ARTICLE IN PRESS

0278-4343/$ - se

doi:10.1016/j.cs

�Correspondifax: +8610 627

E-mail addre

Continental Shelf Research 25 (2005) 1996–2007

www.elsevier.com/locate/csr

Spatial–temporal distribution of dimethylsulfide in thesubtropical Pearl River Estuary and adjacent waters

Min Hua,�, Lingli Liua, Qiju Maa, Tong Zhua, Xudong Tiana, Minhan Daib

aState Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences,

Peking University, Beijing 100871, PR ChinabMarine Environmental Laboratory, Xiamen University, Xiamen 361005, PR China

Received 5 July 2004; received in revised form 22 February 2005; accepted 15 July 2005

Available online 30 August 2005

Abstract

Three cruises were carried out in the Pearl River Estuary (PRE) and adjacent northern South China Sea (SCS) in July

2000, May 2001 and November 2002 to collect seawater samples. Concentrations of dimethylsulfide (DMS),

chlorophyll a (chl a), nutrients (N, P, Si), salinity and temperature in seawater were measured. The spatial and temporal

distribution of DMS concentrations showed larger fluctuations compared with other estuaries reported in the literature.

The mean DMS concentrations in three cruises ranged from 0.05 nM (nM ¼ 10�9mol l�1) to 52.7 nM (n ¼ 76). The

higher concentrations of DMS were observed at the mouth of the estuary. In wet season, high variations of

environmental salinity might stimulate algae to release more DMSP to adjust osmotic pressure. Most of these DMS

‘hotspots’ were coincident with the area of high chl a concentrations, although no significant correlation between DMS

and chl a was found. The values of DMS/chl a showed a clear trend along the north to the south transect, increasing

sharply from estuary to shelf and open sea. There was no significant correlation between DMS and salinity in the wet

season (July and May), but a significant positive correlation in the dry season. High primary production and more iron

deposition implied that the NE monsoon might influence DMS production in the dry season.

r 2005 Elsevier Ltd. All rights reserved.

Keywords: Dimethylsulfide; Spatial–temporal distribution; Phytoplankton; Nutrients; Pearl River Estuary; South China Sea

1. Introduction

Dimethylsulfide (DMS) is the most abundantvolatile sulfide emitted from the ocean. Once DMS

e front matter r 2005 Elsevier Ltd. All rights reserve

r.2005.07.003

ng author. Tel: +8610 62759880;

59880.

ss: minhu@ces.pku.edu.cn (M. Hu).

is oxidized in the atmosphere, the by-products cancontribute to the acidity of rain, and also theformation of sulfate aerosols, which are a majorsource of cloud condensation nuclei over remoteoceans (Andreae and Crutzen, 1997). DMS isformed from its precursor dimethylsulfoniopro-pionate (DMSP), which is produced by marine

d.

ARTICLE IN PRESS

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–2007 1997

phytoplankton. DMSP is involved in osmoregula-tion in algae and bacteria, and cryoprotection inalgae (Liss et al., 1997; Simo, 2001). The produc-tion of DMSP has been found to vary withphytoplankton species. Prymnesiophytes and di-noflagellates are phytoplankton groups with highintracellular DMSP concentrations (Liss et al.,1997). Generally, DMS concentration is high whenhigh DMSP-producing groups dominate the phy-toplankton population. Diatoms have low intra-cellular DMSP concentrations and it is generallybelieved that diatoms are less important thanPrymnesiophytes and dinoflagellates in DMSproduction. However, in some diatom-dominatedwaters with high biomass, DMS concentrationswere as high as in those dominated by those majorDMSP-producing groups (Iverson et al., 1989). Inaddition, several investigators are now showingthat microzooplankton grazing of particulateDMSP to dissolved DMSP, and bacterial conver-sion of this to DMS is also important. Jones et al.(1998) found that in polar waters increasedmicrozooplankton grazing in diatom-dominatedwaters, may lead to above-average concentrationsof DMS. This does not appear to be the case whenthe biomass was dominated by dinoflagellates insubantarctic waters.Due to the growing human population and

urbanization processes, coastal waters are heavilydisturbed. The amount of natural and anthropo-genic materials, which are carried from land tocoast, has increased by a factor of 1.5–2 over thepast 50 years (Christophe and Fred, 2001).Discharges of industrial and municipal waste havecontributed to the eutrophication of the coastalareas of oceans on a global scale. Red tideoutbreaks are hence most frequent in near-shorewaters. For example, 53 red tides were recordedfrom 1981 to 1992 in the Pearl River Estuary(PRE) (Qian and Liang, 1999; Yan et al., 2001).The eutrophication process changes the chemi-

cal and physical characteristics of seawater. Thesechanges may affect the phytoplankton biomass,species composition and physiological processes.In the northwestern Black Sea, eutrophicationinduced a shift in plankton speciation fromdiatoms to flagellate-dominated populations(Mee, 1992; Lancelot et al., 1987). VandenBerg

et al. (1996) suggested that the trend in eutrophi-cation could be explained by a shift of planktonpopulation to Phaeocystis species in the coastalzones of the North Sea, leading to an increase inDMS emissions. A model about the effects ofeutrophication on DMS production in the south-ern part of the North Sea suggested that theanthropogenic eutrophication of the southernNorth Sea had caused an increase of a factor 2.5in the mean annual emission of DMS to theatmosphere in the period of the 1900–1980(VandenBerg et al., 1996). Andreae (1990) ob-served high DMS concentration in oligotrophicareas of the Sargasso Sea. The culture experimentsconducted by Turner et al. (1988) suggested thatthe cocclithophore Emiliania huxleyi produces lessinternal DMSP in higher nitrate medium. Further-more, if nitrate is added to a culture medium withlow nutrients, the intracellular DMSP concentra-tion will decrease within 24 h. Gage et al. (1997)outlined a pathway of DMSP biosythesis inmarine algae, and explained why nitrogen defi-ciency enhances DMSP production.Estuaries are transition zones between rivers

