Seawater, atmospheric dimethylsulfide and aerosol ions in the Pearl River Estuary and the adjacent northern South China Sea

Seawater, atmospheric dimethylsulfide and aerosol ions in the Pearl River Estuary and the adjacent northern South China Sea

Journal of Sea Research 53 (2005) 131 – 145 Seawater, atmospheric dimethylsulfide and aerosol ions in the Pearl River ...

648KB Sizes 0 Downloads 61 Views

Journal of Sea Research 53 (2005) 131 – 145

Seawater, atmospheric dimethylsulfide and aerosol ions in the Pearl River Estuary and the adjacent northern South China Sea Qiju Maa, Min Hua,*, Tong Zhua, Lingli Liua, Minhan Daib a

State Joint Key Laboratory of Environmental Simulation and Pollution Control, College for Environmental Sciences, Peking University, Beijing 100871, PR China b Marine Environmental Laboratory, Xiamen University, Xiamen 361005, PR China Received 3 November 2003; accepted 9 June 2004

Abstract The spatial and temporal distribution of dimethylsulfide (DMS) was investigated in surface seawater and in the marine atmosphere in the Pearl River Estuary and northern South China Sea during three cruises in July 2000, May 2001 and November 2002. Sea-to-air fluxes of DMS were subsequently estimated based upon seawater DMS concentration, temperature of surface seawater and wind speed over sea. The seawater DMS concentration of the three cruises ranged from 0.1 to 52.7 nmol l1 (n=76). DMS concentrations showed remarkable spatial and temporal distributions and highest values were observed at the mouth of the Pearl River Estuary. Throughout the study area we observed high levels of DMS in the water with great sea-to-air flux and relatively low levels of atmospheric DMS (1.70F1.16 nmol m3 in May 2001 and 2.25F0.38 nmol m3 in November 2002). Aerosol components, potentially linked with DMS oxidation, were also measured. The atmospheric concentrations of nss-sulfate and nitrate were much higher in the Pearl River Estuary than in the offshore area, with mean values of 12.11 and 4.45 Ag m3 for nss-sulfate, 4.88 and 2.21 Ag m3 for nitrate. Aerosol mass and components’ concentrations decreased from the inner estuary to outer waters. High concentrations of nss-sulfate and nitrate in sea salt particles imply that oxidation of atmospheric DMS is related with anthropogenic sources and heavy ozone, NOx and SO2 pollution in the study area. D 2004 Elsevier B.V. All rights reserved. Keywords: Dimethylsulfide; Sea-to-air flux; Nss-sulfate; Nitrate; Pearl River Estuary; South China Sea

1. Introduction Investigations on the possible climatic role of the biogenic sulfur cycle are numerous (Charlson et al., * Corresponding author. Tel.: +86 10 62759880; fax: +86 10 62759880. E-mail address: [email protected] (M. Hu). 1385-1101/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.seares.2004.06.002

1987; Andreae and Crutzen, 1997). Dimethylsulfide (DMS) is produced by marine phytoplankton species. Once released to the atmosphere, DMS is mainly oxidised by OH (by day) and NO3 (by night) radicals to form various sulphur-containing products, such as non-sea-salt sulfate (nss-SO42), methanesulfonate (MSA) and dimethylsulfone (Berresheim and Eisele, 1998). In polluted areas with high NOx concentrations


Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

the reaction of DMS with NOx results in a coupling between the nitrogen and the sulfur cycles in the atmosphere (Jensen et al., 1991). HNO3 is a main product of this reaction and can be rapidly removed by dry deposition or adsorption onto aerosols or droplets. This is an efficient night time sink for NOx in the marine troposphere. Very high levels of nss-SO42 and NO3 concentrations were measured in aerosols in the East China Sea (Gao et al., 1996). Non-sea-salt sulfate has both biogenic and anthropogenic sources, whereas MSA is thought to derive exclusively from DMS oxidation. Therefore MSA is considered to be a marker of biogenic sources (Savoie and Prospero, 1989). Ocean margins have been identified as a significant source of DMS (Sciare et al., 2002). Estuaries and their plumes in the open sea could also be important sources of atmospheric DMS (Iverson et al., 1989; Turner et al., 1996; Simo et al., 1997). Studies on DMS in American and European estuaries (e.g. Iverson et al., 1989; Cerqueira and Pio, 1999; Sciare et al., 2002) have focused on the temperate zone. Shenoy and Patil (2003) report on temporal variations in DMSP and DMS in a tropical estuary (Zuari Estuary, India). Some work on DMS in seawater has been done in China. Hu et al. (1995) measured the concentration of DMS in surface waters of the Bo Sea and Gulf of Jiaozhou, both belonging to the East China Sea. Both DMS concentrations and sea-air fluxes showed seasonal variations with a maximum in spring and minimum in winter. Uzuka et al. (1996) studied DMS distribution in the temperate coastal zone of the East China Sea. They observed remarkable temporal and spatial variations of seawater DMS concentrations with its fluxes. Seawater DMS concentrations were measured in the East China Sea (Yang et al., 2000) and the South China Sea (Yang, 2000). In surface water of the East China Sea, DMS ranged from1.8 to 5.7 nmol l1. Highest values were observed on the continental shelf with high biological productivity. They found that only zooplankton biomass and nitrate content were closely related to DMS concentrations. DMS concentrations in surface seawater of the South China Sea ranged from 1.9 to 4.6 nmol l1, with a mean of 2.6 nmol l1. However, generally there is little information on DMS in the atmosphere and its oxidation products in

