha spectrometry (EG&G Ortec 576).After appropriate ingrowth corrections, the 210  It is noted that, except for the ORI-845 profile, 210 Pb p generally increases with depth, similar to findings in the open ocean (Craig et al., 1973;Somayajulu and Craig, 1976 at 3000 m.

Literature data comparison
There are two published 210 Po and 210 Pb profiles near the SEATS station (Fig. 1).
Vertical profiles of 210 Po t and 210 Pb t were reported by Obata et al. (2004) and two sets of dissolved and particulate profiles of 210 Pb and 210 Po measured in different years were reported by Chung and Wu (2005) in the South China Sea.To compare the consistency of the data, vertical profiles of 210 Po t and 210 Pb t redrawn from the two studies and the average values of the four cruises from this study are shown in Fig. 3.It is noted that the sampling stations of these studies were within 300 km of each other in the semi-enclosed basin of the South China Sea (Fig. 1); an unreasonably large discrepancy exists between the data reported by Chung and Wu (2005) Chung and Wu (2005) are higher than those of the other two datasets.Not only were systematically higher 210 Po t and 210 Pb t but also a deficiency of 210 Po relative to 210 Pb, shown as the shaded area in Fig. 3, was found in Chung and Wu (2005).In fact, like many profiles reported in the literature (e.g., Bacon et al., 1976;Sarin et al., 1994), both Obata et al. (2004) and our data showed an excess of 210 Po in the layer underlying the euphotic layer.In addition to the vertical profiles, it is also noted that the  Chung and Wu (2005) comes from comparison with data from nearby stations in Luzon Strait, which is reported by Chen and Chung (1997) (Fig. 1).We have redrawn 210 Pb t and 210 Po t from stations W3 and W5 of the former study and stations 3 and 7 of the latter study in Fig. 5.A systematic discrepancy was apparently found between the two data sets.Based on comparison of data from the same depth, 210 Pb t and 210 Po t from Chung and Wu (2005) are higher by 10-15 dpm (100 L) −1 and 5-10 dpm (100 L) −1 , respectively, than that of Chen and Chung (1997).Consequently, the systematic error results in an unrealistically large 210 Po deficiency in the deep basin of the South China Sea, which leads to the questionable conclusion of Chung and Wu (2005): "Most of the missing 210 Po in the upper layer has probably been consumed by bacteria or cyanobacteria and transferred to higher trophic levels of organisms in the food web".(Bacon et al., 1976), the average 210 Pb t / 226 Ra ratio decreases dramatically in the thermocline then maintains a constant value of 0.5 below 1000 m.Compared to the values summarized in the compilation of Nozaki et al. (1997), the 210 Pb t / 226 Ra ratio is higher than that found in the Bay of Bengal (Cochran et al., 1983;Sarin et al., 1994), the Bismarck Sea (Nozaki et al., 1997) and the Sea of Japan (Nozaki et al., 1973) but lower than in the deep water of the North Pacific (Craig et al., 1973;Nozaki et al., 1980) and the North Atlantic (Bacon et al., 1976).Minor differences of the ratio seem to exist among cruises, however, the variation is too small to reveals apparent temporal variation.The 210 Pb t / 226 Ra ratio showed a decreasing trend toward sediments in the bottom layer, indicating intensified scavenging at the water-sediment interface (Spencer et al., 1980).This boundary-scavenging phenomenon was also reported in the North Pacific (Nozaki et al., 1980), the Indian Ocean (Cochran et al., 1983), and the continental slope of the South Atlantic Bight (Bacon et al., 1988).Vertical contrary, in the water column deeper than 1000 m, a relatively constant 210 Po t / 210 Pb t ratio of 0.75 was found, which is similar to the ratio found in the Santa Monica Basin (0.72, Wong et al., 1992), the Okinawa Trough (0.71, Nozaki et al., 1990), the eastern South Pacific (0.80, Thomson and Turekian, 1976), and the Sargasso Sea (0.70, Kim and Church, 2001).Recent results of an inter-calibration project also showed a signif-  et al., 1976;Cochran et al., 1983;Nozaki et al., 1997).
The inventories of dissolved and particulate 210 Pb and 210 Po in the water column above the four depths from the four cruises are presented in Table 2.The average inventories of total 210 Pb and total 210 Po in the upper 1000 m are 85.5 × 10 3 dpm m −2 and 64.7×10 3 dpm m −2 , respectively, which result in an integrated 210 Po deficit of 20.7× 10 3 dpm m −2 .Significantly higher 210 Po deficits were found in the Philippine Sea, 86 × 10 3 dpm m −2 , and in the Okinawa Trough, 118 × 10 3 dpm m −2 , which was attributed to the focusing effect of 210 Pb accumulation by Kuroshio transport (Nozaki et al., 1990).
In the oligotrophic Sargasso Sea, Kim (2001)   in the interaction with particles in the ocean.It is known that 210 Po shows a higher affinity to organic particles whereas 210 Pb tends to associate with inorganic particles (Bacon et al., 1976;Shannon et al., 1970).A shorter residence time is found for 210 Po compared to 210 Pb in the surface ocean (Nozaki et al., 1998), which reflects the relative intensity of interaction between the two radionuclides and particulate matter.The vertical profiles (Fig. 2) showed that, except in the subsurface water underlying the euphotic layer, 210 Po is lower than 210 Pb in the dissolved phase whereas the opposite was found in the particulate phase, leading to a higher percentage of 210   , 1976;Wei and Murray, 1994).The two coefficients can be calculated by The vertical profiles of K d (Pb) and K d (Po) are shown in Fig. 7.The K d values fall in the range of 10 5.0 -10 6.8 mL g −1 and 10 4.7 -10 7.2 mL g −1 for 210 Pb and 210 Po, respectively.
Based on the average value of K d , 210 Po shows a higher affinity to particles, which is consistent with the results in the western North Atlantic (Bacon et al., 1988), the Equatorial Pacific (Murray et al., 2005), and the Arabian Sea (Sarin et al., 1994).A broad subsurface minimum of K d (Po) underlying the euphotic layer was found, which is similar to the findings in the Sea of Japan ( 2008).The minimum can be ascribed to the process of particle decomposition and remineralization (Bacon et al., 1976;Hong et al., 2008).The fractionation factor, F Po/Pb , was used by Bacon et al. (1988) to compare the relative affinity of 210 Po and 210 Pb to the particles.F Po/Pb is calculated by loads (Wei et al., 2012), an inverse relationship between K d and the concentration of TSM are evident in the open ocean.The negative correlation between the K d s and the TSM indicates the partitioning between dissolved and particulate phases in the deep ocean is controlled by particle-particle interactions, which was proposed by Honeyman et al. (1988) and Honeyman and Santschi (1989) based on 234 Th data in the ocean.In the context of the Browning pumping model (Honeyman and Santschi, 1989), scavenging of particle-reactive elements in the deep ocean that are characterized by extremely low particle concentrations is controlled by slow particle-particle interactions and the slope of the logTSM-logKd correlation reflects the fraction of elements associated with colloids in filterable pool.and Wu (2005) found that the 210 Po/ 210 Pb in suspended particles collected from the euphotic layer at SEATS ranges from 0.66 to 0.77.