and marine ecosystems. Pritchard (1967) definedan estuary as a semi-enclosed coastal body ofwater which has free connection with the open sea,and within which seawater is measurably dilutedwith fresh water derived from land drainage. Highconcentrations of DMS have been reported fromseveral estuary plumes (Turner et al., 1996; Simoet al., 1997; Amouroux et al., 2002). A few studiesabout DMS in American and European estuarieshave been reported (Iverson et al., 1989; Cerqueiraand Pio, 1999; Sciare et al., 2002), and all theresearches were focused on the temperate zone.Estuaries in the subtropical zone are quitedifferent from the temperate ones due to abundantprecipitation and higher temperature. Studyingspatio-temporal distribution of DMS in subtropi-cal estuaries will help us to understand thebiogeochemical characteristics of DMS in estuar-ine waters. The PRE has long suffered from heavyeutrophication pressures (Huang et al., 2003).Studying DMS spatial and temporal distributionin this area may also give us some insight in howhuman activities can modify the natural cycleof DMS.

ARTICLE IN PRESS

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–20071998

2. Study area

2.1. Background information of the study area

The Pearl River is considered the second largestriver in China, next to the Yangtze River, and 13thlargest in the world by discharge volume(3.3� 1011m3 yr�1). The PRE is situated at thesouth of the Tropic of Cancer with an annualaverage temperature of 22 1C and annual averageprecipitation of 1470mm. The Pearl River hasthree principal tributaries, namely the XijiangRiver, Beijiang River and Dongjiang River, andthe waters discharge into the South China Sea(SCS) through eight outlets. About 50–55% of thewater enters Lingdingyang Bay through the foureastern outlets (Humen, Jiaomen, Hongqimen andHengmen). Lingdingyang Bay is the biggestestuary bay of the Pearl River Delta (PRD), andalso the area where this present study was carriedout (Fig. 1) (Han, 1998).The PRD includes several municipalities such as

Guangzhou, Shenzhen, Hong Kong, Macao, witha population approaching 50 million. PRD hasbeen the fastest-growing economic area of China

Fig. 1. Map of study area and loc

during the past two decades. Environmentalpollution in the region has become worse in theseyears due to rapid industrialization and urbaniza-tion (Zhang et al., 1999). Already, a large amountof domestic, industrial and agricultural sewage isdischarged into the estuary. The major pollutantsinclude inorganic nitrogen, phosphate and lead.The PRE has become one of the most heavilypolluted marine areas in China. A large area of thenorthern SCS can be affected by the discharges ofthe Pearl River due to the combination of largedischarge volume and the high nutrient concentra-tions (Huang et al., 2003; Cheung et al., 2003).

2.2. Description of study area

The study was conducted in Lingdingyang Baycommonly called PRE, and its outer waters within20–231N and 113–1161E (Fig. 1). Following thecriteria of Iverson et al. (1989), the study area wasbe divided into three sections: (1) estuarine section,where salinity is lower than 30, from sta. E1 to sta.E18; (2) shelf section, which is between the 200misobath and the seaward boundary of the estuary,from sta. S1 to sta. S7; (3) open sea section, which

ations of sampling stations.

ARTICLE IN PRESS

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–2007 1999

is located at the south of 200m isobath, from sta.O1 to sta. O4.

3. Sampling and analyses

Samplings were conducted during three cruiseson board R/V Yanping II in July 2000, May 2001and November 2002, respectively. Sampling sta-tions are shown in Fig. 1. According to thetopography of study area, the inner LingdingyangBay is in the north of 22.301N and samplingstations E1 to E18 are located in this area; Sta. S2lies in the connection of the PRE and the northernSCS; the stations in south of Sta. S2 are innorthern SCS. The cruise of July 2000 lasted for 13days from July 12 to July 25. The 2001 cruiselasted for 16 days from May 14 to May 30. The2002 cruise lasted for 13 days from November 3 toNovember 16. Generally, during the three cruises,the vessel first got Sta. S2, and then sampled waterin the estuary from south to north until arriving atthe mouth of Pearl River (from sta. E18 to sta.E1). After finishing the sampling in the PRE, thevessel went back to Sta. S2, and then measure-ments were made from north to south until to theopen sea.Surface seawater was collected at the depth of

�1m with Goflo samplers on a rosette equippedwith CTD and then seawater was stored inpolyethylene bottles (100ml) without headspace.Then samples were immediately stored in the darkat �18 1C in a freezer until analysis at thecontinental lab (within 1 week after the cruise).The recovery of the storage test was 8076%(n ¼ 5). Seawater samples were analyzed for DMSusing the cryogenic purge-and-trap technique(Andreae and Barnard, 1983; Bates et al., 1987;Burgermeister et al., 1990). The determinationmethod was described in detail in Hu et al. (1997).In brief, a seawater sample of 5–25ml was pre-concentrated by purging with pure nitrogen at theflow of 35–40mlmin�1 for 30min, and then thecryogenic was trapped by passing through a U-tube packed with Chromosorb R in liquid nitro-gen. DMS was analyzed by a Shimadzu 8A gaschromatography, equipped with Carbopack Bpacked column (2m� 3mm Teflon column, Car-

bopack B/1.5% XE-60/1.0% H3PO4 60–80) andflame photometric detection. Temperature rampprogram (initial 50 1C for 5min and final 130 1Cwith 32 1Cmin�1 ramp) gave a retention time forDMS around 7.1min. Peak areas were recordedwith a HP 3395 integrator. The detection limit forDMS was 0.5 ng DMS. The linear detection rangeis from 0.5 to 15 ng DMS. The precision of thismethod was within 10%.The shipboard temperature and salinity were

determined with an SBE-19-plus Conductivity–Temperature–Depth/Pressure (CTD) unit. Thedetermination of chlorophyll a (chl a) was carriedout according to the spectrophotometric method(Parsons et al., 1984). Phosphate (PO4

3�) andSilicate (H2SiO3, hereafter SiO3

2�) were determinedcolorometrically using a flow injection analyzer(Tri-223 autoanalyzer) provided by S.C. Pai of theNational Taiwan University. Nitrate (NO3

�) plusNO2

� was measured by reducing nitrate to nitriteusing the same Tri-223 autoanalyzer with an on-line Cd coil. The precision is 1.9% for PO4

3�, 0.1%for NO3

�, 0.2% for NO2� (Cai et al., 2004) (The

data of surface temperature, salinity, chl a,phosphate, silicate and nitrate were provided byXiamen University).

4. Results

4.1. Hydrographic parameters

A strong seasonal variation of salinity isobserved in the estuary. Due to a large volumeof freshwater in May and July (wet season), theestuary is dominated by water of high temperatureand low salinity. In November (dry season), theinfluence of fresh water from the Pearl Riverdecreases, and more offshore high-salinity waterintrudes into the estuary. The estuary in thatperiod is dominated by water of high salinity andlow temperature. In the shelf and open sea area,the seasonal variations of salinity and temperatureare small (Fig. 2a and b).The Pearl River transports large amounts of

nutrients in dissolved or particulate form. Threemain nutrients, dissolved inorganic nitrogen(DIN), phosphate and silicate were measured

ARTICLE IN PRESS

0

5

10

15

20

25

30

35

40

2020.52121.52222.523latitude (°N)

latitude (°N)

Sal

inity

July 2000

May 2001

Nov. 2002

(a)

10

15

20

25

30

35

2020.52121.52222.523

Sur

face

sea

tem

pera

ture

(°C

)

(b)

July 2000

May 2001

Nov. 2002

Fig. 2. (a) Surface salinity and (b) surface temperature as

function of latitude.

DIN

(µ

mol

l-1)

Pho

spha

te (

µ m

ol l-1

) S

ilica

te (

µ m

ol l-1

)

July 2000

May 2001

Nov. 2002

120.0

80.0

40.0

0.0

1.5

1.0

0.5

0.0

200.0

150.0

100.0

50.0

0.0

23.0 22.0 21.0 20.0

Lat. (°N)

Fig. 3. Surface nutrient concentrations vs. latitude of the three

cruises: (a) DIN, (b) phosphate and (c) silicate.

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–20072000

during the cruises. The concentrations of the threenutrients in the estuary were very high upstreamand decreased rapidly in the mouth of estuary(near 221N) (Fig. 3a–c). The DIN, phosphate andsilicate show significantly negative correlationswith salinity (Table 1). The average mole ratio ofN to P ranged from 66 to 204 in different seasons,indicating P limitation. The average mole ratio ofN to Si ranged from 0.5 to 0.6, indicating potentialN limitation, especially in the oceanic waters.

4.2. Spatial and temporal distribution of DMS

concentration in surface water

As described in Section 2.2, the study transect isdivided into three sections: estuarine section, shelfsection and an open sea section. Fresh waterdischarged from Pearl River influences the loca-tion of the estuarine section’s seaward boundary,which is closer to the land in November (dryseason) than in May and July (wet season). The

surface water values of DMS concentration, chl a

concentration and salinity are plotted vs. latitudefor the three cruises are given in Fig. 4a–c. Theaverage concentrations of DMS and chl a in thethree sections (estuary, shelf and open sea) aregiven in Table 2.Fig. 4a shows the data collected during July

2000 cruise. DMS concentration in surfacewater ranged from 0.05 to 25.9 nM. The highest

ARTICLE IN PRESS

05

10152025303540

2020.52121.52222.523

Lat. (°N )

DM

S (

nM)

and

salin

ity

05101520253035

Chl

a (

ug l-1

)C

hl a

(ug

l-1)

Chl

a (

ug l-1

)

Estuary Shelf Open sea

(a)

0102030405060

2020.52121.52222.523Lat. (°N)

Lat. (°N)

DM

S (

nM)

and

salin

ity

0246810121416

Estuary Shelf Open sea

(b)

05

10152025303540

2020.52121.52222.523

DM

S (

nM)

and

salin

ilty

01234567

Estuary Shelf Open sea

(c)

SalinityDMSChl a

Fig. 4. DMS concentration and salinity vs. latitude in three

cruises: (a) July 2000, (b) May 2001 and (c) November 2002.

