210Po/210Pb Disequilibria and Its Estimate of Particulate Organic Carbon Export Around Prydz Bay, Antarctica

Hu, Huina; Liu, Xiao; Ren, Chunyan; Jia, Renming; Qiu, Yusheng; Zheng, Minfang; Chen, Min

doi:10.3389/fmars.2021.701014

ORIGINAL RESEARCH article

Front. Mar. Sci., 15 July 2021

Sec. Marine Biogeochemistry

Volume 8 - 2021 | https://doi.org/10.3389/fmars.2021.701014

This article is part of the Research Topic Geochemical Cycling of 210Po and 210Pb in Marine Environments View all 12 articles

²¹⁰Po/²¹⁰Pb Disequilibria and Its Estimate of Particulate Organic Carbon Export Around Prydz Bay, Antarctica

$\r\nHuina Hu,$ Huina Hu^1,2

Xiao Liu¹

Chunyan Ren¹

Renming Jia¹

Yusheng Qiu¹

Minfang Zheng¹

Min Chen^1*

¹College of Ocean and Earth Sciences, Xiamen University, Xiamen, China
²South China Sea Environmental Monitoring Center, South China Sea Bureau, Ministry of Natural Resources, Guangzhou, China

Due to the remoteness and difficulty of sampling, the ²¹⁰Po and ²¹⁰Pb data are scarce in the Southern Ocean. Here, the activity concentrations of ²¹⁰Po and ²¹⁰Pb around Prydz Bay in austral summer were determined to understand their spatial variation and evaluate the dynamics of particle organic matter (POM). The activity concentrations of dissolved ²¹⁰Po (D²¹⁰Po) and ²¹⁰Pb (D²¹⁰Pb) range from 0.47 to 3.20 Bq⋅m^–3 and from 1.15 to 2.97 Bq⋅m^–3, respectively, with the lower values in the shelf. The particulate ²¹⁰Po (P²¹⁰Po) and ²¹⁰Pb (P²¹⁰Pb) are lower in the open ocean and increase to the coastal waters, among which the circumpolar deep water (CDW) is the lowest. The activity concentration of total ²¹⁰Pb (T²¹⁰Pb) ranges from 1.26 Bq⋅m^–3 to 3.16 Bq⋅m^–3, with a higher value in CDW, which is ascribed to radiogenic production from ²²⁶Ra and subsequent lateral transport. Occasionally a high value of T²¹⁰Po occurs in deep water (>3.00 Bq⋅m^–3), which may be caused by the remineralization of POM. The disequilibria between T²¹⁰Po and T²¹⁰Pb appears throughout the water column at most stations. The average T²¹⁰Po/T²¹⁰Pb)_A.R. in the euphotic zone is 0.66, reflecting the effect of strong particle scavenging. There is a good positive correlation between the solid-liquid ratio of ²¹⁰Po and POC, while ²¹⁰Pb does not, indicating that particulate organic matter regulates the biogeochemical cycle of ²¹⁰Po around Prydz Bay. Based on the ²¹⁰Po/²¹⁰Pb disequilibria, the export flux of POC in the water column is estimated to be 0.8–31.9 mmol m^–2 d^–1, with the higher values in the shelf.

Introduction

²¹⁰Po (T_1/2 = 138.4 days) and ²¹⁰Pb (T_1/2 = 22.3 years) are radionuclides in ²³⁸U decay chain. Naturally occurring ²¹⁰Po is a β^–-decay product of ²¹⁰Pb via short-lived ²¹⁰Bi (T_1/2 = 5.0 days), and ²¹⁰Pb is produced throughout ²²⁶Ra decay via several short-lived isotopes (²²²Rn, ²¹⁸Po, ²¹⁴Pb, etc.). ²¹⁰Po mainly comes from in-situ decay of ²¹⁰Pb in seawater, and ²¹⁰Pb has three sources: atmospheric deposition, terrestrial runoff, and in situ production via ²²⁶Ra decay (Bacon et al., 1976; Moore and Smith, 1986; Nozaki et al., 1997; Wei et al., 2011; Kaste and Baskaran, 2012). Unlike ²¹⁰Pb mainly being adsorbed to particle surfaces, ²¹⁰Po is additionally assimilated by phytoplankton (Nozaki et al., 1976; Cochran, 1992; Verdeny et al., 2009). The behavior difference results in a depletion of ²¹⁰Po as compared to ²¹⁰Pb in water column, which provides a tool for quantifying the adsorption rate by particles, and export flux on a seasonal-to-decadal timescale (Fisher et al., 1983; Cherrier et al., 1995; Stewart et al., 2005; Rigaud et al., 2014; Tang et al., 2019).