Sinking fluxes of 210 Pb and 210 Po
There are two major sources of 210 Pb in the ocean: radioactive decay of 226 Ra in seawater and atmospheric 210 Pb input via dry and wet deposition into the surface layer (Bacon et al., 1976).Hence, assuming a steady state, the removal flux of 210 Pb via particle settling can be estimated by  In most studies (e.g., Bacon et al., 1976;Hong et al., 2008;Nozaki et al., 1990;Obata et al., 2004), steady-state was generally assumed to estimate the scavenging flux (J Po-210 ) and removal flux (F Po-210 ) of 210 Po in the deep ocean by where λ 2 is decay constant of 210 Po (= 0.005 d −1 ). reported by Kim and Church (2001) in the upper 500 m of the Sargasso Sea, no systematic time-series of 210 Pb and 210 Po data had been available to evaluate the effect of temporal variation on particle removal in the deep basins of the ocean.Following the formulation of Kim and Church (2001), neglecting atmospheric input and taking into account temporal variation, the equations for the scavenging flux and removal flux of 210 Po are given by where the summation symbol represents the inventory of radionuclides and ∆t is the interval between consecutive sampling times designated by superscripts t 1 and t 2 The results of scavenging and removal fluxes calculated by steady-state (SS) and nonsteady-state (NSS) at four depths were shown in Fig. 8, in which the uncertainty was estimated based on the propagation of error.There were four sampling periods available for the estimation of the sinking flux of 210 Po at 1000 m, whereas only three sets 2. seasonal variation of J Po-210 and F Po-210 fluxes at 1000 m was found, implying the scavenging and removal of 210 Po may be controlled by production and decomposition of biological particles in the euphotic layer, and 3. a general increasing trend with depth was found for both J Po-210 and F Po-210 The largest increase rate was found between 1000 m and 2000 m, indicating 210 Po is continuously scavenged while particles sink through the water column.
One independent method with which to check the validity of the model estimates is to compare the flux estimates with directly measured sinking fluxes from sediment traps.Although limited 210 Po measurements were made on the sinking particles collected by the sediment trap, the data listed in Table 1 provide an independent check of the sinking fluxes of 210 Po.The F Po-210 calculated by Eq. ( 8) is 412-470 dpm m −2 d −1 at 3000 m and 481-567 dpm m −2 d −1 at 3500 m, which is much higher than the measured flux by more than one order of magnitude (Table 1).In the Santa Monica Basin The residence time of 210 Pb with respect to the particle removal rate in the deep ocean shows a large range, from 2-3 yr in anoxic basins to 300 yr in the central gyre of the Pacific (Nozaki et al., 1997).The residence time of 210 Pb, τ Pb , with respect to particle removal was calculated by dividing the inventories of 210 Pb t by the F Pb-210 from Eq (4).
The τ Pb ranges from 12 to 17 yr in the deep layer of the South China Sea, which is consistent with the value estimated by Obata et al. (2004).The τ Pb in the South China Sea is also comparable with the values reported in other marginal seas of the western North Pacific like the East China Sea and the Sea of Japan, in which an average residence time of 15 yr was estimated (Nozaki et al., 1990(Nozaki et al., , 1973)).In the deep layer of the Bismarck Sea and the Bay of Bengal, a very short τ Pb of 8 yr was reported (Cochran et al., 1983;Nozaki et al., 1997).A shorter residence time of 210 Pb in marginal seas than in the open ocean, which is about 50-300 yr (Bacon et al., 1976;Chung and Craig, 1983;Craig et al., 1973;Nozaki and Tsunogai, 1976;Nozaki et al., 1997), demonstrates the boundary scavenging phenomenon incurred by enhanced particle removal in the regions near land masses.It is noted that the τ Pb in the deep basin is shorter than the mixing time required for the water exchange between the western Pacific and the South China Sea.Residence times of 100 yr based on 14 C tracer (Broecker et al., 1986) and 30-71 yr based on deep water transport (Chang et al., 2010) were estimated for the seawater in the South China Sea.Shorter τ Pb than the residence time of water implies a lower 210 Pb concentration in the South China Sea than in the western Philippine Sea.
Indeed, compared with data from the eastern Luzon Strait, from which the deep water of the South China Sea originates (Gong et al., 1992), our unpublished data showed that the 210 Pb t is about 5-10 dpm (100 L) −1 lower in the deep water of the South China Sea.
Similarly, the residence time of 210 Po, τ Po , with respect to particle removal was calculated by the dividing the inventories of 210 Po t by the F Po-210 from Eq. ( 6) and Eq. ( 8), based on the SS and NSS models, respectively.Since the temporal variation of 210 Po Introduction

Conclusions References
Tables Figures

Back Close
Full inventory in the water column is not large enough to significantly affect the F Po-210 calculation, the τ Po calculated by either the SS or NSS models showed no significant difference.The residence time of 210 Po ranged from 0.9 to 2.2 yr for the upper 1000 m and from 1.2 to 1.7 yr for the upper 3000 m of the water column.Except in regions of high productivity, the disequilibrium of 210 Po and 210 Pb is not evident in the deep layer of the open ocean, hence, a longer residence time of 210 Po with respect to particle removal, 2-4 yr, was estimated (Bacon et al., 1976;Thomson and Turekian, 1976).Even in marginal seas, a relatively long residence time of 210 Po was found in deep water.
For example, Hong et al. (2008)  China Sea.We speculate the enhancement is related to the episodic settling of coccolithophore ballast, which may serve as an efficient carrier of organic particles and associated 210 Po.

Conclusions
This study reported the vertical profiles of dissolved and particulate

Time
Depth Full  Full  Full Discussion Paper | Discussion Paper | Discussion Paper | to estimate the export fluxes of carbon from the euphotic layer of the SEATS time-series station.Here we emphasize the 210 Pb and 210 Po geochemistry in the deep basin of the South China Sea.In addition to water column data, sinking fluxes of 210 Pb and 210 Po in the deep basin were determined by data from sediment traps.The goals of this study are: (1) to resolve the discrepancy between vertical distributions Discussion Paper | Discussion Paper | Discussion Paper | ), onboard the R/V Ocean Researcher I(ORI-821: 12-19 January 2007, ORI-845: 21-30 October 2007,  ORI-866: 28 May-3 June 2008)  and R/V Ocean Researcher III (ORIII-1239: 28 July-3 August 2007).A CTD/20 L Go-Flo system was used to collect large volumes of seawater for 210 Pb and 210 Po determinations.At each sampling depth, 20 L seawater was collected and divided into two 10 L subsamples to determine 210 Pb and 210 Po, respectively.Seawater was immediately pressure-filtered with compressed air through a preweighed 142 mm Nuclepore filter (0.45 µm) mounted in a Plexiglas filter holder.The analytical procedures described inWei et al. (2009) were followed.In short, the filtrate from the 210 Po sample was acidified with about 10 mL concentrated HCl and spiked with 2.2 dpm of 209 Po and 30 mg of Fe carrier.Given 2 days' isotopic equilibration time, concentrated NH 4 OH was then added to raise the pH to approximately 8 to precipitate Fe(OH) 3 .The Fe(OH) 3 precipitate was collected by decanting and centrifuging, was dissolved in HCl, digested with HNO 3 , and then 210 Po and 209 Po were spontaneously plated onto silver plates(Flynn, 1968).The particulate samples collected on the Nuclepore filters were dried in a desiccator and weighed to estimate the concentration of total suspended matter.The filter was than decomposed and digested by a mixture of HCl/HF/HNO 3 /HClO 4 .The same procedures as for dissolved samples were carried out to plate 210 Po and 209 Po onto silver plates.The filtered seawater and filter from the 10 L Figures Back Close Full Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Pb and 210 Po activities in seawater and trap samples at sampling time were calculated 3 Results Data for total suspended matter (TSM) concentration, dissolved and particulate 210 Pb (denoted by 210 Pb d and 210 Pb p , respectively), and dissolved and particulate 210 Po (denoted by 210 Po d and 210 Po p , respectively) are given in Appendix A. Vertical profiles of TSM concentration measured from the four cruises are shown in Fig. 2a, d, g, and k.The TSM falls in a small range of 0.1-0.5 mg L −1 throughout the water column.Vertical distributions of 210 Pb d and 210 Pb p measured from the four cruises are shown in Fig. 2b, e, h, and l.The total 210 Pb ( 210 Pb t ) calculated by the summation of 210 Pb d and 210 Pb p is also shown in the figures.Only a small proportion of 210 Pb, with an average fraction of ∼ 5 % in the upper 200 m and ∼ 10 % in the deep layer, is associated with particulates.Discussion Paper | Discussion Paper | Discussion Paper | profiles of the 210 Po t / 210 Pb t ratio of the four cruises to the SEATS station are shown in Fig. 6b.The general pattern of 210 Po distribution shows a deficiency in the surface layer and an excess in the thermocline, similar to the phenomenon observed in the open ocean (Bacon et al., 1976).It can be seen that the deficiency of 210 Po in the thermocline varies greatly among cruises, indicating temporal variation of 210 Po regeneration caused by particle decomposition in the twilight zone of the water column.On the Discussion Paper | Discussion Paper | Discussion Paper | also found a much higher 210 Po deficit of 56 × 10 3 dpm m −2 due to an unexpectedly high 210 Pb inventory of 135 × 10 3 dpm m −2 in the upper 1000 m, which was attributed to the process of 210 Po being transported by macrozooplankton instead of removed by particle settling.On the contrary, the 210 Po deficit in the South China Sea is higher than those found in the eastern South Pacific (5 × 10 3 -15 × 10 3 dpm m −2 , (Thomson and Turekian, 1976) and comparable with those found in the western Equatorial Pacific (26×10 3 -43×10 3 dpm m −2 , Nozaki et al., 1997).A higher deficiency of 210 Po compared to the open ocean implies faster removal by particle settling in the South China Sea.Both the South China Sea and Sea of Japan are semi-enclosed marginal seas with limited exchange with the open ocean; hence, it is interesting to compare the 210 Pb-Discussion Paper | Discussion Paper | Discussion Paper | Po than 210 Pb being associated with particles.The partition coefficients of 210 Pb and 210 Po, K d (Pb) and K d (Po), respectively, have been used as an indicator of the affinities of radionuclides for particulate matter (Discussion Paper | Discussion Paper | Discussion Paper | et al. of F Po/Pb is shown in Fig. 7b.Except in the subsurface layer underlying the euphotic zone in October, F Po/Pb is greater than unity in the whole water column.There are limited data on the partitioning of 210 Pb and 210 Po in the dissolved and particulate phases for calculation of K d (Pb) and K d (Po) in the deep basin of the open ocean.The compilation of literature data from the deep ocean on the correlations of K d (Pb) and K d (Po) with TSM are shown in Fig. 8a and b, respectively.Unlike the findings in in the Yellow Sea (Hong et al., 1999) and coastal waters with high particle Discussion Paper | Discussion Paper | Discussion Paper | 4) where F Pb-210 is the removal flux, I Pb-210 is the atmospheric 210 Pb deposition flux, λ 1 is the radioactive decay constant of 210 Pb (= 8.48 × 10 −5 d −1 ), and the summation Discussion Paper | Discussion Paper | Discussion Paper |symbol represents the inventory of the radionuclides.It should be pointed out that the steady-state assumption is justified by the small temporal variation of the 210 Pb inventory in the whole water column of the SEATS site (Table2).The atmospheric 210 Pb flux at the sampling site is not known; however, taking the average value of 0.4 dpm cm −2 a −1 byFeichter et al. (1991) and 0.76 dpm cm −2 yr −1 by Xu et al. (2010), an I Pb-210 of 0.6 dpm cm −2 yr −1 (16.4 dpm m −2 d −1 ) was assumed in the southern South China Sea(Wei et al., 2011).The F Pb-210 estimated from the 210 Pb deficiencies in the water column falls in relatively small ranges, which are 30.5-32.1, 40.3-44.6,and  44.5-51.3dpm m −2 d −1 , above 2000 m, 3000 m, and 3500 m, respectively.The 210 Pb flux measured from the sediment trap samples, which showed an average value of 34.3 dpm m −2 d −1 at 3000 m and 46.3 dpm m −2 d −1 at 3500 m ( However, time-series profiles of dissolved and particulate 210 Pb and 210 Po have enabled us to estimate the scavenging and removal fluxes of 210 Po without assuming steady state.Even in a deep basin like the Sea of Japan, Hong et al. (2008) found that temporal variation of 210 Po profiles may occur.They found that the inventory of total 210 Po may decrease by 15 % during a period of six months at the same site in the deep basin of the Sea of Japan (Hong et al., 2008).Unfortunately, except for the data Discussion Paper | Discussion Paper | Discussion Paper | of profiles were available for the modeling task at depths deeper than 1000 m.Several inferences can be made regarding the scavenging and removal fluxes of 210 Po in the South China Sea: 1. except for the J Po-210 flux at 1000 m in June 2007, both J Po-210 and F Po-210 calculated from the steady-state and the non-steady-state models showed little difference within the uncertainty, indicating a minor effect of temporal variation on the scavenging and removal of 210 Po in the deep layer, Discussion Paper | Discussion Paper | Discussion Paper | , Wong et al. (1992) also found the measured 210 Po flux by sediment traps was significantly lower than the predicted 210 Po flux from the 210 Po/ 210 Pb disequilibrium in the water column.The large discrepancy suggests three possible causes: (1) under-trapping of sediment traps, (2) 210 Po removal by processes other than particle sinking, and (3) episodic particle removals not caught by sediment traps.Since the current velocity in the deep layer (3000 m) of the South China Sea is low (0.5-2 cm s −1 , Wang et al., 2011) and, as discussed previously, 210 Pb fluxes in agreement between trap-measured and modeled fluxes were found, it is very unlikely that the discrepancy of predicted and measured fluxes is caused by under-trapping of the sediment traps.To explain a larger deficiency of 210 Po with respect to 210 Pb in the water column of oligotrophic regimes compared to that in more productive regimes, Kim (2001) proposed a very different view on the mechanism of polonium removal from the ocean, i.e., instead of the conventional view that polonium is removed by settling particles, the large 210 Po deficiency found in oligotrophic ocean was attributed to efficient uptake by cyanobacteria and transfer to higher trophic levels.Based on the questionable data, Chung and Wu (2005) also Discussion Paper | Discussion Paper | Discussion Paper | ascribed the 210 Po removal to trophic transfer in the South China Sea.However, Hong et al. (2013) recently reported results of 210 Po flux data measured by sediment traps in the deep layer of the oligotrophic basin at the BATS station, which was similar to the 210 Po fluxes calculated from the water-column deficiency.In the context of Kim (2001), to balance the amount of 210 Po removed by swimmers, a large quantity of excess 210 Po would exist somewhere in the South China Sea to balance the "missing flux" that was transported away from the SEATS site by nekton via cyanobacterial uptake of 210 Po.This hypothesis can only be tested with more profiles to reveal the spatial variability of the distributions of 210 Po and 210 Pb in the South China Sea.We propose that sporadic wash-out by calcareous ballasts poses a plausible cause of the large discrepancy between the modeled 210 Po flux and the measured flux at the site.Chen et al. (2007) investigated the spatial and temporal variation of coccolithophore biomass and the vertical fluxes of coccolithophores in the northern South China Sea.Induced by nutrient inputs associated with the northeast monsoon, coccolithophore abundance significantly increased during the winter, which resulted in a 6-fold increase in the sinking flux of calcareous coccoliths in comparison with the summer in the deep layer of the SEATS site (Chen et al., 2007).The F Po-210 listed in Table 1 only represented the average flux during a 15 day period, the duration set by the rotary sampler of the sediment trap, during January and July of 2007.It is possible a much higher flux of 210 Po-laden particles was missed.Sherrell et al. (1998) proposed episodic particle sweeping events to explain the temporal variation of suspended particle concentrations in the water column off California.The TSM profiles shown in Fig. 2 display the temporal variation.The inventories of TSM can vary by 30 %, from 23 to 31 g m −2 , in the upper 1000 m and by 20 %, from 817 to 972 g m −2 , in the whole 3500 m water column, depicting sporadic events that strip particles from the water column.Nonetheless, no conclusive proof can be made yet until more flux data are available to demonstrate the temporal variation of 210 Po flux in the deep layer.This result warrants continuous measurements of the sinking flux using sediment traps deployed in the deep layer of the basin to investigate the 210 Po removal by particle settling in the South China Sea.
Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | -series station, SEATS, in the South China Sea.The time-series 210 Pb and 210 Po data allow us to evaluate the importance of temporal variations of particle removal in the deep basin of the South China Sea.Limited data of 210 Pb and 210 Po concentrations in the sinking particles collected by the sediment traps showed that the measured 210 Pb flux is consistent with the removal rate predicted from the 210 Pb-226 Ra deficiency whereas the measured 210 Po flux is significantly lower than the expected removal from the 210 Po-210 Pb deficiency in the water column.Future studies aimed at sediment trap deployments in the deep basin to reveal temporal variability of the 210 Po sinking flux are warranted.A short residence time, 12-17 yr, relative to the particle removal rate was estimated for 210 Pb in the deep Discussion Paper | Discussion Paper | Discussion Paper | basin of the South China Sea.The significantly shorter residence time than that found in the open ocean demonstrates the boundary scavenging effect caused by enhanced scavenging at the water-sediment interface.Discussion Paper | Discussion Paper | Discussion Paper | production in the South China Sea, Biogeosciences, 8, 3793-3808, doi:10.5194/bg-8-3793-2011,2011.Wei, C.-L., Lin, S.-Y., Wen, L.-S., and Sheu, D. D.: Geochemical behavior of 210 Pb and 210 Po in the nearshore waters off western Taiwan, Mar.Pollut.Bull., 64, 214-220, 2012.Wei, C.-L., Chen, P.-R., Lin, S.-Y., Sheu, D. D., Wen, L.-S., and Chou, W.-C.: Distributions of Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper |

Figure 1 .Figure 2 .Figure 3 .
Figure 1.Location of the SEATS station and sites of previous studies in the South China Sea.Bathymetry contours of 200 m and 2000 m are shown as black and gray lines, respectively.Stations of previous studies are: M2 of Chung et al. (2004), PA-11 of Obata et al. (2004), C, W3, W5 of Chung and Wu (2005), 3, 7 of Chen and Chung (1997).

Figure 4 .Figure 5 .Figure 6 .Figure 7 .Figure 8 .
Figure 4. Correlation of 210 Pb t and 210 Po t for the data at station PA-11 of Obata et al. (2004) (solid square), at station C of Chung and Wu (2005) (solid circle), and at station SEATS of this study (open circle).Slopes of 0.5 and 1 are shown as solid lines.
) Vertical distributions of 210 Po d and 210 Po p measured during the four cruises are shown in Fig. 2c, f, i, and m.The total 210 Po ( 210 Po t ) calculated by summing 210 Po d and 210 Po p is also shown in the figures.The 210 Po d is lowest in the mixed layer, increases to and those of the other two studies.While showing a similar vertical structure, our210Po data seems Introduction

4.2 210 Pb/ 226 Ra and 210 Po/ 210 Pb disequilibria
To evaluate the degree of deviation of the secular equilibrium, the SiO 2 -226 Ra correlation, 226 Ra (dpm (100 L) −1 ) = 5.15 + 0.14SiO 2 (µmol L −1 ), which is based on the data determined at station PA-11 by Nozaki and Yamamoto (2001) (Fig. 1), is used to calculate 226 Ra activity.Vertical profiles of the 210 Pb t / 226 Ra ratio obtained from the four cruises to the SEATS station are shown in Fig. 6a.Due to atmospheric input, an excess of 210 Pb relative to 226 Ra in the upper 250 m is generally found.In contrast to the upper water column, 210 Pb shows a large deficiency relative to 226 Ra in the deep water, indicating fast removal of 210 Pb in the deep basin of the South China Sea.Similar to the findings in the open ocean deficiency of 210 Po relative to 210 Pb in the deep water, with a 210 Po/ 210 Pb ratio of 0.74 ± 0.06 at 2000 m at BATS in the North Atlantic and 0.80 ± 0.11 at 3000 m at the baseline station in the North Pacific, respectively (Church et al., 2012).It should be noted that, contrary to the 210 Po deficiencies in these environments, the 210 Po t / 210 Pb t ratio is commonly close to unity in the deep layer (> 1000 m) of the ocean (1988; Bacon

Partitioning of 210 Pb and 210 Po in dissolved and particulate phases
Hong et al. (2008))n the two basins.However, only210Po data is available in the deep water of the Sea of Japan, in which 210 Po inventories ranged from 43.5 × 10 3 to 65.8 × 10 3 dpm m −2 and from 78.2 × 10 3 to 100.6 × 10 3 dpm m −2 in the upper 1000 m and 2000 m, respectively(Hong et al., 2008).Our results of 210 Po inventories in the upper 1000 m of the South China Sea are comparable with those from the Sea of Japan.describedinHonget al. (2008), may be caused by enhanced removal by terrestrial inputs of particles.Unfortunately, no 210 Pb data is available for revealing the degree of 210 Pb-210 Po disequilibrium in the deep water of the Sea of Japan.
Although both are particle-reactive, subtle differences are found for210Po and 210 Pb Nozaki et al. (1998) reported 210 Pb and 210 Po activities in the surface water at three stations in the South China Sea.Compared with other regions, as a result of low biological activity and the terrestrial input of detrital material, Nozaki et al. (1998) found 210 Po is removed at a faster rate than 210 Pb in the South China Sea.Due to this frac- Nozaki et al. (1998)ng particles is only 0.58, lower than that in other regions.Corroborating with the prediction ofNozaki et al. (1998), Chung

Table 1 )
, are generally in agreement with the estimated values.Similar 210 Pb flux predicted from the deficiency of 210 Pb in the water column to the F Pb-210 measured by the sediment trap indicates that removal by particle settling instead of bottom scavenging at the water-sediment interface is the dominant process for 210 Pb removal in the South China Sea.
estimated the τ Po of 2.3-5.5 yr in the Japan Sea and Wong et al. (1992) reported the τ Po of 3.2 yr in the Santa Monica Basin.Compared with literature results found in the open ocean and other marginal seas, the τ Po in deep water is smaller, indicating enhanced 210 Po removal from the water column of the South

Table 1 .
Fluxes of total mass (F mass ), 210 Pb (F Pb-210 ), and 210 Po (F Po-210 ) measured by sediment trap.Standard deviations of radionuclide data are based on propagated counting error (σ).

Table 2 .
Inventories of dissolved and particulate 210 Pb and 210 Po, deficiency of 210 Pb, and deficiency of 210 Po in the water column above four depth levels during the four cruises to the SEATS station.

Table A1 .
Depth, total suspended matter (TSM) concentration, dissolved ( 210 Pb d ) and particulate ( 210 Pb p ) 210 Pb, and dissolved ( 210 Po d ) and particulate ( 210 Po p ) 210 Po measured from different cruises to the SEATS station.Standard deviations of radionuclide data are based on counting error (1σ).