Table 1

Correlation coefficients between parameters of nutrients and salinity

Correlation coefficients July 2000 May 2001 Nov. 2002

r (DIN) �0.93 (n ¼ 23) �0.74 (n ¼ 24)

r (Phosphate) �0.88 (n ¼ 23) �0.82 (n ¼ 25) �0.72 (n ¼ 29)

r (Silicate) �0.80(n ¼ 23) �0.92 (n ¼ 25) �0.98 (n ¼ 24)

All correlations are significant at 0.01 level (2-tailed).

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–2007 2001

concentrations of DMS were observed at themouth of the estuary, with a sharp increase fromabout 2 to 22 nM. The maxima of DMS concen-trations coincided with the maxima of the surfacechl a at the mouth of the estuary. However, thistrend does not hold in the low salinity region. Theresults from the cruise of May 2001 are shown in

Fig. 4b. The highest average DMS concentration(9.1 nM) was found during the three cruises, withthe largest variation (0.7–52.7 nM). The maxima ofboth DMS and chl a concentrations were observedat the mouth of estuary; the pattern of DMS,however, was not entirely consistent with that ofchl a. The data distribution from the cruise ofNovember 2002 is shown in Fig. 4c. The wholecruise exhibited less variation of DMS concentra-tion between stations than those in May and July,and lower DMS concentrations were measured,with an average of 4.2 nM (ranged from 1.36 to8.24). As in the previous cruise, a maximum of5.6 nM was also found at the south boundary ofthe estuary.Comparing the salinity of three cruises, in the

estuarine section, a more irregular distribution isfound along the estuary in the wet season (Mayand July) than that in the dry season (November).The DMS concentration in the estuary section washigher in average with greater variation in Mayand July (Table 2). However, in November DMSlevels increased steadily from north to south in theestuary, along with an increase in salinity. Spear-man’s correlation analysis suggested that there wasno significant correlation between DMS andsalinity in July and May in the estuarine section.However, in November DMS showed a significantpositive correlation with salinity (r2 ¼ 0:79,N ¼ 15).The average DMS concentrations in the shelf

section are about the same or higher than those inthe estuary, although the average concentrationsof chl a are much lower compared to the estuary,which means that the DMS production in the shelfsection is probably more influenced by phyto-plankton species than by biomass. This situation isdifferent from the results of the estuarine section,

ARTICLE IN PRESS

Table 2

The average concentrations of DMS and Chl a in estuary, shelf, and open sea of the three cruises

Area DMS (nM) Chl a (mg l�1)

July 2000 May 2001 Nov 2002 July 2000 May 2001 Nov 2002

Estuary Mean 4.5 8.6 3.0 7.1 3.8 1.1

Std. deviation 6.8 14.8 1.1 8.5 2.1 1.3

N 15 15 17 15 15 17

Shelf Mean 3.8 14.4 6.3 0.2 2.9 0.2

Std. deviation 1.7 17.0 1.4 0.2 5.9 0.2

N 5 6 7 5 6 7

Open sea Mean 0.9 2.9 5.6 0.1 0.2 0.1

Std. deviation 0.6 2.0 1.9 0.01 0.1 0.02

N 3 4 4 3 4 3

Total Mean 3.9 9.1 4.2 4.7 3.0 0.7

Std. deviation 5.6 14.2 2.0 7.6 3.4 1.1

N 23 25 28 23 25 27

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–20072002

where generally most DMS ‘hotspots’ coincidewith high chl a concentration areas.In May and July, the lowest average DMS

concentration was found in the open sea section,where chl a concentration is much lower, probablybecause of less productivity due to very lownutrient concentrations. Nevertheless phytoplank-ton species exist in the open sea section that canproduce high DMS emissions. For instance, prettyhigh DMS concentration was observed in the opensea section in November, and the chl a concentra-tions pattern of this cruise is similar to those inJuly and May. This effect may also be due to themonsoon, as will be discussed later.

4.3. DMS and phytoplankton

Liss (1999) found that there was a roughcorrespondence between areas of high DMSconcentrations and the blooming of algae withhigh intracellular DMSP concentrations, such ascoccolithophorids. However, a statistical analysisof almost 16,000 measurements of DMS in surfaceseawater showed only a weak correlation betweenDMS and chl a (Kettle et al., 1999). Simo andDachs (2002) have published an empirical algo-rithm between DMS, chl and MLD (climatologicalmixed layer depth). It’s very interesting to

compare our data with their results. But it’s verypitiful that the MLD data were not available for usnow. So we just discuss the correlation betweenDMS and chl a. In our study, no significantcorrelation between DMS and chl a was foundduring the wet reason, but in the dry season astrong negative correlation between DMS and chla was observed (r2 ¼ �0:79, N ¼ 27). Such acorrelation, however, may reflect the spatialvariation of salinity from the estuary to the opensea rather than the relation between the DMSconcentration and the phytoplankton. And nocorrelation was found between DMS and chl a

(r2 ¼ 0:19, N ¼ 27) (corrected for the effects ofsalinity).The large spatial and seasonal variability of

DMS concentration can be caused by changesin the biomass and taxonomic compositionof phytoplankton (Turner et al., 1988). Chl a isused as a standard measure for phytoplanktonbiomass. To find out the spatial distribution ofthe DMS production ability per unit biomass, weused the chl a-normalized DMS concentration(DMS/chl a).The values of DMS/chl a showed a clear trend

along the north to south transect, increasingsharply from the estuary to the shelf and the opensea (Fig. 5a–c). The highest DMS/chl a values were

ARTICLE IN PRESS

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–2007 2003

found in the shelf section in all seasons. Thevariation of DMS/chl a was also greatest in theshelf area (Table 3). The average DMS/chl a of theNovember 2002 cruise exceeded those of other twocruises in wet season by a factor of 1.9–3.2, and

60.0

40.0

20.0

0.0

EstuaryShelfOpen sea

100.0

80.0

60.0

40.0

20.0

0.0

260.0

240.0

80.0

60.0

40.0

20.0

0.00.0 10.0 20.0 30.0 40.0

Lat. (°N)

(a)

(b)

(c)

Fig. 5. DMS/chl a vs. salinity: (a) July 2000, (b) May 2001 and

(c) Nov. 2002.