The Southern Ocean divides the polar parts from the warm tropical ocean, including the southern part of the Pacific, Atlantic, and Indian Oceans. Although it represents only 10% of the ocean surface area, it accounts for approximately 25% of the oceanic uptake of atmospheric CO₂ (Takahashi et al., 2002; Arrigo et al., 2008). The Prydz Bay, located in the Indian sector of the Southern Ocean, is the third largest bay in the Antarctica, following the Weddell Sea and the Ross Sea. Previous studies on ²¹⁰Po and ²¹⁰Pb in the Southern Ocean mainly focused on the Antarctic Circumpolar Current (ACC) (Shimmield et al., 1995; Friedrich and Rutgers van der Loeff, 2002), there is still very little research on Prydz Bay. ²¹⁰Po/²¹⁰Pb and ²³⁴Th/²³⁸U disequilibria have been used to estimate POC export flux in the Southern Ocean, but the results obtained by the two methods are sometimes different. The POC export fluxes estimated by ²¹⁰Po/²¹⁰Pb disequilibria were significantly lower than those by ²³⁴Th/²³⁸U disequilibria in the Bellingshausen Sea (Shimmield et al., 1995). In contrast, the POC export fluxes estimated by the two methods were similar in the ACC (Rutgers van der Loeff et al., 1997). Therefore, comparing the two methods not only helps to deepen the understanding of their applicability, but also helps to more accurately understand the temporal and spatial variability of the POC export flux.

Here, we report the activity concentrations of ²¹⁰Po and ²¹⁰Pb around Prydz Bay, including dissolved ²¹⁰Po (D²¹⁰Po), particulate ²¹⁰Po (P²¹⁰Po), dissolved ²¹⁰Pb (D²¹⁰Pb), and particulate ²¹⁰Pb (P²¹⁰Pb). The main objectives include: (1) revealing the distribution of ²¹⁰Po and ²¹⁰Pb around Prydz Bay; (2) assessing the factors affecting the disequilibria between²¹⁰Po and ²¹⁰Pb; and (3) quantifying the export flux of POC via ²¹⁰Po/²¹⁰Pb disequilibria, and compared with the result by ²³⁴Th/²³⁸U disequilibria.

Materials and Methods

Sampling

Seawater sample was collected using a Teflon-coated Niskin bottle assembled on a Sea-Bird SBE-911 rosette system (Sea-Bird Electronics Inc., United States) from January 31 to February 3, 2013 onboard R/V XUELONG. A total of 104 water samples (about 5 L each) was collected at different depths at six stations on the transect P7, covering the continental shelf (water depth from 200 to 400 m), continental slope (water depth of about 1,000 m), and the open ocean (water depth of about 3,500 m) (Figure 1). The water sample was filtered through a 0.4 μm polycarbonate membrane to separate the dissolved and particulate phases. The filtrate was acidified to pH <2 with approximately 20 mL concentrated HCl immediately. The particulate matter was frozen and stored at −20°C for further processing in the onshore laboratory.

FIGURE 1

Figure 1. Sampling locations around Prydz Bay. Stations P7-15, P7-14 and P7-11 are located in the shelf, station P7-09 is located in the slope, and stationsP7-07 and P7-05 are located in the basin.

Measurements

Temperature and conductivity were measured by SBE-911 CTD, the accuracy of which was 0.001°C and 0.0003 S/m, respectively. The water sample used for nutrient determination was filtered through a 0.45 μm cellulose acetate membrane and stored in a 100 mL acid-cleaned HDPE bottle by adding saturated HgCl₂ solution. The nitrate (NO₃-N), phosphate (PO₄-P), and silicate (SiO₃-Si) was determined by the zinc-cadmium reduction method, molybdenum-blue method, and molybdate-blue method, respectively (Grasshoff et al., 1983).

The particulate matter in the water sample (5–10 L) was filtered on a pre-combusted 47 mm GF/F membrane (450°C, 4 h) for POC measurement. After the sample was dried at 60°C, it was stored frozen at −20°C. In the laboratory, the filter was fumigated with concentrated HCl for 48 h to remove inorganic carbonate. The sub-sample was encapsulated in a tin boat, and sent to the elemental analyzer (NC2500, Carlo Erba) for POC determination. The detection limit of POC is 0.1 μmolC, and the precision is 0.2% (Ren, 2015).

The dissolved sample was enriched with ²¹⁰Po by co-precipitating with Fe(OH)₃, and purified by spontaneous deposition on a silver disk (4 h at 85°C). The mixed acid of HNO₃, HClO₄, and HCl was used to digest the particulate sample, and the subsequent procedure was the same as that of the dissolved sample. The radioactivity of ²¹⁰Po was counted by an alpha spectrometer (Octête^® Plus, EG&G) in Xiamen University. After the first self-deposition of ²¹⁰Po, the sample solution was left for more than 9 months, and the second spontaneous deposition of ²¹⁰Po was performed to determine the radioactivity of ²¹⁰Po produced by the decay of ²¹⁰Pb (Yang, 2005).

The radioactivity of ²¹⁰Po in the sample was calculated by correcting the ²¹⁰Po activity measured from the first self-deposition to the sampling time point. The calculation formula is as follows:

A_{2}^{0} = \frac{A_{2}^{1} - A_{2}^{2} e^{λ_{1} t_{1}} (e^{- λ_{1} t_{0}} - e^{- λ_{2} t_{0}}) / (e^{- λ_{1} t_{2}} - e^{- λ_{2} t_{2}}) R_{1}}{e^{- λ_{2} t_{0}} R_{2}} - B

where $A_{2}^{0}$ is the activity concentration of ²¹⁰Po at the time of sampling (Bq⋅m^–3); $A_{2}^{1}$ is the activity concentration of ²¹⁰Po at the time of first self-deposition (Bq⋅m^–3); t₀ represents the time interval from sampling to the first self-deposition; R₂ represents the chemical recovery of ²¹⁰Po, as determined by the ²⁰⁹Po tracer; B denotes the blank in the analysis process.

In the calculation of ²¹⁰Pb activity, considering that ²¹⁰Pb has a long half-life (22.3 years) and the time interval from sampling to co-precipitation is short, the effects of ²¹⁰Pb decay and ingrowth from sampling to co-precipitation could be ignored. Therefore, the ²¹⁰Pb activity was corrected back to the time point of co-precipitation. The calculation formula is as follows:

A_{1}^{0} = \frac{A_{2}^{2} (λ_{2} - λ_{1})}{λ_{2} e^{- λ_{1} t_{1}} (e^{- λ_{1} t_{2}} - e^{- λ_{2} t_{2}}) R_{1}} - B

where $A_{1}^{0}$ is the activity of²¹⁰Pb at the time point of co-precipitation (that is, the activity of ²¹⁰Pb in the sample, Bq⋅m^–3); $A_{2}^{2}$ is the ²¹⁰Po activity measured from the second self-deposition sample (Bq⋅m^–3); λ₁ and λ₂ are the decay constants of ²¹⁰Pb (0.031 a^–1) and ²¹⁰Po (1.828 a^–1), respectively; t₁ and t₂ represent the time interval from co-precipitation to the first self-deposition and the time interval from the first to second self-deposition, respectively; R₁ represents the chemical recovery of ²¹⁰Pb, which was determined by the stable Pb content measured by an atomic absorption spectroscopy; B represents the blank in the analysis process.

The error of ²¹⁰Po and ²¹⁰Pb activity reported here is ±1σ counting uncertainty.

Estimation of POC Export Flux

Friedrich and Rutgers van der Loeff (2002) found that ²¹⁰Po activity has a stronger correlation with POC and biogenic silica in the Southern Ocean, indicating that ²¹⁰Po/²¹⁰Pb disequilibria is a reliable method for estimating the export flux of POC in the Southern Ocean. Similar to the ²³⁴Th approach (Buesseler et al., 1992, 2006; Stewart et al., 2007), the export flux of POC is calculated by multiplying the export flux of P²¹⁰Po (i.e., the removal flux of ²¹⁰Po) by the POC/P²¹⁰Po ratio at the export interface (Shimmield et al., 1995). This is an empirical approach. It is not necessary to assume that POC and P²¹⁰Po have the same residence time. The calculation equation is as follows:

F_{P O C} = F_{P^{210} P o} . \frac{P O C}{P^{210} P o}

where F_POC and F_{P²¹⁰ P_o} represent the export flux of POC and P²¹⁰Po, respectively; $\frac{P O C}{P {}^{210}P o}$ is the ratio of POC to P²¹⁰Po at the export interface.

The F_{P²¹⁰ P_o} was calculated by a classical steady-state irreversible scavenging model after ignoring the atmospheric deposition flux of²¹⁰Po (Bacon et al., 1976; Friedrich and Rutgers van der Loeff, 2002; Masqué et al., 2002; Yang et al., 2006).

Here, the integration depth interval of P²¹⁰Po export is delineated according to the physical properties of water masses, namely, from 0 to 25 m (the mixed layer), from 50 to 200 m (the thermocline water), from 300 to 1,000 m (the CDW), and from 2,000 m to the bottom (the AABW), respectively. As the water depth of the continent shelf is relative shallow, it is divided into surface (from 0 to 25 m), subsurface (from 50 to 100 m), and deep layer (from 200 m to the bottom).

Results

Hydrological Characteristics

The temperature in the upper water increased to the north outside Prydz Bay (Figure 2A). The summer surface water in the top 50 m layer has a salinity of about 33.8 and a maximum temperature of 0.8°C. Below the surface water, extremely low temperature and higher salinity appear, which are characteristics of the Winter Water (WW, T < −1.5°C, 34.2 < S < 34.56). The formation of the WW is due to the weak mixing in the upper waters in summer, which causes the water to retain the winter characteristics for a long time (Pu et al., 2002a). The water with a depth of 500–2,000 m shows the characteristics of high temperature and high salinity of Circumpolar Deep Water (CDW) (T > 1.0°C, 34.5 < S < 34.75, Figures 2A,B). It is widely distributed in a large area around the continental shelf and is also one of the main water masses of Antarctic Circumpolar Current (ACC) (Pu et al., 2002b). Below the CDW (>2,000 m), the water mass with the highest salinity and the low temperature is called Antarctic Bottom Water (AABW), and its average temperature and salinity are −0.153 ± 0.022°C and 34.651 ± 0.001, respectively.

FIGURE 2

Figure 2. The sectional distribution of (A) temperature (°C), (B) salinity (psu), (C) nitrate (μmol dm^–3), (D) phosphate (μmol dm^–3), (E) silicate (μmol dm^–3), and (F) POC (μmol dm^–3).

Nutrients

Nitrate gradually increases from the shelf to the open ocean, while nitrite shows an increasing trend from north to south. Although the concentration of nitrate in surface water is low, it is still detectable. The concentration of nitrate in the bottom water increases, which may be due to the nitrification of ammonia released from the sediments (Figure 2C). The concentration of phosphate in the upper ocean is low, but it remains abundant at depths deeper than the mixed layer (Figure 2D). The vertical change of silicate in the shelf (Stns. P7-15, P7-14, and P7-11) is not obvious, but the profiles in the slope and the open ocean are different, in which the silicate increases as the depth increases. The highest silicate concentration appears near the bottom, which is attributed to the dissolution of biogenic silica in the sediments and subsequent transport to the overlying water (Figure 2E).

POC

The POC concentration varies from 0.35 to 5.37 μmol/dm³, with an average of 1.51 μmol/dm³. The POC concentration in surface water falls in a range of 1.77–5.37 μmol/dm³, with an average of 3.83 μmol/dm³. In the upper 100 m water column, POC decreases significantly as the depth increases, which is resulted from the photosynthesis of phytoplankton and the degradation of organic matter. In addition, the POC concentration in the shelf is higher than those in the slope and the open ocean, reflecting the spatial variation of primary productivity. Note that the POC concentration in near-bottom water tends to increase, especially at stations P7-05, P7-07, and P7-09, which may reflect the effect of sediment resuspension (Figure 2F).

²¹⁰Po and ²¹⁰Pb

The activity concentrations of ²¹⁰Po and ²¹⁰Pb are shown in Table 1, and the sectional distribution along ∼78.0 °E is shown in Figure 3.

TABLE 1

Table 1. The activity concentration of dissolved and particulate ²¹⁰Po and ²¹⁰Pb, and the activity ratio of T²¹⁰Po/T²¹⁰Pb.

FIGURE 3

Figure 3. The sectional distribution of (A) dissolved ²¹⁰Po (Bq/m³), (B) particulate ²¹⁰Po (Bq⋅m^–3), (C) total ²¹⁰Po (Bq⋅m^–3), (D) dissolved ²¹⁰Pb (Bq⋅m^–3), (E) particulate ²¹⁰Pb (Bq⋅m^–3), and (F) total ²¹⁰Pb (Bq⋅m^–3).

The activity concentration of D²¹⁰Po ranges from 0.47 to 3.20 Bq⋅m^–3, with an average of 1.44 ± 0.15 Bq⋅m^–3. The D²¹⁰Po in the shelf water (avg. 1.04 ± 0.11 Bq⋅m^–3) is lower than those in the slope (avg. 1.51 ± 0.15 Bq⋅m^–3) and open ocean (avg. 2.08 ± 0.20 Bq⋅m^–3). Compared with other water masses, the D²¹⁰Po in CDW is higher (p < 0.0001, one-way ANOVA, Figure 3A). The activity concentration of P²¹⁰Po varies in a wide range from 0.03 to 0.66 Bq⋅m^–3, with an average of 0.18 ± 0.04 Bq⋅m^–3. The vertical variation of P²¹⁰Po in the shelf decreases as the depth increases. The activity concentration of P²¹⁰Po in the slope is mostly stable at about 0.13 Bq⋅m^–3, but an abnormally high value appears at a depth of 2,000 m at station P7-07. The activity concentration of P²¹⁰Po in surface water (upper 50 m) of the open ocean is higher than that in the shelf and slope, with an average of 0.27 ± 0.06 Bq⋅m^–3 (Figure 3B). The activity concentration of T²¹⁰Po is mainly contributed by D²¹⁰Po, and its value is between 0.49 and 3.34 Bq⋅m^–3 (avg. 1.63 ± 0.15 Bq⋅m^–3). The T²¹⁰Po in the surface and the bottom layer is lower than that in the CDW. In general, the T²¹⁰Po shows a decreasing trend from open ocean (avg. 2.26 ± 0.21 Bq⋅m^–3) to the slope (avg. 1.71 ± 0.16 Bq⋅m^–3) and the shelf (avg. 1.21 ± 0.12 Bq⋅m^–3) (Figure 3C). This spatial variation is attributed to the high biological productivity and the active sediment resuspension in the shelf, resulting in the rapid scavenging of particle-reactive radionuclides (Chen et al., 2012). The vertical change of T²¹⁰Po in the shelf is small, which indicates that the particle scavenging is strong in the entire water column.

The distribution pattern of D²¹⁰Pb is generally similar to that of D²¹⁰Po (Figures 3A,D). The variation range of D²¹⁰Pb is from 1.15 to 2.97 Bq⋅m^–3, with the lowest in the shelf and the highest in the open ocean (p < 0.001, one-way ANOVA, Figure 3D). The D²¹⁰Pb is lower in surface water due to more effective particle scavenging, while the D²¹⁰Pb increases in the subsurface and deep water. The higher D²¹⁰Pb in CDW is affected by the remineralization of particulate organic matter (Bacon et al., 1976; Somayajulu and Craig, 1976; Thomson and Turekian, 1976; Wei et al., 2011). The activity concentration of P²¹⁰Pb ranges from 0.07 to 0.59 Bq⋅m^–3. In the shelf, the P²¹⁰Pb is higher in surface water, and lower in deep water (0.07–0.11 Bq⋅m^–3) (Figure 3E). The activity concentration of T²¹⁰Pb ranges from 1.26 to 3.16 Bq⋅m^–3, with an average of 2.20 ± 0.17 Bq⋅m^–3 (Table 1). The T²¹⁰Pb in the surface water is the lowest, especially in the shelf (Figure 3F), indicating that ²¹⁰Pb is effective scavenged by particles. The activity concentration of T²¹⁰Pb in CDW (2.49–3.02 Bq⋅m^–3) is higher than that in other water masses (p < 0.002, one-way ANOVA), which is consistent with the highest T²¹⁰Pb (∼3 Bq⋅m^–3) in ACC reported previously (Rama et al., 1961; Somayajulu and Craig, 1976; Chung and Applequist, 1980; Chung, 1981).

Discussion

²¹⁰Po/²¹⁰Pb Disequilibria

The activity ratio of T²¹⁰Po/T²¹⁰Pb ranges from 0.36 to 1.33, and shows a widespread deficiency of ²¹⁰Po with respect to ²¹⁰Pb around Prydz Bay (Figure 4). The average value of T²¹⁰Po/T²¹⁰Pb)_A.R. in the upper 100 m water column is 0.66, reflecting the influence of biological activity and particle scavenging. Compared with ²¹⁰Pb, organisms preferentially absorb ²¹⁰Po, and organic tissues have a stronger affinity for ²¹⁰Po (Bacon et al., 1976; Cochran et al., 1983; Stewart and Fisher, 2003; Stewart et al., 2007; Yang et al., 2009), which leads to more effective removal of ²¹⁰Po from the euphotic zone. Under the combined action of biological absorption and particle adsorption, the T²¹⁰Po/T²¹⁰Pb ratio in the shelf is lower than those in the slope and open ocean (compare the right and left panels in Figure 4), which is consistent with the observed low nutrients and high POC in the shelf (Figure 2).

FIGURE 4

Figure 4. Profiles of T²¹⁰Po/T²¹⁰Pb activity ratio at sites around Prydz Bay. The left panel is stations in the slope and the basin, and the right panel is stations in the shelf.

Some previous studies have shown a secular equilibrium is reached between ²¹⁰Pb and ²¹⁰Po in water deeper than 1000 m, such as the North Atlantic (Bacon et al., 1976), the South Pacific (Turekian and Nozaki, 1980), and the Indian Ocean (Chung and Finkel, 1988). However, a large deficiency of ²¹⁰Po has been found in deep waters of some seas, such as the East China Sea and the Philippine Sea (Nozaki et al., 1990), the equatorial Pacific and the Bering Sea (Nozaki et al., 1997), the Sargasso Sea (Kim and Church, 2001), and the South China Sea (Chung and Wu, 2005; Hong et al., 2013). In this study, most of the ²¹⁰Po in deep water around Prydz Bay are in equilibrium with ²¹⁰Pb, but there is an excess or deficiency of ²¹⁰Po at certain depths at some sites (Figure 4). These disequilibria may be caused by differences in particle composition. Hong et al. (2013) found a positive correlation between the flux of P²¹⁰Po and the flux of calcium carbonate, while no correlation between P²¹⁰Pb and calcium carbonate. However, Niedermiller and Baskaran (2019) found a significant negative correlation between the inventories of T²¹⁰Pb and particulate Al, but no correlation between ²¹⁰Po and particulate Al, suggesting much less removal of ²¹⁰Po by lithogenic material compared to ²¹⁰Pb. Therefore, detailed mineralogical and chemical composition of particles play an important role in deep water scavenging.

The activity ratios of T²¹⁰Po/T²¹⁰Pb are less than 1.0 in the bottom water of some sites, such as Stns. P7-15, P7-14, and P7-07 (Figure 4), indicating that boundary scavenging results in preferential removal of ²¹⁰Po over ²¹⁰Pb. The resuspension of sediments may increase the concentration of particles in near-bottom water, thereby enhancing the scavenging and removal of ²¹⁰Po in the benthic boundary layer, similar to what happens in the euphotic zone. Ma (2006) found that ²¹⁰Po in the bottom water of the Northeast Pacific was depleted relative to ²¹⁰Pb (the average ratio of T²¹⁰Po/T²¹⁰Pb was 0.64), which was attributed to the effect of benthic boundary scavenging.

High Activity Concentrations of ²¹⁰Po and ²¹⁰Pb in CDW

The CDW around Prydz Bay mainly exists in a depth range of 200–2,000 m, and it upwells during southward movement due to the effect of the seabed topography (Smith et al., 1984; Pu and Dong, 2003; Pu et al., 2007). Our results show that CDW has the highest activity concentrations of ²¹⁰Po and ²¹⁰Pb compared with other water masses (p < 0.01, one-way ANOVA, Figure 5). The reason is worthy of in-depth study.

FIGURE 5

Figure 5. Scatter plots of T²¹⁰Po (left panel) and T²¹⁰Pb activity concentration (right panel) versus potential temperature and salinity.

The half-life of²¹⁰Pb is relatively long (22.3 years), so lateral transport is one of the important factors affecting its redistribution in seawater (Smoak et al., 1996; Moran et al., 1997). Ku and Lin (1976) measured ²²⁶Ra in deep water south of the Antarctic Convergence Zone, and found that the distribution of ²²⁶Ra is related to the latitudinal transport of the circumpolar current from the Pacific sector to the Atlantic sector. Hanfland (2002) found that the highest activity concentration of ²²⁶Ra (about 3.08 Bq⋅m^–3) was stable in the ACC. ²²⁶Ra is a soluble radionuclide with a half-life of 1602 years. The movement of ACC around the Antarctic continent causes ²²⁶Ra to accumulate in the CDW and present a uniform characteristics. Therefore, the high activity concentration of ²¹⁰Pb in CDW is attributed to the in situ decay of ²²⁶Ra.

However, for the ²¹⁰Po in CDW, only considering the in situ production by ²¹⁰Pb decay cannot explain the ²¹⁰Po excess we observed (Figure 5). Due to the short half-life of ²¹⁰Po (138.4 days), the horizontal transport of the water mass cannot retain the excess signal of ²¹⁰Po for a long time. Therefore, the ²¹⁰Po in CDW is more likely to be supplied by local source. In the process of remineralization of POM, both ²¹⁰Po and ²¹⁰Pb are released into the dissolved phase, but the priority of ²¹⁰Po can lead to an excess of ²¹⁰Po over ²¹⁰Pb, because more ²¹⁰Po is bound to organic matter (Shimmield et al., 1995; Nozaki et al., 1997; Wei et al., 2011). Therefore, the remineralization of particulate organic matter is most likely to be responsible for the high ²¹⁰Po and its excess in the CDW. A similar situation was observed in the 100–300 m layer in the eastern North Atlantic and the intermediate layer in the northwestern North Pacific, which was attributed to the remineralization of biogenic particles or measurement errors (Bacon et al., 1976; Kawakami et al., 2008).

Estimates of POC Export Flux

Our results show that there is a good positive correlation between the solid to liquid ratio of ²¹⁰Po (i.e., P²¹⁰Po/D²¹⁰Po) and POC, while ²¹⁰Pb does not (Figure 6). This confirms that particulate organic matter regulates the biogeochemical cycle of ²¹⁰Po, and it is feasible to use ²¹⁰Po/²¹⁰Pb disequilibria to estimate POC export flux around Prydz Bay.

FIGURE 6

Figure 6. The relationship between P²¹⁰Po/D²¹⁰Po ratio, P²¹⁰Pb/D²¹⁰Pb ratio, and POC concentration (A: P²¹⁰Po/D²¹⁰Po ratio vs POC, B: P²¹⁰Po/D²¹⁰Po ratio vs POC).

Based on the ²¹⁰Po/²¹⁰Pb disequilibria, the POC export flux in the entire water column varies from 0.8–31.9 mmol⋅m^–2⋅d^–1, among which the variation ranges in the open ocean, the slope and the shelf are 0.8–20.2, 1.5–4.8, and 4.8–31.9 mmol⋅m^–2⋅d^–1, respectively (Table 2). Note that the calculated F_POC at station P7-15 is negative, which is caused by a slight excess of ²¹⁰Po over ²¹⁰Pb. In this case, the removal flux of ²¹⁰Po calculated by the model may not represent the real situation and will not be considered in the subsequent discussion. As shown in Figure 7, the POC export flux in the upper water column in the shelf is higher than that in the slope and the open ocean (p < 0.05, one-way ANOVA). The high POC export is consistent with active biological activities in the shelf. Previous studies have shown that the Chl-a concentration and primary productivity in Prydz Bay are significantly higher than those outside the bay, and there is a positive correlation between the Chl-a and POC in the euphotic zone (Liu et al., 2002; Qiu et al., 2004; Cai et al., 2005; Han et al., 2011; Yu et al., 2011; Sun et al., 2012). The active photosynthesis in the upper water enhances the export of POC to the deep sea in Prydz Bay. In fact, the POC export flux in deep water in the shelf (such as Stn. P7-14 and P7-15) is also higher than that in the slope and the open ocean.

TABLE 2

Table 2. POC export flux at different depth interfaces estimated from ²¹⁰Po/²¹⁰Pb disequilibria.

FIGURE 7

Figure 7. The export flux of POC at different depth interface in the shelf, slope, and basin.

In addition to ²¹⁰Po/²¹⁰Pb disequilibria,²³⁴Th/²³⁸U disequilibria have also been widely used to estimate the POC export flux in the euphotic zone (Murray et al., 1989, 2005; Shimmield et al., 1995; Kim and Church, 2001; Friedrich and Rutgers van der Loeff, 2002; Stewart et al., 2007; Buesseler et al., 2008; Wei et al., 2011). However, the different half-lives of ²³⁴Th and ²¹⁰Po and their different affinities for particulate matter (Po >> Pb ≈ Th >> Ra > U) may cause differences in the POC export fluxes estimated by the two methods (Kharkar et al., 1976; Murray et al., 2005). It is of great significance to compare the POC export flux obtained by the two methods. Since most of the reports based on ²³⁴Th/²³⁸U disequilibria are concerned with the POC export at a depth of 100 m, here we focus on comparing the POC export at this interface. Based on ²¹⁰Po/²¹⁰Pb disequilibria, POC export flux at 100 m depth interface around Prydz Bay is estimated to be in a range of 4.2–9.0 mmol m^–2 d^–1, with an average of 6.9 mmol m^–2 d^–1, which is comparable to most previously reported values in Antarctica seas, whether based on ²³⁴Th/²³⁸U or ²¹⁰Po/²¹⁰Pb disequilibria (Table 3). Yang et al. (2009) estimated by ²¹⁰Po/²¹⁰Pb disequilibria that the POC export flux at a site (64.00 °S, 73.00 °E) outside Prydz Bay is 2.3 mmol m^–2 d^–1, which is slightly lower than this study. Note that its site is located in the northern part of our sites and is more affected by the ACC upwelling, its lower POC export is reasonable. In contrast, the reported POC export fluxes at the 100 m interface in Prydz Bay via ²³⁴Th/²³⁸U disequilibria (17.1−117.2 mmol m^–2 d^–1, avg. 63.5 mmol m^–2 d^–1, He et al., 2007) were 9.2 times of our estimates on average, even though the sampling season and locations are close. We found that the POC concentration in the He et al. (2007) (avg. 17.5 μmol/dm³) was on average 8.3 times that of this study (avg. 2.1 μmol/dm³), which resulted in a significant increase in their POC export flux. Considering that biological activities in the Antarctic seas often show large interannual and temporal variability, it cannot be ruled out that the changes in phytoplankton growth have led to such large differences.

TABLE 3

Table 3. The POC export flux at a depth of 100 m in the Southern Ocean estimated from ²³⁴Th/²³⁸U to ²¹⁰Po/²¹⁰Pb disequilibria.

The difference in the POC export flux obtained between the ²³⁴Th method and the ²¹⁰Po method has also been found in the study of the equatorial Pacific Ocean (Murray et al., 1989), the Mediterranean Sea (Stewart et al., 2007), the Bellingshausen Sea (Shimmield et al., 1995), and the Antarctic Circumpolar Current (Friedrich and Rutgers van der Loeff, 2002; Verdeny et al., 2009). The difference in the time scale and biogeochemical behavior between ²³⁴Th and ²¹⁰Po is proposed to be responsible for this. The half-life of ²³⁴Th is relatively short (24.1 days), it records the export of particulate matter in a short time scale, including occasional blooms in productive sea areas, while ²¹⁰Po has a longer half-life (138.4 days), and its record is more likely to be smoothed and homogenized. In addition, in the one-dimensional steady-state model for calculating the export flux of nuclides, the physical processes such as advection, upwelling, and horizontal transport are ignored. This may have different effects on different nuclides, leading to deviations in the calculated export flux. For example, the upwelling brings deep water with a high activity concentration of ²¹⁰Po or ²³⁴Th to the surface, which underestimates the POC export flux. The underestimation of the ²¹⁰Po method may be greater because ²¹⁰Po has a longer half-life. The difference in biogeochemical behavior between ²³⁴Th and ²¹⁰Po is another possibility (Fisher et al., 1983; Murray et al., 2005). In the Bellingshausen Sea, the POC export flux estimated by the ²³⁴Th method is higher than that by the ²¹⁰Po method, which is attributed to the fact that ²³⁴Th tends to be adsorbed by lithogenic materials (Shimmield et al., 1995). Friedrich and Rutgers van der Loeff (2002) found that in the ACC, ²³⁴Th is easier to bind with biogenic silica, while ²¹⁰Po has a stronger affinity for POC. Murray et al. (2005) pointed out that ²¹⁰Po can be absorbed into cells by organisms, while ²³⁴Th is only adsorbed on particle surface, thereby ²¹⁰Po may be more conducive to tracking the export of POC. As the phytoplankton around Prydz Bay grows rapidly in summer (Liu et al., 2002; Qiu et al., 2004; Cai et al., 2005), ²¹⁰Po/²¹⁰Pb disequilibria is an ideal method to estimate the POC export flux in this case.

Although the values of POC export flux estimated by the two methods are different, their spatial distributions are similar around Prydz Bay, showing that the POC export flux at the 100 m interface in the shelf is higher than those in the slope and open ocean (Figure 7). It is worth noting that He et al. (2007) observed an extremely high POC export flux at 50 m in the slope, which was attributed to the rising CDW transporting nutrients to the surface, stimulating the primary productivity and the removal of ²³⁴Th. However, judging from our ²¹⁰Po/²¹⁰Pb results, the upwelling of CDW may lead to an underestimation of ²¹⁰Po export flux due to the impact of organic matter remineralization. In addition, we noticed that the POC/P²¹⁰Po ratios in the slope are closer to those in the shelf. Therefore, horizontal transport across the shelf may partially compensate for the underestimated ²¹⁰Po export’s impact on the POC export flux.

Conclusion

In this study, dissolved and particulate ²¹⁰Po and ²¹⁰Pb were measured in the entire water column around Prydz Bay. Our results show that the D²¹⁰Po and D²¹⁰Pb in the shelf are lower than those in the slope and the open ocean, indicating an enhanced particle scavenging in the shelf. Among the various water masses, the CDW has the highest activity concentrations of T²¹⁰Pb and T²¹⁰Po, reflecting the effects of ²¹⁰Pb decay and POM remineralization. Our results show that there is a good positive correlation between the solid-liquid ratio of ²¹⁰Po and POC, while ²¹⁰Pb does not. This indicates that particulate organic matter regulates the biogeochemical cycle of ²¹⁰Po, and ²¹⁰Po/²¹⁰Pb disequilibria is a reliable method for estimating the POC export flux around Prydz Bay. The estimated POC export flux based on ²¹⁰Po/²¹⁰Pb disequilibria ranges from 0.8 to 31.9 mmol m^–2 d^–1. The higher POC export at the 100 m interface in the slope is attributed to the horizontal transport across the shelf. Although the difference in biogeochemical behavior and time scale between ²¹⁰Po and ²³⁴Th may affect the estimated POC export flux, the spatial variation of POC export flux estimated by these two methods is consistent, which shows that ²¹⁰Po/²¹⁰Pb disequilibria is a reliable method for POC export.

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

HH and MC designed the study and wrote the manuscript. XL sampled onboard and measured ²¹⁰Po. CR determined POC. RJ sampled and revised the manuscript. YQ and MZ contributed the experimental tools. All authors contributed to the article and approved the submitted version.

Funding

This work was supported by Impact and Response of Antarctic Seas to Climate Change (IRASCC 01-01-02C and 02-01-01) and funded by Ministry of Natural Resources of the People’s Republic of China and Chinese Arctic and Antarctic Administration, National Natural Science Foundation of China (41721005), and Program funded by China Ocean Mineral Resources R&D Association (No. DY135-13-E2-03).

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

As with most endeavors, we owe thank to a number of people: Jianming Pan and Zhengbing Han for providing nutrient data, the captains and crews of R/V XUELONG for their assistance in sampling, and colleagues in the SCSEMC for supporting efforts.

References

Arrigo, K. R., Van Dijken, G. L., and Bushinsky, S. (2008). Primary production in the Southern Ocean, 1997–2006. J. Geophys. Res. 113:C08004. doi: 10.1029/2007JC004551