China’s coastal zones. Only Gao et al. (1996) reported on the East China Sea: atmospheric MSA concentrations ranging from 0.029 to 0.066 Ag m3, nsssulfate from10 to 12 Ag m3, and nitrate from 5.6 to 7.7 Ag m3. Moreover, the available information on atmospheric volatile sulfur compounds in Chinese seas is mainly based on observations during one campaign only, the Pacific Exploratory Mission (PEM-West A&B). Airborne studies from Japan, Hong Kong to the US Air Force Base in Guam show that the DMS concentrations in the free troposphere range from 10–120 pptv with a mean value of 65 pptv (Li et al., 1996). Arimoto et al. (1996) reported at Kato (Hong Kong, 16 August 1991 to 5 February 1992) that only 4.8% of total nss-sulfate was from biogenic sources and that the DMS-derived fraction of nss-SO42 is highest in summer. Besides the important role of DMS in the sulfur cycle, CCN formation and the global climate, there are three other reasons to investigate DMS in China at present: Firstly, ocean margins have been identified as a significant source of DMS and several results suggest that estuaries and corresponding plumes in the open sea could represent an important source of atmospheric DMS (Iverson et al., 1989; Turner et al., 1996; Sciare et al., 2002). China has over 18 thousand km of coastline and 0.38 million km2 of sea, but little information is available on DMS in its seas, and hardly any investigation has been done in its estuaries. Secondly, what is the role of DMS in the acid deposition in the Pearl River Delta? As a consequence of the rapid economic development of the east coast, huge amounts of domestic, industrial and agricultural sewage are discharged into the estuaries. The Pearl River Estuary is particularly severely polluted: its water quality is beyond the fourth grade of the National Seawater Quality Standard (medium polluted, only suitable for sea ports and exploitations). The ecological environment in this region has also changed; there is a higher frequency of algal blooms and red tides as well as a decrease in biomass and biological diversity (Hu et al., 2001). These changes will inevitably affect the algae and DMS production in the seawater. The distribution of DMS in the seawater and sea-to-air flux will also be affected. Acid deposition is a major environmental problem in China: 30% of the country is suffering from acid rain

Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

(State of the environment report in China, 2000). The Pearl River Delta in Guangdong Province has a high frequency of acid rain (40.5%); the mean pH of the acid rain in Guangdong Province is 4.64 (Guangdong EPA, 2002 ). In coastal areas the non-sea-salt sulfate aerosol from DMS oxidation in the marine troposphere is the major contributor to the acidity of natural precipitation (Charlson and Rodhe, 1982). Thirdly, the Chinese government have in recent years taken measures to improve the air quality in many cities. The SO2 and NOx concentrations in many areas have decreased. For instance, in Guangzhou city (near the study area) the concentrations of SO2, NO2 and TSP have decreased by 29.7%, 49.5% and 40.29%, respectively, during 1996–2000 (Guangzhou EPA, 2000), and the contribution of biogenic sources to sulfate in the atmosphere should be re-assessed. In this paper we present the concentrations of DMS in the surface water and in the atmosphere and aerosol components of the Pearl River Estuary and the adjacent northern South China Sea. This is the first attempt to investigate DMS levels in estuarine waters in China. The dataset offers a unique opportunity to investigate a possible link between the DMS variation and its oxidation product.


men, Hongqimen and Hengmen) collect about 50– 55% of the total runoff and discharge into Lingdingyang Bay, which is the biggest estuary of the Pearl River. The Pearl River Delta (PRD) is a highly developed area which includes large cities such as Hong Kong, Macao, Guangzhou and Shenzhen. The fast economic development and population growth in the PRD have caused severe water pollution in the Pearl River Estuary. The PRD receives 64% of the industrial sewage and 74% of the domestic sewage of the entire Guangdong Province. The runoff of the Pearl River contains huge quantities of nutrients that affect the northern South China Sea (SCS) directly. In 2002, the discharged amounts of inorganic nitrogen (IN), phosphate and heavy metals were 4.38105 t, 1.46104 t and 3.10103 t, respectively (China Oceanic Environment Quality Report, 2002). The major pollutants in the Pearl River Estuary were IN, phosphate and lead. The air pollution problems in the PRD focus on ozone, particles, NOx and acid rain. In Guangdong Province the annual mean SO2 of 2002 was 22 Ag m3, NO2 was 0.027 mg m3 and pH of rainwater was 4.64. The northern SCS is located on the western rim of the Pacific Ocean and the water quality achieved the second grade of National Seawater Quality Standard (clean seawater), except the area near the PRD.

2. Study area 2.2. Description of study area 2.1. Background information on the study area The Pearl River is the biggest river in southern China and its flux is the second largest in China (only surpassed by the Yangtze River). Its length, drainage area and annual mean flux are 2214 km, 4.52105 km2 and 10 524 m3 s1, respectively. The Pearl River drainage area is located in the subtropical zone with an annual mean temperature of 14–22 8C and annual mean precipitation of 1200–2200 mm. The precipitation is distributed so unevenly over the year that 80% of it falls in the April–September period (wet season) and only 20% in October–March (dry season) (Zhao, 1990). The Pearl River drainage area is made up of the West River, East River, North River and many streams in the Pearl River Delta. All these tributaries flow into the South China Sea through eight outlets. The four eastern outlets (Humen, Jiao-

The study area was Lingdingyang Bay commonly called Pearl River Estuary (PRE) and its outer waters within 20–238N and 113–1168E (Fig. 1). Following the criteria of Iverson et al. (1989), the study area can be divided into three sections: (1) an estuarine section, with a salinity lower than 30, from Station E1 to station E18; (2) a shelf section, between the 200 m isobath and the seaward boundary of the estuary, from station S1 to station S7; (3) an open sea section, south of the 200 m isobath, from station O1 to station O4.

3. Sampling and analysis Sampling was conducted during three cruises on board RV dYanping IIT in July 2000, May 2001, and November 2002, respectively. Sampling stations are


Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

Fig. 1. Map of the study area and locations of sampling stations.

shown in Fig. 1. Sampling stations E1 to E18 are located in the inner Lingdingyang Bay north of 22.308N; St. S2 lies between the PRE and the northern SCS; the stations south of St. S2 are in the northern SCS. The 2000 cruise was from 12 to 25 July (13 days), the 2001 cruise from 14 to 30 May (16 days), the 2002 cruise from 3 to 16 November (13 days). Generally during the three cruises, the vessel first visited St. S2, and then did the sampling in the estuary from south to north up to the mouth of Pearl River (from St. E18 to St. E1). After sampling in the PRE, the vessel returned to St. S2, and then measurements were made from north to south as far as the open sea. The data on water temperature and salinity were collected simultaneously by a CTD (conductivitytemperature-depth). Meteorological data, such as wind speed and air temperature, were obtained from the meteorological station aboard RV dYanping IIT. The determination of Chl-a was carried out according to the spectrophotometric method (Parsons et al., 1984).

based laboratory (within one week of the cruise). The samples were analysed for DMS using the cryogenic purge-and-trap technique (Andreae and Barnard, 1983; Bates et al., 1987; Bu¨rgermeister et al., 1990). The determination method was described in detail in Hu et al. (1997). In brief, a seawater sample of 5–25 ml was pre-concentrated by purging with pure nitrogen at the flow of 35 to 40 ml min1 for 30 min, and then trapped by passing through a U-tube packed with Chromosorb R in liquid nitrogen. DMS was analysed by a Shimadzu 8A gas chromatography, equipped with Carbopack B packed column (2 m3 mm Teflon column, Carbopack B/1.5%XE-60/1.0%H3PO4 60– 80) and flame photometric detection. A temperature ramp program (initial 50 8C for 5 min and final 130 8C with 32 8C /min ramp) gave a retention time for DMS of around 7.1 min. Peak areas were recorded with a HP 3395 integrator. The detection limit for DMS was 0.5 ng DMS. The linear detection range is from 0.5 ng to 15 ng DMS. The precision of this method was within 10%.

3.1. Seawater DMS measurement 3.2. Atmospheric DMS Surface seawater was collected at a depth of ~1 m with Goflo samplers on a rosette equipped with CTD and stored in polyethylene bottles (100 ml) without any air. The samples were immediately stored in the dark and frozen at 18 8C until analysis in the land-

Atmospheric DMS was collected using a vacuum pump, which takes air in through an absorbent cotton oxidant scrubber and sample tubes for 2 h at 0.5 l min1. The inlet of the sample tube was about 10 m

Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

above the seawater surface. Oxidant scrubber materials (NaOH, KOH and KI) were contained in a glass tube of 20 cm length and 2 cm I.D. (Davison and Allen, 1994). The sample tube was made of Teflon or stainless steel of 20 cm length and 4 mm I.D., containing Tenax-TA (80–100 mesh, CHROMPACK). Before sampling, the sample tubes were aged with nitrogen flow at 200 8C for at least 2 h for used tubes and 8 h for new ones. The maximal adsorption amount of one tube for DMS was 36 mg, and its recovery was 80%. The atmospheric samples were analysed in the same way as the seawater samples. DMS was desorbed from Tenax-TA at a temperature of 150F10 8C and a carrier flow of 30 ml/min nitrogen. The calibration of DMS was performed with a DMS permeation tube (VICI Metronics, Inc, Standard service type) and Calibrator Model 8550 (monitor labs Inc.) and by plotting the detector response vs. the trapped amounts of DMS in a log-log scale. The permeation rate of this DMS tube determined gravimetrically was 71.39 ng min1 at 50F0.1 8C.


weighing on an electronic balance (0.00001 g, temperature 20F1 8C and relative humidity 45F5%). Filters were extracted in quartz beakers for 20 min in 20 ml of Milli-Q water in an ultra-sonic bath. Extracts were analysed by ion chromatography (Dionex 600, USA). A Dionex AS14 column with ASRS-I suppressor was used for the analysis of anions (Cl, NO3and SO42), which were determined with 3.5 mmol l1 Na2CO3/1.0 m mol l1 NaHCO3 eluent and a flow rate of 2.0 ml min1. For cations (NH4+ and Na+ etc.) a CS-12A column with a CSRS-I suppressor was used. Separation was achieved under the elution at 1.5 ml min1 of 20 mM MSA. The reproducibility was better than 5%, and the detection limits of Cl, NO3, SO42, NH4+ and Na+ were 0.03, 0.01, 0.01, 0.05, 0.03 Ag m3, respectively. A field blank was also analysed and found to be below the detection limits.

4. Results and discussion

3.3. Sampling and analysis of aerosol

4.1. Spatial and temporal distribution of DMS in the surface waters

The aerosol samples were collected by Micro Orifice Uniform Deposit Impactor (MOUDI, Model 100, MSP Corporation, USA) at a flow rate of 30 l min1. The inlet of MOUDI was about 10 m above the seawater surface. The 50% cut-off diameters of the MOUDI were 18, 10, 5.6, 3.2, 1.8, 1.0, 0.56, 0.32, and 0.18 Am for the eight stages, respectively. Atmospheric particles were collected on Teflon membranes (Geltman Teflok filters, 0.2 Am pore size, 47 mm diameter). The aerosol was sampled for 24–72 h periods. After sampling, the filters were stored in the refrigerator in sealed polyethylene petri dishes until analysis. The particulate mass was determined by

The average concentrations of seawater DMS in the three areas are presented in Table 1. The DMS concentration in the open sea section was lower than those of other two sections. Also the concentrations of DMS in the surface of the estuary and shelf sections were relatively higher than in other estuaries and seas reported in the literature. Uher et al. (2000) reported on the DMS distribution at the European western continental margin. The means of DMS concentrations were 2.8 nmol l1 in September 1994 and 7.2 nmol l1 in July 1995. Sciare et al. (2002) investigated dissolved sulfur compounds during nine oceanographic cruises carried in six major tidal

Table 1 Mean seawater DMS concentration (nmol l1) and DMS/Chla (nmol Ag1) in three areas during the cruises DMS (nmol l1)


Estuary Shelf Open sea

Mean N Mean N Mean N

DMS/Chla (nmol Ag1)

July 2000

May 2001

Nov. 2002

July 2000

May 2001

Nov. 2002

4.1F7.1 13 3.8F1.7 5 0.9F0.6 3

8.6F14.3 16 8.9F7.0 4 2.9F2.0 4

3.1F1.2 18 6.4F1.7 7 5.6F1.9 4

1.2F1.5 13 27.4.0F19.2 5 14.5F8.0 3

2.9F5.9 16 35.6F30.6 4 20.1F11.4 4

4.9F3.2 18 71.1F82.4 6 47.3F16.4 3


Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

European estuaries between July 1996 and May 1998. Very low levels of DMS were reported: only 0.6F0.3 nmol l1. The DMS concentration, salinity and Chl-a concentration vs. latitude in the three cruises is given in

Fig. 2. The stations in the three sections are distributed along the latitude. There was obvious seasonal variation of salinity in the estuary. In July 2000 and May 2001 (wet season) the large volume of freshwater made the boundary of the estuary and shelf sections

Fig. 2. DMS concentration and salinity vs. latitude in the three cruises: (a) July 2000, (b) May 2001, (c) November 2002.

Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

move south of 228N. In November 2002 (dry season) the drainage of freshwater was smaller and more offshore high-salinity seawater intruded into the estuary, so the boundary moved toward the north, closer to the continent. Of the three cruises, the variation of DMS concentrations was largest in May 2001, ranging from 0.7 to 52.7 nmol l1 with an average of 9.1 nmol l1, also the highest average value of the three cruises. The highest variability was observed at the estuary mouth, increasing rapidly from 1.1 to 31.3 nmol l1, and then to 52.7 nmol l1. The maxima of both DMS and chlorophyll-a concentrations were observed at the mouth of the estuary. During the July 2000 cruise, the DMS concentration of the surface water ranged from 0.1 to 25.9 nmol l1 with an average of 3.9 nmol l1. The highest concentrations of DMS were also observed at the mouth of the estuary, with a sharp increase from 2.2 to 25.9 nmol l1. Maximum DMS concentrations coincided with maxima of surface Chl-a. During the November 2002 cruise, DMS concentrations ranged from 1.4 to 10.0 nmol l1 with an average of 4.2 nmol l1. There was no obvious trend in the variation of DMS concentrations found at the estuary mouth. Throughout the November 2002 cruise, DMS concentrations were less variable than during the other cruises. The concentration gradient over the three areas (estuary, shelf, open sea) was observed to differ among the three cruises. In July 2000 the highest mean DMS concentration occurred in the estuary. During the other two cruises, however, the highest mean DMS concentration was found in the shelf section. Chlorophyll-a was used as a standard measure of phytoplankton biomass. To investigate the spatial distribution of the DMS production ability per unit biomass, we calculated chlorophyll-normalised DMS concentrations. There was also a difference between the distribution of DMS concentrations and that of DMS/Chl-a. The gradient of DMS/Chl-a could be clearly seen and the highest DMS/Chl-a values were found in the shelf section for all three cruises in the three seasons, where the DMS/Chl-a values of the sampling stations also exhibited the greatest variations (Table 1). Firstly, the phytoplankton population structure and biomass play an important role in DMS production. In shelf waters, the composition of the phytoplankton community is different from those in the PRE, where brackish-water algae normally


dominate, such as Chaetoceros affinis Laude, Skeletonema costatum (Huang et al., 1997). These species are low DMS production algae. However, in the shelf water of the South China Sea, the biomass of high DMS production phytoplankton, such as dinoflagellates, is increased. The transparency of the estuary is less than 1 m (Han, 1998). Secondly, the suspended materials may significantly affect the photosynthetic process due to radiation limitation. Experiments indicate that exposure to UV radiation will lead to a 10 to 25% increase in the per-cell amount of DMSP in Emiliania huxleyi, and that the intracellular DMSP concentration is always higher in PAR (photosynthetically active radiation) + UV-exposed E. huxleyi than in PAR-exposed E. huxleyi (Slezak and Herndl, 2003). Due to the lower concentration of suspended material in the shelf water, the UV radiation intensity is higher there than in the estuary, and hence the primary production in the shelf is even higher than in the estuary (Zhang et al., 1999). These processes may lead to higher DMS/ Chl-a values in the shelf area. 4.2. Calculation of DMS sea-to-air flux The flux of DMS from sea to air, F, is given by the formulation of air-sea gas exchange of Liss and Slater (1974)  F ¼ K Cg =H  Cl ð1Þ where K is the DMS transfer velocity, H is Henry’s law constant; Cg and Cl are DMS concentrations in the gas and liquid phase, respectively. Since aqueous DMS concentrations are typically greater than atmospheric concentrations (Berresheim et al., 1991), the flux can be approximated by F ¼  K  Cl


where K (m d1) can be parameterised in terms of wind speed and seawater temperature according to the formula of Liss and Merlivat (1986) K ¼ 0:17½AðTÞ2=3 d u

for uV3:6m s1


K ¼ 0:17½AðTÞ2=3 d u þ 2:68½AðTÞ1=2 d ð u  3:6Þ for 3:6bub13m s1



Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

K ¼ 0:17½AðTÞ2=3 d u þ 2:68½AðTÞ1=2 d ðu  3:6Þ 1=2

þ 3:05½AðTÞ

concentration on the shelf was lower than in the estuary. In November 2002, the shelf flux was the highest in all three sections but the estuary flux was two times greater than the open sea flux, while the open sea DMS concentration was two times the estuary value. This suggest that the transfer velocity(K) is as important as the seawater concentration to evaluate the sea-to-air flux.

ðu  13Þ for uz13m s1 ð5Þ

where u is the wind speed (m s1); A (T) is the ratio of the DMS Schmidt number at T 8C and CO2 Schmidt number at 20 8C (equal to 595). Sc DMS(T) / ScCO2(20)=ScDMS(T)/595; ScDMS(T): the Schmidt number of DMS at temperature T(8C) is calculated by the equation from Saltzman et al. (1993): ScDMSðTÞ ¼ 2674:0  147:12T þ 3:726T

4.3. Atmospheric DMS



Atmospheric DMS samples were collected during two cruises in May 2001 and November 2002 (see Table 3). In May 2001, the DMS concentration in the air increased from the estuary to the open sea, but in November 2002 the concentration distributions were similar in all three areas, with a total mean of 2.25F0.38 nmol m3. In the eastern Mediterranean, the mean concentration of atmospheric DMS was 2.35F1.38 nmol m3 with highest values in summer (3.73F0.81) and lowest in winter (0.87F0.32) (Kouvarakis and Mihalopoulos, 2002). There are reports of 5.36F3.04 nmol m3 DMS during the Albatross campaign in the Atlantic Ocean (Sciare et al., 2000) and of 17.63F12.72 nmol m3 at Amsterdam Island during summer (Sciare et al., 2001). Even at Palmer Station, a remote ocean location in Antarctica, the atmospheric DMS concentration was as high as 5.31F4.20 nmol m3 (Berresheim et al., 1998). Comparing our data with these literature data shows that our atmospheric DMS concentrations were at the same level as those in the Mediterranean, but lower than in those remote areas. The major factors controlling the DMS in the atmosphere are the sea-to-air flux (strength of source,


The calculated sea-to-air flux of DMS and other parameters are shown in Table 2. Overall the flux levels in the estuary and the open sea were comparable with literature values. Turner et al. (1996) calculated a DMS flux in the North Sea of 5.9 Amol m2 d1 in summer and 1.8 Amol m2 d1 in winter. The winter and summer average values in the coastal zone of the East China Sea were 6.1 and 1.5 Amol m2 d1, respectively, with a mean of 3.8 Amol m2 d1 (Uzuka et al., 1996). On the West European continental margin, fluxes of 6.6 Amol m2 d1 were found in September and 14.4 Amol m2 d1 in July (Uher et al., 2000). The DMS flux on the shelf was relatively large compared with those found in the other two areas. The high DMS concentrations in the water caused large sea-air fluxes. The wind speeds did not vary greatly during all three cruises, so there was only a limited difference in K values of a factor of five, but the mean concentration of DMS in the water could vary about ten fold. In July 2000, the shelf flux was greater than the estuary flux, but the DMS

Table 2 The calculated Sc and K values and sea-to-air fluxes of DMS in the study area Period


DMS concentration (nmol l1)

Temperature (8C)


Wind speed (m s1)

K (m d1)

Flux (Amol m2d1)

Std. Dev. (Amol m2d1)

July 2000

Estuary Shelf Open Sea Estuary Shelf Open Sea Estuary Shelf Open Sea

4.1 3.8 0.9 8.6 8.9 2.9 3.1 6.4 5.6

30.1 28.9 29.3 27.4 27.2 29.3 24.7 23.2 22.0

585.2 617.2 606.4 658.6 664.7 606.4 739.5 790.4 836.7

4.6 7.7 4.5 5.1 5.6 4.5 6.8 6.5 4.3

3.41 12.20 3.14 4.67 6.03 3.14 8.67 7.69 2.21

14.0 46.4 2.8 40.2 53.7 9.1 26.9 49.2 12.4

21.3 20.7 1.9 66.8 42.2 6.3 10.4 13.1 4.0

May 2001

Nov. 2002

Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145 Table 3 The mean concentrations of atmospheric DMS in the study area (nmol m3) Period



Atmospheric DMS concentration

Std. Deviation

May 2001

Estuary Shelf Open Sea Total Estuary Shelf Open Sea Total

9 5 8 22 1 7 6 14

0.81 1.45 2.85 1.70 2.81 2.12 2.30 2.25

0.75 1.29 2.44 1.16

Nov. 2002

0.58 0.41 0.38

influenced by the local wind), mixing and dilution in the marine boundary layer (MBL) and loss by chemical reactions with oxidants (OH and NO3 radicals) (Sciare et al., 2001). The seawater DMS concentrations and DMS sea-to-air fluxes reported in this study were relatively large compared with the other relevant ocean areas. Few studies have investigated the vertical distribution of DMS in the MBL or examined the possible impact of the MBL height on the DMS mixing ratios measured at ground level (Sciare et al., 2001). Although the variation in dilution induced by changes in the MBL height should influence the DMS levels, no clear relationship has been established between atmospheric DMS and MBL height, nor was a clear seasonal trend found at Amsterdam Island (Sciare et al., 2001). In general, the height of the MBL over the ocean varies much less than over a continent. Therefore, and also for lack of data on MBL height, the influence of variation in MBL height was ignored. High atmospheric DMS concentrations were expected in the study area, in view of the large emission fluxes. However, the concentrations actually observed were rather low, suggesting that the rate of DMS removal from the marine boundary is really fast in our study area. On the base of three-day backward trajectory analysis, the quick DMS oxidation in the atmosphere could be partly explained by the origin of air masses reaching the sampling stations as illustrated in Fig. 3. The sampling stations in the PRE and the shelf during the sampling period were influenced by the air parcels from the continent. Firstly, the continental flux of DMS was very low, and secondly the continental air was more polluted. As mentioned, major atmospheric pollutants in the Pearl River Delta are ozone, particles


and NOx, which may help the formation of OH radicals by day and NO3 radicals by night. The formation of oxidants in the air of the PRE was heavily influenced by anthropogenic sources, which can be confirmed by the high level of nss-sulfate and nitrate (discussed in Section 4.4.). Average O3 concentrations at Tung Chung Station of Hong Kong (22.298N, 113.938E), near the mouth of PRE, were about 60 Ag m3 in May 2001 and 75 Ag m3 in November 2002. PRE and the adjacent SCS are located in the subtropical zone. Enough radiation and ozone concentration in May can cause high OH radical concentrations (Ren et al., 2001). The concentration of NO2 amounted to about 50 Ag m3 in May 2001 and 70 Ag m3 in November 2002 (data from At night, the role of NO3 radicals cannot be ignored (NO2 + O3 YNO3 +O2), especially in a polluted atmosphere. The great oxidation capacity of the atmosphere in the PRD will cause fast oxidising DMS, i.e. the lifetime of DMS in the atmosphere will be short. In a remote oceanic area, average OH concentration was in order of 1105 cm3 and lifetime of DMS was as long as about one week (Berresheim et al., 1998). In a polluted atmosphere OH concentration could rise to about 1107 cm3 and the lifetime of DMS was only several hours to one day (Putaud et al., 1999). 4.4. Concentrations of nss sulfate, nitrate and other ions in aerosols 4.4.1. Mass concentration During the November 2002 cruise, three sets of aerosol samples were collected in the PRE, northern South China Sea, and St. S2, respectively. The mass concentration as a function of the size distribution of particles is shown in Fig. 4. Obviously, the mass concentration in the PRE was the highest for any size interval; St. S2 was the second highest, and the South China Sea was the lowest. Accumulating the last seven stages of MOUDI, whose cut-off points were smaller than 10 Am, PM10 concentrations were calculated as 103.7, 43.4, 23.5 Ag m3 in the three areas, respectively. This variation from the inner estuary to the open ocean can be explained by the distance from the continent. The anthropogenic influence in the PRE is much larger than in the South China Sea. Mass size distribution of ambient


Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

Fig. 3. Three-day backward trajectory analysis: (a) Stations E18, E14 and E10 on 4 November 2002, (b) Stations S3, S4 and S6 in the shelf area on 12 November 2002, (c) Stations O3, O3 and O4 in the open sea on 14 November 2002.

particles clearly showed a bimodal pattern with one peak in 0.56–1.0 Am (Dp=0.78 Am) and another in 3.2–5.6 Am (Dp=4.4 Am). The peaks in the PRE

were sharp and high. The ions in these aerosol samples also showed similar variations from the estuary to the open sea.

Fig. 4. Mass size distribution of particles during the November 2002 cruise.

Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

4.4.2. Nss-sulfate The concentration of nss-sulfate was calculated by subtracting sea-salt sulfate (ss-SO42) from the total sulfate. Sea-salt sulfate was estimated by the following formula, in which the coefficient of 0.252 is the typical sulfate-to-sodium mass ratio in seawater (Millero and Sohn, 1992). 2 þ Nss  SO2 4 ¼ SO4  0:252  Na


Fig. 5a presents the size distribution of nss-sulfate concentration in aerosols. Overall the concentrations of nss-sulfate over the PRE were much higher than those over the northern SCS. The sum of the 8 stages was considered to be PM18, in which the nss-SO42 of the PRE, St. S2 and the SCS were 12.1, 6.3 and 4.5 Ag m3, respectively. The concentration level in the PRE was comparable with that of coastal cities in China,


such as 12F10 Ag m3 in Qingdao and 10F4.3 Ag m3 at Xiamen (Gao et al., 1996). The nss-sulfate concentration in the northern SCS was still much higher than in remote oceanic regions. Over the Pacific, mean values of nss-sulfate were 0.49 Ag m3 in Oahu (Savoie and Prospero, 1989) and 0.37 Ag m3 in American Samoa (Savoie et al., 1994). The much higher concentrations of nss-sulfate in the PRE and SCS strongly suggested input of nss-sulfate from non-biogenic sources. The nss-sulfate concentration (4.45 Ag m3) in the northern SCS was comparable to values of 2.61–5.16 Ag m3 found in Sapporo, Japan (Ohta and Okita, 1990) and 3.70 Ag m3 in Cheju Island, Korea (Kim et al., 2000). Gao et al. (1996) measured a mean concentration of nss-sulfate of 4.0F1.3 Ag m3 during the cruise in the East Sea, corresponding very closely to the value of the SCS found in the present study. All these areas lie on the

2 2 Fig. 5. Size distribution of (a) nss-SO2 4 , and (b) nss-SO4 /SO4 .


Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

western rim of the Pacific Ocean and show the same concentration levels of nss-sulfate. The peak of nsssulfate concentration lay at 0.56–1.0 Am, and the peak value in the PRE was more than two times larger than in the SCS. Fig. 5a shows that more than 75% of the mass of nss-SO42concentrated in particles b1.0 Am,

denoting the bulk of these compounds, are secondary aerosols formed in gas-particle conversion. This is consistent with the results from other regions. Putaud et al. (1999) reported that over 70% of nssSO42concentrated in particles smaller than 0.6 Am. At Cheeka Peak, Washington, the maximum nss-


+ + Fig. 6. Mass concentrations/particle-size distribution patterns: (a) NO 3 , (b) Na (c) HH4 .

Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

SO42 occurred in particles between 0.37 and 0.75 Am in diameter (Quinn et al., 1993). This is also clear in Fig. 5b: the ratios of nss-SO42/SO42 in particles b1.8 Am were very close to 1; with the increase in particle diameters, nss-SO42/SO42 decreased. The percentage of sea-salt sulfate evidently increased from the PRE to the SCS. Nitrate, which results mainly from oxidation of NOx, is considered a tracer of anthropogenic pollution (Savoie et al., 1992). Fig. 6a shows that nitrate concentration in the PRE was much higher than in the SCS. The sums of the 8 stages in the PRE, St. S2 and the SCS were 4.88, 1.60, and 2.21 Ag m3, respectively. In addition, the nitrate peak in b1.0 Am particles disappeared in aerosols of St. S2 and the SCS. As mentioned, in the SCS nitrate was mainly formed through NO2 and HNO3 reaction with sea-salt particles. The influence of anthropogenic sources in the northern SCS was much less than that in the PRE. 4.4.3. Nitrate The nitrate size distribution in the PRE was clearly bimodal (Fig. 6a), with 38% of its mass present on b1.0 Am particles, and 42% on particles with diameters larger than 3.2 Am. One peak in nitrate concentration is associated with particles between 3.2 and 5.6 Am in diameter, similar to the sea-salt maximum (Fig. 6b). The second peak was on particles of 0.56–1.0 Am, similar to that for nss-SO42. This bimodal distribution indicates two possible sources of NO3. One was the dissolution of NO2 and HNO3 in the alkaline sea-salt droplets, usually found in coarse particles. The other source was through HNO3 formed from NOx condensing on b1.0 Am anthropogenic particles (Putaud et al., 1999). As suggested by Gao et al. (1996), the b1.0 Am aerosol nitrate may be formed through the gas phase reactions involving HNO3 from fossil fuel burning and possibly NH3 from agricultural activities, with NO3 and NH4+ as end products. High concentrations of NH4+ associated with b1.0 Am nitrate were observed (Fig. 6c).

5. Conclusions The spatial and temporal distributions of seawater DMS, atmospheric DMS, sea-to-air flux of DMS and major components in aerosols were measured along


the PRE and the adjacent northern South China Sea during three cruises. The main conclusions are summarised as follows: !



Surface water content of DMS revealed spatial and seasonal variations with the highest values at the mouth of the estuary. Larger variation of DMS in seawater from the inner estuary to outer waters was observed in the wet season than in the dry season. The seawater DMS concentrations in the estuary and shelf sections were generally higher than those in the open sea section, while the values of DMS/Chl-a in the estuary were less than in the shelf and open sea sections, which indicated the difference in phytoplankton speciation from the estuary to open sea. Surface water concentration and the transfer velocity are both main parameters controlling DMS fluxes to the atmosphere. Atmospheric DMS concentrations are mainly controlled by DMS flux and the oxidation capacity of the atmosphere. High concentration levels of DMS in the water and relatively low DMS concentrations in the atmospherie indicated fast DMS removal from the marine boundary layer. Much higher concentrations of nss-sulfate and nitrate in the atmosphere were observed during the 2002 autumn cruise. The concentrations of aerosol and its components decreased from the inner estuary to outer waters. The high concentrations of O3 and NOx in the atmosphere of the PRD could result in a great potential for OH and NO3 radical formation. This made fast atmospheric DMS oxidation possible. With the decrease of SO2 emission from anthropogenic sources in the PRE the DMS contribution to SO42 will increase. On the basis of measured MSA/nss-SO42 the biogenic sulfur contribution can be estimated.

Acknowledgements We would like to thank Zhai Weidong for supplying the salinity, temperature and other meteorological data. We are grateful to the Ocean Carbon group of Xiamen University for the help during sampling. We are indebted to China National Nature Science Foundation (# 20177002, # 20131160731) for


Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145

financial support. The sampling cruise was supported by CNNSF through grants # 49825111, #40176025 and #49976021. We thank Prof. Sjaak Slanina (The Netherlands Energy Research Foundation) for his helpful comments and two anonymous referees for their efforts and useful comments on the ms.

References Andreae, M.O., Barnard, W.R., 1983. Determination of trace quantities of dimethylsulfide in aqueous solutions. Anal. Chem. 55, 608 – 612. Andreae, M.O., Crutzen, P.J., 1997. Atmospheric aerosols: biogeochemical sources and role in atmospheric chemistry. Science 276, 1052 – 1058. Arimoto, R., Duce, R.A., Savoie, D.L., Prospero, J.M., Talbot, R., Cullen, J.D., Tomza, U., Lewis, N.F., Ray, B.J., 1996. Relationships among aerosol constituents from Asia and the North Pacific during PEM-West A. J. Geophys. Res. 101 (D1), 2011 – 2023. Bates, T.S., Chlarson, R.J., Gammon, R.H., 1987. Evidence for the climatic role of marine biogenic sulphur. Nature 329, 319 – 321. Berresheim, H., Eisele, F.L., 1998. Sulfur chemistry in the Antarctic tropospheric environment: an overview of project SCATE. J. Geophys. Res. 103 (D1), 1619 – 1627. Berresheim, H., Andreae, M.O., Iverson, R.L., Li, S.M., 1991. Seasonal variations of dimethylsulfide emissions and atmospheric sulfur and nitrogen species over the western North Atlantic Ocean. Tellus 43B, 353 – 372. Berresheim, H., Huey, J.W., Thorn, R.P., 1998. Measurements of dimethyl sulfide, dimethyl sulfoxide, dimethyl sulfone, and aerosol ions at Palmer Station, Antarctica. J. Geophys. Res. 103 (D1), 1629 – 1637. Bqrgermeister, S., Zimmermann, R.L., Georgii, H.W., Bingemer, H.G., Kirst, G.O., Jansen, M., Ernst, W., 1990. On the biogenic origin of dimethylsulfide: Relation between chlorophyll, ATP, organismic DMSP, phytoplankton species, and DMS distribution in Atlantic surface water and atmosphere. J. Geophys. Res. 95 (D12), 20607 – 20615. Cerqueira, M.A., Pio, C.A., 1999. Production and release of dimethylsulphide from an estuary in Portugal. Atmos. Environ. 33, 3355 – 3366. Charlson, R.J., Rodhe, H., 1982. Factors controlling the acidity of natural rainwater. Nature 295, 683 – 685. Charlson, R.J., Lovelock, J.E., Andreae, M.O., Warren, S.G., 1987. Sulfur phytoplankton, atmospheric sulfur, cloud albedo and climate. Nature 326, 655 – 661. Davison, B.M., Allen, A.G., 1994. A method for sampling for dimethyl sulfide in polluted and remote marine atmospheres. Atmos. Environ. 28, 1721 – 1729. Gao, Y., Arimoto, R., Duce, R.A., Chen, L.Q., Zhou, M.Y., Gu, D.Y., 1996. Atmospheric non-sea-salt sulfate, nitrate and methanesulfonate over the China Sea. J. Geophys. Res. 101 (D7), 12601 – 12611.

Han, W.Y., 1998. Chemical Oceanography in South China Sea. Ocean Press, Beijing. Hu, M., Tang, X., Li, J.L., 1995. Measurement of dimethyl sulfide in the Bo Sea and Gulf of Jiaozhou. Pure Appl. Chem. 67, 1481 – 1485. Hu, M., Tang, X., Li, J.L., Yu, Z.J., 1997. Dimethylsulfide in sea water in the gulf of Jiaozhou. Acta Scient. