*Corresponding author.

Continental Shelf Research 19 (1999) 1049}1064

Lead-210 and polonium-210 in the winterwell-mixed turbid waters in the mouth

of the Yellow Sea

Gi-Hoon Hong!,*, Sun-Kyu Park!, M. Baskaran",Suk-Huyn Kim!, Chang-Soo Chung!, Sang-Han Lee!

!Korea Ocean Research and Development Institute, Ansan P.O. Box 29, Seoul 425-600, South Korea"Department of Marine Science, Texas A and M University, Galveston, TX 77551, USA

Received 2 February 1998; accepted 16 December 1998

Abstract

Concentration pro"les of 210Pb and 210Po were measured along a traverse of the mouth ofthe Yellow Sea in February 1993. Winter time suspended particulate matter concentration wasmore than 10}100 mg l~1 in the coastal domain and less than 10 mg l~1 in the central domain.Concentrations of dissolved 210Pb over the area were low ((5 dpm kg~1) due to the e$cientremoval of 210Pb from the water column over the shelf. Evidence for release of 210Po is seen ina sub-surface layer, close to the sediment}water interface, where 210Po is enriched in thedissolved form and depleted in the particles. The high concentration of SPM in the mouth of theYellow Sea appears to determine dissolved and particulate 210Pb and 210Po activities. Theatmospheric input of 210Pb is the major source of 210Pb in the region with minor contributionfrom the 210Pb rich Kuroshio Water. The K

$values of 210Po varied by a factor of 50 while the

corresponding values of 210Pb varied only by a factor of 4. It appears that in waters whereparticle concentrations are high ('10 mg l~1), the K

$appears to be independent of particle

concentration. ( 1999 Elsevier Science Ltd. All rights reserved.

Keywords: 210Pb; 210Po; Turbid waters; Yellow sea

1. Introduction

210Po (t1@2

"0.378 y) has a potential utility as tracer for biogeochemical processes,such as primary production, zooplankton grazing, degradation of particles, etc. ontime scales less than a few years because of its half-life and highly mobile nature

0278-4343/99/$ - see front matter ( 1999 Elsevier Science Ltd. All rights reserved.PII: S 0 2 7 8 - 4 3 4 3 ( 9 9 ) 0 0 0 1 1 - 4

Fig. 1. Sampling locations of YS9302 expedition.

(Shannon et al., 1970; Nozaki et al., 1990). Earlier studies have concluded thatpolonium behaves more like the nutrient elements (Kharkar et al., 1976; Fisher et al.,1983). Several studies of oceanic 210Pb (t

1@2"22.3y) distribution have shown that

scavenging from the water column is enhanced at the continental margins relative tothe open ocean (e.g. Bacon et al., 1976). Extensive measurements on 210Po and itsparent 210Pb in open - ocean surface waters have shown that the 210Po/210Pb activityratio, although geographically variable, averages about 0.5 (Tsunogai and Nozaki,1971; Nozaki et al., 1976; Bacon et al., 1976). The mean ratio corresponds to thebox-model mean residence time of 0.6 y, which is a factor of 2}3 shorter than that ofatmospherically derived 210Pb, suggesting the preferential removal of 210Po relativeto 210Pb from the surface water. However, the 210Po/210Pb activity ratio below 100 mare close to unity, since the deep-sea 210Pb residence times are generally of the orderof a few decades (Craig et al., 1973; Broecker and Peng, 1982). It was observed that210Po is enriched relative to 210Pb in soft tissues of marine organisms (Shannon et al.,1970; Kharkar et al., 1976). In oxygenated seawater the biological uptake may bemore important than inorganic adsorption for 210Po scavenging while the opposite istrue for 210Pb (Fisher et al., 1983; Kadko, 1993; Wei and Murray, 1994). The

1050 G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064

preferential scavenging of 210Po relative to 210Pb in photic zone caused primarily bybiological uptake and such a mechanism would not cause any signi"cant radioactivedisequilibrium between 210Po and 210Pb in the deep water as a whole (Nozaki et al.,1990; Bacon et al., 1988). In areas where hydrothermal activity is strong (East Paci"cRise), enhanced scavenging of 210Po in association with hydrothermal activity wasobserved (Kadko et al., 1987).

The Yellow Sea (including Bohai Sea), bounded by contiguous landmass of Chinaand Korean Peninsula, has a total area of 0.42]106 km2 with average depth of 44 m(Fig. 1). The sea is also characterized as extremely turbid water caused by the amplesupply of riverine particulate matter and resuspension from the bottom. The YellowSea is also experiencing seasonal strati"cation during summer. Nozaki et al. (1991)reported that during water column strati"cation period in summer, 210Po/210Pbactivity ratio averages about 0.3 and the residence times of 210Pb and 210Po in thewaters were about 2 months in the mouth of the Yellow Sea. Total 210Po and 210Pbconcentrations could also undergo seasonal change in the temperate regions. Tanakaet al. (1983) reported that total 210Po and 210Pb concentrations were high in winterand low in summer in the seasonally strati"ed Funka Bay, Japan. However, relativelyfew studies have been undertaken simultaneously to determine 210Pb and 210Poconcentrations in the dissolved and particulate phases in seawater (Chung and Finkel,1988; Bacon et al., 1988; Nozaki et al., 1990; Wei and Murray, 1994). We havemeasured 210Pb and 210Po in the dissolved and particulate phases during winter inorder to (1) establish the detailed 210Po } 210Pb disequilibrium relationship in thewater column and (2) elucidate the fractionation of Po and Pb between dissolved andparticulate phases as a function of suspended particulate loading.

2. Sampling and methods

In the framework of Yellow Sea Coastal Ocean Flux Program which was aimed toaddress the origin and biogeochemical provenance of water masses using temper-ature, salinity and d18O, primary productivity using 14C incubation and heavy metaldistribution in the sea for the period of 1991 to 1993, winter cruise using RV Eardowas carried out from 4 February to 18 February 1993 in the mouth of the Yellow Sea.Fig. 1 shows the station locations at which 210Po and 210Pb pro"les were measured.Stations D7 and D9 are in the relatively less turbid central water domain, stationsD1}D5 are in the western shallow water domain, and stations D10}D11 are in theeastern shallow water domain (Fig. 2c). Continuous temperature and salinity (con-ductivity) pro"les and discrete water samples were obtained by a Sea-Bird CTD 25assembly "tted with twelve 10-l Niskin bottles. Discrete salinity samples were taken tocheck the accuracy of the CTD data. Samples for suspended particulate matter (SPM)were "ltered through the pre-combusted at 5503C for 3 h and pre-weighed 0.7 lmGF/F Glass "ber "lters, and washed 5}6 times with distilled water to remove salt inthe "lter and kept in frozen on board. The frozen "lters were freeze-dried in thelaboratory. The dried "lters were weighed and SPM concentrations were determinedby subtracting original "lter weight. The precision was estimated to be less than 4%.

G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064 1051

Fig. 2. Depth distribution of temperature (a), salinity (b), and SPM (c) concentration in the mouth of theYellow Sea.

Two aliquots of 3 l samples were passed through a 0.7 lm pore size, 47 mmdiameter glass "ber "lter for collection of particulate phase of 210Pb and 210Po, andthe "ltered water was stored in a polyethylene bottle. Shortly after collection, the"ltered water samples were acidi"ed to pH 2 with concentrated HCl. The 209Po spike(0.18 dpm) and iron carrier (50 mg as Fe) were added to one of the aliquots and 210Po(together with 209Po) was coprecipitated with ferric chloride and isolated from 210Pbby plating Po onto a silver disk. This was done in the laboratory within a week aftersampling. The electroplated disk was counted for polonium in a high-resolution alphaspectrometer coupled to a multichannel analyzer. The 210Po activity was decaycorrected from the time of collection to mid-counting date. The other aliquot wasstored for more than 2 y, during which time, the in situ 210Po decayed awaycompletely and 210Po produced from 210Pb reached an almost equilibrium value with210Pb in the water and then Po was electroplated onto silver disks and alpha counted.

1052 G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064

For particulate 210Po activity determination, one of the two "lters used was digestedwith Conc. HF, HNO

3, HCl successively after adding 209Po spike. Polonium was

electroplated from the "nal solution and alpha counted. The other "lter was alsostored for more than 2 y, during which time, 210Po reached an almost equilibriumvalue with 210Pb in SPM on the "lter. The 210Pb activity was calculated from the210Po value (Nozaki et al., 1991). The 210Pb decay correction for the time allowed forthe ingrowth of 210Po was applied.

3. Results and discussion

The hydrographic data are shown in Fig. 2. The water column is well mixed at thistime of the year (Fig. 2b). The distribution and variability of water properties in theYellow Sea are in#uenced by strong seasonal variations of river discharge, air}seainteraction, tidal mixing and the Kuroshio Current. The seasonal cycle of freshwaterdischarge from these rivers dominates the surface distribution of water properties inthe sea, especially during summer when the river input is the largest. Air}sea interac-tion is responsible for the vertical strati"cation of water masses in the sea where thewater is mixed vertically by strong surface cooling and wind mixing during winter andrestrati"ed by strong surface heating during summer. Non-linear interaction betweenstrong semi-diurnal tidal currents over the bottom in the shallower regions can causeboth strong tidal mixing and generate subtidal #ow (Lie, 1989; Suk et al., 1996).During 4}10 February 1993 in the mouth of the Yellow Sea, water temperature wasthe highest (123C) in the central region of the Yellow Sea trough, and lowest in thecoastal regions of China and Korea. Salinity was the highest (34.20 psu) in the YellowSea trough, and got lower approaching coasts of China and Korea. Temperature andsalinity are higher in the Korean coast than in the Chinese coast. Both temperatureand salinity were vertically homogenous. Suspended particulate matter (SPM) con-centrations were very high (SPM '30 mg l~1) in the shallow region (less than 40 mdepth), and relatively low ((10 mg l~1) in the central region. Winter condition in themouth of the Yellow Sea may be characterized by the warm saline water that intrudeinto the Yellow Sea from the southeast to the northwest. The Yangtz River-derivedwater was reported to extend only up to 1243E 32.53N (Hong et al., 1995a). The resultsof 210Po and 210Pb are given in Table 1 and plotted against depth in Fig. 3.

3.1. 210Pb proxles

In general, dissolved 210Pb activities were less than 3 dpm/100 l in the westernturbid shallow domain and less than 2 dpm/100 l in the central clear water region (Fig.3). Particulate 210Pb activities increased with depth and high in the shallow domains(Fig. 4). Speci"c 210Pb activities in SPM ranged between 4 and 23 dpm (g SPM)~1

with values low in the western shallow domain and high values in the eastern shallowdomain. These values are considerably higher than the 0}1 cm section of the bottomsediment core. 210Pb activity in the sur"cial sediment varies from 2.8 to 8.0 dpmg~1 (Hong et al. unpublished, DeMaster et al., 1985). The Yellow Sea water is

G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064 1053

Fig. 3. Depth distributions of dissolved 210Po (a), particulate 210Po (b), dissolved 210Pb (c), and particulate210Pb (d) in the mouth of the Yellow Sea.

characterized by 210Pb de"ciency (dissolved phase as well as total) relative to 226Ravalue (Elsinger and Moore, 1984; Nozaki et al., 1991), which is indicative of removalby scavenging. Most of the 210Pb in the upper part of the water column is supplied bydeposition from the atmosphere, which is presumed to be relatively uniform over the

1054 G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064

Fig. 4. Speci"c activities of 210Po (a) and 210Pb (b) in the suspended particulate matter in the mouth of theYellow Sea.

entire area. The strong concentration gradient across the front is due to variations inthe e$ciency of removal by particles relative to regeneration in the water column.This horizontal gradient should give rise to a net #ux of 210Pb across the frontal zoneonto the central region.

3.2. 210Po proxles

Dissolved 210Po activities peak at the sub-surface at the western turbid waters(4}8 dpm/100 l) and are low ((2 dpm/100 l) and vertically homogenous in the salineclear central waters (Fig. 3). Particulate 210Po activities gradually increase with depth(Fig. 4). Speci"c 210Po activities on suspended particles ranged between 4 and 13 dpm(g SPM)~1 and particles in the western shallow domain is more enriched with 210Pothan those in the eastern shallow domain. The behavior of 210Po is best observed bytaking its activity di!erence relative to the parent nuclide 210Pb (excess 210Po). Thusa value of zero will occur when 210Po is in radioactive equilibrium with 210Pb. Thedissolved excess 210Po was nearly zero in the central domain and enriched in thewestern and eastern shallow domains (Fig. 5). The particulate excess 210Po was nearlyzero in the central domain and enriched in the surface waters of the eastern shallowdomain, while particles in the bottom layer show a depletion of 210Po with respect to210Pb indicating regeneration of 210Po from the particulate (Fig. 5). The desorption of210Po from the particulate matter only partly balances the excess in the dissolvedform. This indicates a net transport of dissolved 210Po out of the mouth of the YellowSea.

G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064 1055

Fig. 5. Excess dissolved 210Po (a) and particulate 210Po (b) in the mouth of the Yellow Sea. The excessamount was obtained by subtracting 210Pb activity from 210Po activity.

It has been suggested that zooplankton grazing and fecal pellet production may beimportant for the vertical transport of 210Po (Beasley et al., 1978). For 210Po,biological uptake is more important than inorganic adsorption on to suspendedparticles (Fisher et al., 1983; Kadko, 1993) while the opposite is true for 210Pb. Thiscontrasting behavior of 210Pb and 210Po for particles can be seen from the empiricaldistribution coe$cients and fractionation factor for the two nuclides. The distributioncoe$cient K

$is given by

K$"C

1/(C

$]TSPM),

where C1and C

$are concentrations of the nuclides in particulate and dissolved forms,

respectively, and TSPM is the total suspended particulate matter concentration. TheK

$values for 210Po varied between 3.9]104 cm3g~1 and 331]104 cm3g~1. The

K$values for 210Po were high in the central domain and an order of magnitude lower

in the shallow domains due to di!erences in the concentration of SPM. The K$values

for 210Pb were more uniform, varying between 1.7]105 cm3 g~1 and 7.2]105 cm3

g~1 (Table 1). This relatively constant values of K$

for 210Pb could suggest that thedistribution of 210Pb is less dependent on the concentration of particles. The frac-tionation factor, F

P0@P", is simply the ratio of the distribution coe$cients, KP0

$/KP"

$(Bacon et al., 1988). The fractionation factor varied between 0.19 and 6.4. For 210Pothe highest distribution coe$cients are generally found in the surface layer, and highfractionation factors A1 indicate the preferential uptake of 210Po by phytoplanktonrelative to 210Pb by inorganic particles (Shannon et al., 1970). Minimum values of

1056 G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064

KP0$

and FP0@P"

are generally found in the near sediment-water interface zone wheremost of the particles are inorganic; biogenic particles are likely regenerated beforereaching the sediment-water interface (Fig. 6). Wei and Murray (1994) also observedthat the K

$for 210Po in the euphotic zone is much greater than that of 210Pb. In

terrigenous particles, such as aluminosilicates, the 210Po is expected to be radioactiveequilibrium with its parent, 210Pb while in biogenic particles, gross disequilibrium ispossible. The fractionation factor F

P0@P"could be an important parameter to charac-

terize biotic versus abiotic oceanic particles.

3.3. Mass balance for 210Pb in the mouth of the Yellow Sea

The major source of 210Pb to the continental shelf is the atmospheric deposition.The measured atmospheric depositional #ux of 210Pb for Ansan, mid-western coast ofKorean Peninsula (for a period of 18 months) is 2.0 dpm cm~2 y~1 (Nozaki et al. 1991;Park, 1993). Supply of 210Pb by rivers is generally assumed to be insigni"cant due torapid scavenging of 210Pb by riverine suspended sediment particles and e$cienttrapping of sediments in estuaries (Benninger et al., 1975). An additional supply of210Pb to the shelf is from the Kuroshio Current, due to higher concentrations ofdissolved 210Pb o!shore and exchange across the shelf-Kuroshio front (Nozaki et al.,1991). This input can be estimated from a knowledge of the average residence time ofwater in the mouth of the Yellow Sea (3}4 years, Nozaki et al., 1991), the in#ux ofKuroshio Water (&6]1012 m3 y~1, Hong et al., 1995b), and concentration of 210Pbin the Kuroshio Water (26.6 dpm /100 l, Nozaki et al., 1991). Using these values, theoceanic input of 210Pb is estimated to be about one-"fth of the atmospheric input.

In order to identify if the scavenging of particle-reactive nuclide in this area iscontrolled by diapycnal (vertical scavenging) processes or isopcynal transport (lateraltransport), the steady state #ux of 210Pb to the bottom sediment can be compared tothe inputs (atmospheric fallout # input from Kuroshio Current) for shallow waters.The 210Pb #ux to the bottom sediment along the mouth of the Yellow Sea have beendetermined using excess 210Pb down core distribution and the values range between0.1 and 2.9 dpm cm~2 yr~1 in the central (0.14, 0.74, 0.80 dpm cm~2 yr~1) and thewestern shallow (2.9 dpm cm~2 yr~1) domains (Hong GH, unpublished data). Sedi-ment inventories of excess 210Pb vary from 4 to 94 dpm cm~2 (4.3, 23.84, 25.55, 94.51dpm cm~2) depending upon the sediment accumulation rate (Hong, unpublisheddata). Total 210Pb inventory in the water column is about 1 dpm cm~2 (Table 1)which is much less than the sediment inventory. Although the sediment inventories ingeneral are comparable to the input #ux, isopycnal transport is stronger thandiapycnal transport in the sea. Strong tidal current could potentially redistribute theparticles in shallow waters. There are other shelf areas where the sediment inventoriesare comparable to the atmospheric fallout. Bacon and Belastock (1991) observed thatthe sediment inventories of 210Pb are similar to the inventory predicted from theatmospheric supply in the shelf sediments of mid-Atlantic Bight since the 210Pb ladensilt- and clay- particles are e$ciently trapped on the shelf because deposition isfollowed by downward mixing into the sediment column by benthic organisms. Thesediment inventories of 210Pb in shelf and slope regions of the Gulf of Mexico are

G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064 1057

Table 1210Pb/210Po disequilibria and oceanographic variables in the mouth of the Yellow Sea during winter 1993(YS9302).

Stn. Depth 210Pb (dpm/100 l) 210Po(dp m/100 l) 210Po/210Pb(m) Dissolved Particulate Dissolved Particulate Dissolved

D1 0 1.99$0.25 29.8$0.8 2.8$0.8 29.4$0.9 1.41$0.1810 1.90$0.27 31.9$0.7 1.5$0.5 45.7$1.4 0.79$0.1130 1.75$0.29 41.3$0.9 9.3$3.0 42.6$1.3 5.31$0.90

D3 0 1.84$0.31 42.8$0.9 1.2$0.6 $1.3 0.65$0.1110 3.33$0.37 55.7$1.1 11.1$1.5 41.2$1.2 3.33$0.3830 1.31$0.25 55.0$1.3 1.4$0.4 64.3$1.9 1.07$0.21

D5 0 1.99$0.27 14.7$0.6 0.3$0.2 14.1$0.4 0.15$0.0210 1.79$0.26 13.9$0.5 2.6$0.8 12.0$0.4 1.45$0.2230 2.23$0.26 19.2$0.7 1.0$0.2 16.8$0.5 0.45$0.0550 2.46$0.44 15.0$0.7 2.2$0.7 12.5$0.4 0.89$0.16

D7 0 1.84$0.28 5.7$0.3 1.6$0.4 1.2$0.0 0.87$0.1410 1.48$0.23 6.2$0.4 1.4$0.3 6.6$0.2 0.95$0.1530 1.52$0.18 9.5$0.4 0.5$0.2 10.0$0.3 0.33$0.0450 1.64$0.19 8.1$0.4 1.6$0.4 7.8$0.2 0.98$0.1275 1.49$0.21 7.3$0.6 1.4$0.4 8.5$0.3 0.94$0.14

D9 0 1.81$0.25 5.8$0.4 5.6$1.5 6.9$0.2 3.09$0.4410 2.30$0.39 6.4$0.5 2.8$0.6 5.4$0.2 1.22$0.2130 2.69$0.25 5.0$0.3 1.2$0.3 4.5$0.1 0.45$0.0450 1.93$0.24 4.6$0.3 1.6$0.6 3.9$0.2 0.83$0.11

100 1.95$0.20 29.9$0.8 2.3$0.7 33.3$1.1 1.18$0.13D 10 0 3.25$0.33 19.1$0.7 4.7$1.0 16.0$0.5 1.45$0.15

30 3.11$0.28 26.4$0.8 3.8$0.9 23.5$0.7 1.22$0.1250 3.01$0.27 38.4$1.0 3.7$0.8 40.6$1.2 1.23$0.1265 2.50$0.33 59.1$1.2 4.1$1.1 20.5$0.6 1.64$0.22

D11 0 ND ND ND 08.5$0.3 ND10 5.28$0.54 21.4$0.8 5.7$1.0 61.4$1.8 1.08$0.1230 4.06$0.60 31.3$1.0 6.4$1.1 29.9$0.9 1.58$0.2455 4.11$0.78 33.6$1.0 7.6$1.5 32.0$1.0 1.85$0.36

comparable to the expected inventory based on atmospheric fallout (Baskaran andSantschi, 1999). Similar observation was also reported for the shelf regions of the EastChukchi Sea, Alaskan Arctic (Baskaran and Naidu, 1995).

3.4. Water column stratixcation and seasonal variations of the concentrations of SPM,210Po, and 210Pb

The surface water concentrations of 210Pb and 210Po were much lower in summerthan in winter. Nozaki et al. (1991) sampling was carried out during May}June 1987and thus may represent summer conditions and our data were obtained duringFebruary 1993 and could represent winter conditions (Hong et al. 1995a). May}June1987 sampling period appears to be strati"ed as evidenced by the depletion ofdissolved silica concentration (Nozaki et al., 1991). Total (dissolved and particulate)

1058 G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064

210Pb R¹ K$(cm3g~1) K

$(cm3g~1) KP0

$/KP"

$Temp. Sal SPM

(Month) 210Po 210Pb (3C) (psu) (mg/l)

2.67E#05 3.80E#05 0.70 8.72 32.1 39.33.96E#05 2.19E#05 1.81 8.76 32.1 76.8

0.41 8.32E#04 4.29E#05 0.19 8.79 32.1 55.06.63E#05 4.17E#05 1.59 8.54 31.6 55.83.85E#04 1.74E#05 0.22 8.47 31.7 96.3

0.52 6.22E#05 5.68E#05 1.09 8.46 31.7 73.83.31E#06 5.19E#05 6.38 11.45 34.0 14.22.05E#04 3.42E#05 0.60 11.46 33.9 22.65.68E#05 2.92E#05 1.95 11.47 33.9 29.5

0.77 1.93E#05 2.08E#05 0.93 11.46 33.9 29.31.37E#05 5.48E#05 0.25 12.51 34.2 5.65.55E#05 4.96E#05 1.12 12.51 34.3 8.51.84E#06 5.76E#05 3.20 12.51 34.3 10.94.38E#05 4.42E#05 0.99 12.51 34.3 11.1

0.84 5.34E#05 4.32E#05 1.24 12.49 34.3 11.42.78E#05 7.22E#05 0.39 ND ND 4.43.03E#05 4.40E#05 0.69 10.27 33.6 6.35.78E#05 2.86E#05 2.02 10.33 33.7 6.52.34E#05 2.28E#05 1.03 10.94 33.8 10.5

1.54 3.97E#05 4.20E#05 0.94 11.87 34.0 36.52.16E#05 3.72E#05 0.58 7.70 33.1 15.82.19E#05 3.00E#5 0.73 7.71 33.1 28.32.17E#05 2.52E#05 0.86 7.72 33.1 50.6

1.43 9.35E#04 4.43E#05 0.21 7.48 33.1 53.4ND ND ND 8.53 33.2 6.64.96E#05 1.87E#05 2.65 8.53 33.2 21.71.38E#05 2.27E#05 0.61 8.64 33.2 34.0

1.60 ND ND ND

activities of 210Pb and 210Po varied between 4 and 10 dpm/100 l and between 0.8 and3.4 dpm/100 l, respectively, in summer (Nozaki et al., 1991), and 8 and 44 dpm/100l and 3 and 46 dpm/100 l (Table 1), respectively, in winter. Total 210Po and 210Pbconcentrations were also observed to be high in winter and low in summer in thenorthern temperate seasonally strati"ed Funka Bay, Japan (Tanaka et al., 1983). SPMloading in the water column undergoes seasonal changes due to seasonality of theforcing functions. In general, SPM concentrations are high during winter and low insummer (Milliman et al., 1986; Wells, 1988).

3.5. Residence time of 210Po and 210Pb in the water column

As discussed before 210Pb and 210Po concentrations are considerably lower in theshelf waters than in the Kuroshio Current (Nozaki et al., 1991), and 210Pb in thesurface water is predominantly of atmospheric origin. The atmospheric #ux of 210Pb

G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064 1059

Fig. 6. Percentage of particulate 210Po and 210Pb in the mouth of the Yellow Sea.

in this region is about 2 dpm cm~2 yr~1 (Nozaki et al., 1991), which is much higherthan in situ production from 226Ra in the water (about 0.03 dpm cm~2 yr~1 ina 100 m water column). The input #ux from Kuroshio Current was estimated to be&0.4 dpm cm~2 yr~1. Since the decay of 210Pb in seawater is also small, the meanresidence time of 210Pb in the entire water column with respect to particle scavengingis obtained simply by dividing the amount of 210Pb (dpm cm~2) in the water columnby the combined input from atmosphere and Kuroshio Current (dpm cm~2 yr~1).The calculated 210Pb residence times are given in Table 1. The 210Pb residence timewas less than 1 month in the turbid western shallow water domain and about2 months in the clear water domain as observed during summer by Nozaki et al.(1991).

Nozaki et al. also suggested that 210Po residence times (RT) could be calculated bythe following equation:

RT(P0)

"(R/(1!R))]1/jP0

,

where R is the (210Po/210Pb) ratio in the water and jP0

is decay constant of 210Po. Inderiving the above equation, the atmospheric #ux of 210Po is ignored because it issmall compared to in situ production from 210Pb in the water (Bacon et al., 1976;Nozaki and Tsunogai, 1976). However, the data in the mouth of the Yellow Sea do notallow us to make a simple box-model calculation, yielding unreasonably long 210Poresidence times. This is presumably due to the high suspended load with the(210Po/210Pb) ratio of &1 (an equilibrium value expected for terrestrial particlesolder than 2 yr, Table 1). As noted by Nozaki et al. (1991), the residence times of 210Pbin the shelf water are short compared to the residence times of the waters of 2}3 y(Nozaki et al., 1991). Thus, it is likely that this reactive nuclide is largely deposited on

1060 G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064

Fig. 7. Plot of Log K$

versus Log SPM for (a) 210Po and (b) 210Pb versus Log SPM.

G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064 1061

the shelf sediments prior to transport from the shelf to the open ocean. This con-clusion probably applies to other land-derived heavy metals and pollutants that havechemical reactivities similar to 210Pb, as suggested by Nozaki et al. (1991).

3.6. Control of trace metal concentrations in coastal seawater through partition ontosuspended particulate matter

It has been shown that the concentration of particle-reactive nuclides in areas wherethe suspended particle concentrations are high ('10 mg l~1), a major fraction of thenuclide exists in the particulate form (for example, Baskaran and Santschi, 1993). Thedistribution coe$cient for particle-reactive nuclides, such as Th, has been observed tobe correlated with SPM concentration (Honeyman and Santschi, 1989). Balls (1988)had shown that the distribution of metal concentration between dissolved andparticulate phases in coastal seawater depends on the amount of SPM loading. Due tothe high loading of SPM, 210Pb and 210Po are mostly distributed in the particulatephase (Fig. 6). Percentage of total 210Pb concentration in particulate phase in themouth of the Yellow Sea is similar to the stable Pb in the coastal waters in the NorthSea (Balls, 1988). However, percentage of total 210Pb concentration in particulatephase is more strongly dependent upon SPM loading than that of total 210Poconcentration in particulate phase.

The distribution coe$cient, K$, of 210Po and 210Pb, is plotted against the particle

concentration in Fig. 7. For other particle-reactive nuclides such as 7Be and 234Th,negative log K

$versus log SPM correlations have been previously observed and this

observation is known as the particle concentration e!ect (McKee, 1986; Honeymanet al., 1988). Several hypotheses have been put forward to explain this observation,including kinetics, irreversible adsorption, "ltration artifacts, particle}particle interac-tion and presence of colloid-bound metals which are often included in the "ltratefraction. Of these, the explanation due to the arti"cial separation of dissolved andparticulate phases (due to presence of colloid-bound metals) is widely accepted(Honeyman et al., 1988). The lack of inverse correlation between log K

$of 210Po

(or 210Pb) with log SPM clearly indicates that there is no particle concentration e!ectin areas where particle concentrations are greater than 10 mg l~1 and the variations inparticle-concentration is less than two orders of magnitude.

4. Conclusions

We have measured the dissolved and particulate concentrations of 210Po, 210Pband SPM concentrations in the mouth of the Yellow Sea. From the present investiga-tion, the following conclusions can be drawn: (i) The high concentration of SPM in themouth of the Yellow Sea appears to determine dissolved and particulate 210Pb and210Po activities. (ii) The residence time of 210Pb was less than 1 month in the turbidwestern shallow water domain and about 2 months in the clear water domain. (iii) Theatmospheric input of 210Pb is the major source of 210Pb in this region with minorcontribution from the 210Pb rich Kuroshio Water. (iv) The K

$values of 210Po varied

1062 G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064

almost by a factor of 85 while the K$for 210Pb varied only by a factor of 4. Thus, the

K$

values of Po as well as the ratio of KP0$

/KP"$

can be utilized to infer the nature ofparticles, viz., biogenic versus abiogenic particles. (v) In coastal waters where particleconcentrations are greater than 10 mg l~1 and variations on particle concentration israther narrow, the K

$s appear to be independent of SPM (no particle concentration

e!ect).

Acknowledgements

We thank the crew of the R/V Eardo for their assistance in sample collection. Wealso thank two anonymous reviewers for their considerate and thorough reviews. Thiswork was supported by Ministry of Science and Technology and Ministry of Environ-ment, Korea, under the Grants in Aid BSPN 203 and BSPN 218.

References

Bacon, M.P., Belastock, R.A., 1991. Lead-210 balance and implications for particle transport on thecontinental shelf, mid-Atlantic bight, USA (abstract). In: Kershaw, R.J., Woodhead, D.S. (Eds.)Radionuclides in The Study of Marine Processes. Elsevier Applied Science, London, pp. 164.

Bacon, M.P., Belastock, R.A., Tecotzky, M., Turekian, K.K., Spencer, D.W., 1988. Lead-210 and polon-ium-210 in ocean water pro"les of the continental shelf and slope south of New England. ContinentalShelf Research 8, 841}853.

Bacon, M.P., Spencer, D.W., Brewer, P.G., 1976. 210Pb/226Ra and 210Po/210Pb disequilibria in seawaterand suspended particulate matter. Earth and Planetary Science Letters 32, 277}296.

Balls, P.W., 1988. The control of trace metal concentrations in coastal seawater through partitioning ontosuspended particulate matter. Netherlands Journal of Sea Research 22, 213}218.

Baskaran, M., Naidu, S.A., 1995. 210Pb-derived chronology, and the #uxes of 210Pb and 137Cs isotopes intocontinental shelf sediments, East Chukchi Sea. Alaskan Arctic. Geochimica et Cosmochimica Acta 59,4435}4448.

Baskaran, M., Santschi, P.H., 1993. The role of particles and colloids in the transport of radionuclides incoastal environments of Texas. Marine Chemistry 43, 95}114.

Baskaran, M., Santschi, P.H., 1999. Particulate and dissolved 210Pb activities in the shelf and slope regionsof the Gulf of Mexico waters. Continental Shelf Research, Submitted for publication.

Beasley, T.M., Heyrand, M.H., Higgo, J.J.W., Cherry, R.D., Fowler, S.W., 1978. 210Po and 210Pb inzooplankton fecal pellets. Marine Biology 44, 325}328.

Benninger, L.K., Lewis, D.M., Turekian, K.K., 1975. The use of natural Pb-210 as a heavy metal tracer inthe river-estuarine system. In: Church, M.T. (Eds.), Marine Chemistry in The Coastal Environment. ACSSymposium Series, 18. American Chemical Society, Washington DC, pp. 202}210.

Broecker, W.S., Peng, T.H. 1982. Tracers in the Sea. Eldigio Press. 689pp.Chung, Y., Finkel, R., 1988. 210Po in the western Indian Ocean: distributions, disequilibria and partitioning

between the dissolved and particulate phases. Earth and Planetary Science Letters 88, 232}240.Craig, H., Krishanawami, S., Somayajulu, B.L.K., 1973. 210Po!226Ra: radioactive disequilibrium in the

deep sea. Earth and Planetary Science Letters 17, 295}273.DeMaster, D.J., McKee, B.A., Nittrouer, C.A., Jiangchu, Q., Guodong, C., 1985. Rates of sediment

accumulation and particle reworking based on radiochemical measurements from continental shelfdeposits in the East China Sea. Continental Shelf Research 4, 143}158.

Elsinger, R.J., Moore, W.S., 1984. 226Ra and 228Ra in the mixing zones of the Pee Dee River-Winyah Bay,Yangtz River and Delaware Bay estuaries. Estuarine, Coastal and Shelf Science 18, 601}613.

G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064 1063

Fisher, N.S., Burns, K.A., Cherry, R.D., Heyraud, M., 1983. Accumulation and cellular distribution of241Am, 210Pb and 210Po in two marine algae. Marine Ecology Progress Series 11, 233}237.

Honeyman, B.D., Santschi, P.H., 1989. A Brownian-pumping model for oceanic trace metal scavenging:evidence from Th isotopes. Journal of Marine Research 47, 951}992.

Honeyman, B.D., Balistrieri, L.S., Murray, J.W., 1988. Oceanic trace metal scavenging: the importance ofparticle concentration. Deep-Sea Research 35, 227}246.

Hong, G.H., Kim, S.H., Chung, C.S., Kang, D.J., Choi, J.K., Lee, T.S., Jung, J.Y., 1995a. Nutrients andbiogeochemical provinces in the Yellow Sea. . In: Tsunogai, S. (Ed.) Proceedings of the 1994 SapporoIGBP Symposium. Hokkaido University, Sapporo, pp. 1225}1239.

Hong, G.H., Kim, S.H., Chung, C.S., Pae, S.J., 1995b. The role of the anthropogenic nutrient input in thecarbon "xation of the coastal ocean Yellow Sea. A preliminary study. In: Handa, N., Ohsumi, T. (Eds.)Direct Ocean Disposal of Carbon Dioxide. Tera Scienti"c Publishing Co, Tokyo, pp. 13}22.

Kadko, D., 1993. Excess 210Po and nutrient cycling within the California coastal transition zone. Journal ofGeophysical Research 98, 857}864.

Kadko, D., Bacon, M.P, Hudson, A., 1987. Enhanced scavenging of 210Pb and 210Po by processesassociated with the East Paci"c Rise near 8345'N. Earth Planetary Science and Letters 81, 349}357.

Kharkar, D.P., Thomson, J., Turekian, K.K., Forster, W.O., 1976. Uranium and thorium decay seriesnuclides in plankton from the Caribbean. Limnology and Oceanography 21, 294}299.

Lie, H.J., 1989. Tidal fronts in the southeastern Hwanghae (Yellow Sea). Continental Shelf Research 9,527}546.

McKee, B.A., 1986 The fate of particle-reactive radionuclides in the Amazon and Yangtze continentalshelves. Ph. D. Dissertation, North Carolina State University.

Milliman, J.D., Li, F., Zhao, Y.Y., Zhang, T.M., Limeburner, R., 1986. Suspended matter regime in theYellow Sea. Progress in Oceanography 17, 215}227.

Nozaki, Y., Ikuta, N., Yashima, M., 1990. Unusually large 210Po de"ciencies relative to 210Pb in theKuroshio Current of the East China and Philippine Seas. Journal of Geophysical Research 95,5321}5329.

Nozaki, Y, Tsubota, H., Kasemsupaya V., Yashima, M., Ikuta, N., 1991. Residence times of surface waterand particle reactive 210Pb and 210Po in the East China and Yellow seas. Geochimica et CosmochimicaActa 55, 1265}1272.

Nozaki, Y., Thomson, J., Turekian, K.K., 1976. The distribution of 210Pb and 210Po in the surface waters ofthe Paci"c Ocean. Earth and Planetary Science Letters 32, 304}312.

Nozaki, Y., Tsunogai, S., 1976. 226Ra, 210Pb, and 210Po disequilibria in the western North Paci"c. Earthand Planetary Science Letters 32, 313}321.

Park, K.M., 1993. Wet deposition of atmospheric 7Be and 210Pb at the mid-western coast of KoreanPeninsula. M.S. Thesis, Ehwa Women's University.

Shannon, L.V., Cherry, R.D., Orren, M.J., 1970. Polonium-210 and lead-210 in the marine environment.Geochimica et Cosmochimica Acta 34, 701}711.

Suk, M.S., Hong, G.H., Chung, C.S., Chang, K.I., Kang, D.J., 1996. Distribution and transport of suspendedparticulate matter, dissolved oxygen and major inorganic nutrients in the Cheju Strait. Journal ofKorean Society of Oceanography 31, 55}63.

Tanaka, N., Takeda, Y., Tsunogai, S., 1983. Biological e!ect of removal of Th-234, Po-210 and Pb-210 fromsurface water in Funka Bay. Japan. Geochimica et Cosmochimica Acta 47, 1783}1790.

Tsunogai, S., Nozaki, Y., 1971. Lead-210 and polonium-210 in the surface water of the Paci"c. GeochemicalJournal 5, 165}173.

Wei, C.L., Murray, J.W., 1994. The behavior of scavenged isotopes in marine anoxic environments: 210Pband 210Po in the water column of the Black Sea. Geochimica et Cosmochimica Acta 58, 1795}1811.

Wells, J.T., 1988. Distribution of suspended sediment in the Korea Strait and southeastern Yellow Sea:onset of winter monsoon. Marine Geology 83, 273}284.

1064 G.-H. Hong et al. / Continental Shelf Research 19 (1999) 1049}1064