Table 3

Average DMS/Chl a values in the three sections of each cruise

PRE, July 2000 PRE, May 2001 PRE, Nov

Estuary 1.672.4 6.2714.0 4.572.9

Shelf 32.0718.8 28.6730.9 60.5779.1

Open sea 14.578.0 20.1711.4 47.376.4

also showed larger variability. DMS/chl a ratios inthe shelf water exceeded by 4.5–20 times thosevalues in estuarine water, and 1.3–2.2 times thosein open seawater. These results are not consistentwith those found in North American estuaries(Iverson et al., 1989), where the highest DMS/chl a

ratios were found at open sea stations, andthe average value were much lower than PRE(Table 3), which may be a reflection of thedifference in phytoplankton population and nu-trient conditions.DMS/chl a values in the estuary were larger and

more variable, in the cruise of May 2001 comparedto the cruise of November 2002, where DMS/chl a

ratios increased steadily with the salinity. Signifi-cant correlations between DMS/chl a and salinitywere found in November (r2 ¼ 0:826, P ¼ 0:00,N ¼ 16) and July (r2 ¼ 0:588, P ¼ 0:017, N ¼ 16),but not in May (r2 ¼ 0:226, P ¼ 0:339, N ¼ 16).

5. Discussion

5.1. The influence of fresh water in the Pearl River

Estuary

Limited studies on DMS production in estuarineecosystems are found in the literature, and most ofthem have been carried out in temperate regions(Iverson et al., 1989; Cerqueira and Pio, 1999;Sciare et al., 2002). Previous studies regardingDMS distribution in the SCS were conducted inopen sea regions (Yang et al., 1999; Yang, 2000).In order to investigate DMS production in asubtropical estuary, our work is focused on thespatial and seasonal distribution of DMS in thePRE and adjacent SCS. We found that higherDMS concentrations and greater seasonal varia-tions are the distinctive features in Pear River

2002 Delaware Bay

(Iverson et al., 1989)

Chesapeake Bay (Iverson

et al., 1989)

0.370.2 0.870.6

3.673.4 2.471.9

15.174.7 24.0713.7

ARTICLE IN PRESS

Table 4

Comparing average DMS concentration in PRE with other estuaries

Region Concentration range (nM) Average concentration (nM) Reference

Estuary in North America 1–18 Iverson et al. (1989)

Scheldt estuary 0–2.5 0.4–0.6 Sciare et al. (2002)

Gironde estuary 0–1.7 0.2–0.7

Elbe Estuary 0–2.5 0.9

Ems 0–1.5 0.2

Rhine estuary 0–10

Loire estuary 0.5–3.6 1.3

Canal de Mira 0–18 2.9–5.3 Cerqueira and Pio (1999)

Pearl River Estuary 0.05–56.7 3.0–8.6 This work

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–20072004

Estuary compared to other studies in temperatezones (Table 4).The salinity of the PRE is strongly influenced by

seasonal patterns of rainfall. Because of thecomplex topography of the PRE, fresh waterenters the SCS from eight channels with variousvolumes. The salinity distribution of the surfacewater is thus complicated and fluctuates greatly.The ‘hotspots’ of DMS production are located atthe southern boundary of the estuary, where thesalinity increased sharply from about 15–30 and asalinity front can be observed. On the continentalshelf of New Zealand, the same phenomenon isalso observed: a sharp increase of DMS coincideswith a surface lens of freshwater (Walker et al.,2000). In Zuari estuary, Goa (India), Shenoya andPatilb (2003) observed the highest DMS concen-trations from July to August; meanwhile, thesalinity showed highest variations during their 14-month study. Laboratory studies of Vairava-murthy et al. (1985) suggested that the DMSproduction from algae increased with increasingsalinity of medium. If algal cells, growing in ahigh-salinity medium, were transferred to a low-salinity medium, DMS was released into themedium, and the DMS amount increased with anincreasing salinity difference between the med-iums.The variation of environmental salinity in an

estuary may act as a kind of stimulation to DMSproduction. In response to the change in salinity,algae release more DMS to adjust the osmoticpressure. A total of 70–80% of the total annualwater of the Pearl River discharge enters the SCSin the wet season (April–September), and 20–30%

in dry season (October–March). July is normallythe river’s flood season, but the salinity front inJuly 2000 was closer to the river exit than in May2001. This may be explained by heavy precipita-tion experienced in the watershed during May2001 (Zhai et al., 2005). The greatest variation ofDMS concentrations in the estuary as well as thehighest average DMS concentration and highestDMS/chl ratios were all found in May 2001. Suchlarge variations of DMS distribution in the wetseason may be due to the influence of largevariations in salinity. However, since we have nocontinuous measurements of both DMS andsalinity in any station, it is impossible to draw afinal conclusion.The volume of fresh water may also affect the

speciation of the phytoplankton. In the SwanRiver estuary, Western Australia, dinoflagellatesand marine diatoms dominate during periods oflow flow, whereas the faster-growing freshwaterdiatoms were associated with higher flow rates(Chan and Hamilton, 2001). Most of the algaewith high DMS production rates belong to themarine phytoplankton, such as dinoflagellates.With the influence of fresh water diminishing inthe dry season, DMS producing algae will possiblyenter the estuary. This may explain why in dryseason DMS/chl a values are high even with lowbiomass and lower temperatures.

5.2. High DMS concentrations and high DMS/chl

a values in the shelf section

The average DMS concentrations in the shelfsection are higher than those in the estuary, and

ARTICLE IN PRESS

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–2007 2005

the highest DMS/chl a values with the greatestvariation were observed in both the wet anddry season. Various processes could causethe observed patterns of DMS concentrations onthe shelf. Firstly, the phytoplankton populationstructure and biomass play an important role inDMS production. In shelf water the compositionof the phytoplankton community is different fromthat in the PRE, where normally it is dominated byfresh water, and brackish water algae, such asChaetoceros affinis Laude, Skeletonema costatum

(Huang et al., 1997). These species are low DMSproduction algae. While in the shelf water of theSCS, the biomass of high DMS productionphytoplankton, such as dinoflagellates, is in-creased.Secondly, the transparency of the estuary is

lower than 1m (Han, 1998). The suspendedmaterials may significantly affect the photosyn-thetic process due to radiation limitation. Experi-ments indicate that exposure to UV radiation willlead to a 10–25% increase in the per-cell amountof DMSP of E. huxleyi. And the intra-cellular DMSP concentration is always higher inphotosynthetically active radiation (PAR)+UV-exposed E. huxleyi than in PAR-exposed E.

huxleyi (Slezak and Herndl, 2003). Due tothe lower concentration of suspended material,the UV radiation intensity is higher in the shelfthan in the estuary, and hence the primaryproduction in shelf water is still higher than inthe estuary (Zhang et al., 1999). These processesmay lead to higher DMS/chl a values in the shelfarea.Thirdly, DMSP is involved in osmoregulation in

algae. Both laboratory studies and fieldworkindicate that high nitrate concentrations mayreduce DMSP/DMS production, while low nitrateconcentration will stimulate DMSP/DMS synth-esis in algae (Turner et al., 1988; Andreae, 1990;Leck et al., 1990). Because of human emissions,the average DIN concentration in the PRE ismuch higher than in the shelf section. In July 2000and May 2001, respectively, the DIN concentra-tion in the estuary was 14 and 4 times higher thanin the shelf. The lower nitrate in the shelf sectionmight induce higher DMSP production in thealgae, and then higher DMS emission.

5.3. The influence of the monsoon in the open sea

Seasonal monsoons are a climatic feature of thePRE and SCS. The NE monsoon prevails (withstronger winds) in winter, and the SW monsoon(weaker winds) dominates in summer (Han, 1998;Yin, 2002) The highest DMS and DMS/chl a in theopen sea were observed in November, althoughthe average chl a concentration was lower than inMay or July. This indicates that more active DMSproduction process occurred in November. Zhanget al. (1999) suggested that Fe2+ is a limitingfactor for phytoplankton photosynthesis in SCS.Generally, most Fe is carried from land byatmospheric transport. Several studies indicatethat iron fertilization will significantly increasethe DMS production in the open sea (Martin et al.,1994; Turner et al., 1996). Fieldwork indicatedthat, biological activity in surface water in the SCSis enhanced by monsoon winds, for instance, thehighest primary production was observed inwinter. This may due to the NE monsoon carriessuspended dust from the northern Chinese main-land (Chen et al., 1998). When the high Fecontaining dust is deposited into the area, phyto-plankton metabolism and DMSP/DMS produc-tion were stimulated. Higher iron depositioncausing greater primary production may offer anexplanation as to how the monsoon can influencethe DMS production.

6. Conclusions

PRE and adjacent areas display different DMSand DMS/chl a distribution patterns comparedwith other estuaries, possibly due to differentclimatic influences, and anthropogenic pollution.The large amount of precipitation during the wetseason causes high salinity variation in the estuaryand the fluctuation may stimulate DMS emissionsby the phytoplankton. To regulate the osmoticequilibrium, algae have to produce or releaseDMSP, the precursor of DMS. Both phytoplank-ton biomass and species composition influenceDMS concentration in the surface water. Highnutrient content may lead to high DMS concen-trations in PRE and shelf area, while the NE

ARTICLE IN PRESS

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–20072006

monsoon may enhance DMS production in theopen sea area.

Acknowledgments

We would like to Dr. Zhai, Weidong, Huang,Tao and their colleagues for supplying the salinity,temperature and nutrient data. We are grateful tothe Ocean Carbon group from Xiamen Universityfor the help during sampling. We are indebted toChina National Nature Science Foundation forfinancial support (#20177002, #20131160731,#30230310). The sampling cruise was supportedby CNNSF through Grants # 49825111,#40176025 and #49976021. We thank Dr. Slanina(Professor of the Netherlands Wageningen Uni-versity, Professor of Peking University) for hissuggestions and comments.

References

Amouroux, D., Roberts, G., Rapsomanikis, S., Andreae, M.O.,

2002. Biogenic gas (CH4, N2O, DMS) emission to the

atmosphere from near-shore and shelf waters of the north-

western Black Sea. Estuarine, Coastal and Shelf Science 54,

575–587.

Andreae, M.O., 1990. Ocean–atmosphere interactions in the

global biogeochemical sulfur cycle. Marine Chemistry 30,

1–29.

Andreae, M.O., Barnard, W.R., 1983. Determination of trace

quantities of dimethylsulfide in aqueous solutions. Analy-

tical Chemistry 55, 608–612.

Andreae, M.O., Crutzen, P.J., 1997. Atmospheric aerosols:

biogeochemical sources and role in atmospheric chemistry.

Science 276, 1052–1058.

Bates, T.S., Cline, J.D., Gammon, R.H., Kelly-Hansen, S.R.,

1987. Regional and seasonal variations in the flux of oceanic

dimethylsulfide to the atmosphere. Journal of Geophysical

Research 92, 2930–2938.

Burgermeister, S., Zimmermann, R.L., Georgii, H.-W., Binge-

mer, H.G., Kirst, G.O., Janssen, M., Ernst, W., 1990. On

the biogenic origin of dimethylsulfide: Relation between

cholorophyll, ATP, organismic DMSP, photoplankton

species, and DMS distribution in Atlantic surface water

and atmosphere. Journal of Geophysical Research 95,

20607–20615.

Cai, W.J., Dai, M.H., Wang, Y.C., Zhai, W.D., Huang, T.,

Chen, S.T., Zhang, F., Chen, Z.Z., Wang, Z.H., 2004. The

biogeochemistry of inorganic carbon and nutrients in the

Pearl River Estuary and the adjacent Northern South China

Sea. Continental Shelf Research 24, 1301–1319.

Cerqueira, M.A., Pio, C.A., 1999. Production and release of

dimethylsulphide from an estuary in Portugal. Atmospheric

Environment 33, 3355–3366.

Chan, T.U., Hamilton, D.P., 2001. The effect of freshwater flow

on biomass and succession of phytoplankton in a seasonal

estuary. Marine and Freshwater Research 52, 869–884.

Chen, J.F., Zheng, L.F., Wiesner, M.G., Chen, R.H., Zheng,

Y.L., Wong, H.K., 1998. Estimations of primary produc-

tion and export production in the South China Sea based on

sediment trap experiments. Chinese Science Bulletin 43 (7),

583–586.

Cheung, K.C., Poon, B.H.T., Lan, C.Y., Wong, M.H., 2003.

Assessment of metal and nutrient concentrations in

river water and sediment collected from the cities in the

Pearl River Delta, South China. Chemosphere 52 (9),

1431–1440.

Christophe, R., Fred, T.M., 2001. Influence of the human

perturbation on carbon, nitrogen, and oxygen biogeochem-

ical cycles in the global coastal ocean. Geochimica et

Cosmochimica Acta 65, 3615–3641.

Gage, D.A., Rhodes, D., Nolte, K.D., Hicks, W.A., Leustek,

T., Cooper, A.J., Hanson, A.D., 1997. A new route for

synthesis of dimethylsulphoniopropionate in marine algae.

Nature 387, 891–894.

Han, W.Y., 1998. Chemical Oceanography in South China Sea.

Ocean Press, Beijing, China.

Hu, M., Tang, X., Li, J.L., Yu, Z.J., 1997. Dimethylsulfide in

sea water in the Gulf of Jiaozhou. Acta Scientiae

Circumstantiae 17 (1), 110–115.

Huang, L.M., Chen, Q.C., Yin, J.Q., Wen, W.Y., He, Y.Q.,

Wang, H.X., 1997. Basic organisms composition and

environment status of the Zhujiang estuary and adjacent

waters. Chinese Marine Environmental Sciences 16 (3), 1–7.

Huang, X.P., Huang, L.M., Yue, W.Z., 2003. The character-

istics of nutrients and eutrophication in the Pearl River

Estuary, South China. Marine Pollution Bulletin 47, 30–36.

Iverson, R.L., Nearhoof, F.L., Andreae, M.O., 1989. Produc-

tion of dimethylsulfonium propionate and dimethylsulphide

by phytoplankton in estuarine and coastal water. Limnol-

ogy and Oceanography 34 (1), 53–67.

Jones, G.B., Curran, M.A.J., Swan, H.B., Greene, R.M.,

Griffiths, F.B., Clementson, L.A., 1998. Influence of

different water masses and biological activity on dimethyl-

sulphide and dimethylsulphoniopropionate in the sub

Antarctic zone of the Southern Ocean during ACE-1.

Journal of Geophysical Research 103 (D13), 16691–16701.

Kettle, A.J., Andreae, M.O., Amouroux, D., et al., 1999. A

global database of sea surface dimethylsulfide (DMS)

measurements and a procedure to predict sea surface

DMS as a function of latitude, longitude, and month.

Global Biogeochemistry 13, 399–444.

Lancelot, C., Billen, G., Sournia, A., Weisse, T., Colljn, F.,

Veldhuis, M.J.W., Davies, A., Wassman, P., 1987. Phaeo-

cystis blooms and nutrient enrichment in the continental

coastal zones of the North Sea. Ambio 16, 38–46.

Leck, C., Larsson, U., Bagander, L.E., Johansson, S., Hajdu,

S., 1990. DMS in the Baltic Sea: annual variability in

ARTICLE IN PRESS

M. Hu et al. / Continental Shelf Research 25 (2005) 1996–2007 2007

relation to biological activity. Journal of Geophysical

Research 95, 3353–3363.

Liss, P.S., 1999. Take the shuttle—from marine algae to

atmospheric chemistry. Science 285, 1217–1218.

Liss, P.S., Hatton, A.D., Malin, G., Nightingale, P.D., Turner,

S.M., 1997. Marine sulphur emissions. Philosophical

Transactions of the Royal Society of London Series B—

Biological Sciences 352, 159–168.

Martin, J.H., Coale, K.H., Johnson, K.S., et al., 1994. Testing

the iron hypothesis in ecosystems of the equatorial Pacific

Ocean. Nature 371, 123–129.

Mee, L.D., 1992. The Black Sea in crisis: a need for concerted

international action. Ambio 21, 278–286.

Parsons, T.R., Maita, Y., Lalli, C.M., 1984. Plant pigments. In:

Parsons, T.R., Maita, Y., Lalli, C.M. (Eds.), A Manual of

Chemical and Biological Methods for Seawater Analysis.

Pergamon Press, Oxford, pp. 101–114.

Pritchard, D.W., 1967. What is an estuary: physical viewpoint.

In: Lauf, G.H. (Ed.), Estuaries. A.A.A.S. Publ. No. 83,

Washington, DC, pp. 3–5.

Qian, H.L., Liang, S., 1999. Study on the red tide in the Pearl

River Estuary and its near waters. Marine Environmental

Science 18, 70–74.

Sciare, J., Mihalopoulos, N., Nguyen, B.C., 2002. Spatial and

temporal variability of dissolved sulfur compounds in

European estuaries. Biogeochemistry 59, 121–141.

Shenoya, D.M., Patilb, J.S., 2003. Temporal variations in

dimethyl sulphoniopropionate and dimethylsulphide in the

Zuari estuary, Goa (India). Marine Environmental Re-

search 56, 387–402.

Simo, R., 2001. Production of atmospheric sulfur by

oceanic plankton: biogeochemical, ecological and

evolutionary links. Trends In Ecology and Evolution 16,

287–294.

Simo, R., Dachs, J., 2002. Global ocean emission of dimethyl-

sulfide predicted from biogeophysical data. Global Biogeo-

chemical Cycles 16 (4), 1078.

Simo, R., Grimalt, J.O., Albaiges, J., 1997. Dissolved

dimethylsulphide, dimethylsulphoniopropionate and di-

methylsulphoxide in western Mediterranean waters. Deep-

Sea Research Part II—Topical Studies in Oceanography 44,

929–950.

Slezak, D., Herndl, G.J., 2003. Effects of ultraviolet and visible

radiation on the cellular concentrations of dimethylsulfo-

niopropionate (DMSP) in Emiliania huxleyi (strain L).

Marine Ecology—Progress Series 246, 61–71.

Turner, S.M., Malin, G., Liss, P., Harbour, D.S., Holligan,

P.M., 1988. The seasonal variation of dimethylsulfide and

dimethylsulfoniopropionate concentrations in nearshore

waters. Limnology and Oceanography 33, 364–375.

Turner, S.M., Nightingale, P.D., Spokes, L.J., Liddicoat, M.I.,

Liss, P.S., 1996. Increased dimethyl sulphide concentrations

in sea water from in situ iron enrichment. Nature 383,

513–517.

Vairavamurthy, A., Andreae, M.O., Iverson, R.L., 1985.

Biosynthesis of dimethylsulfide and dimthylopropiothetin

by Hymenomonas carterae in relation to sulfur source and

salinity variations. Limnology and Oceanography 30,

59–70.

VandenBerg, A.J., Turner, S.M., VanDuyl, F.C., Ruardij, P.,

1996. Model structure and analysis of dimethylsulphide

(DMS) production in the southern North Sea, considering

phytoplankton dimethylsulphoniopropionate- (DMSP)

lyase and eutrophication effects. Marine Ecology Progress

Series 145, 233–244.

Walker, C.F., Harvey, M.J., Bury, S.J., Chang, F.H., 2000.

Biological and physical controls on dissolved dimethylsul-

fide over the north-eastern continental shelf of New

Zealand. Journal of Sea Research 43 (3-4), 253–264.

Yan, T., Zhou, M.J., Zhoum, J.Z., Qian, P.Y., 2001.

Preliminary studies on red tide formation mechanism in

Hong Kong and Pearl River Estuary. Acta Ecology Sinica

21, 1634–1641.

Yang, G.P., 2000. Spatial distributions of dimethylsulfide in the

South China Sea. Deep-Sea Research Part I—Oceano-

graphic Research Papers 47, 177–192.

Yang, G.P., Liu, X.T., Li, L., Zhang, Z.B., 1999. Biogeochem-

istry of dimethylsulfide in the South China Sea. Journal of

Marine Research 57, 189–211.

Yin, K.D., 2002. Monsoonal influence on seasonal variations in

nutrients and phytoplankton biomass in coastal waters of

Hong Kong in the vicinity of the Pearl River Estuary.

Marine Ecology—Progress Series 245, 111–122.

Zhai, W.D., Dai, M.H., Cai, W.J., Wang, Y.C., Wang, Z.H.,

2005. High partial pressure of CO2 and its maintaining

mechanism in a subtropical estuary, the Pearl River estuary,

China. Marine Chemistry 93, 21–32.

Zhang, J., Yu, Z.G., Wang, J.T., Ren, J.L., Chen, H.T., Xiong,

H., Dong, L.X., Xu, W.Y., 1999. The subtropical Zhujiang

(Pearl River) Estuary: Nutrient, trace species and their

relationship to photosynthesis. Estuarine Coastal and Shelf

Science 49, 385–400.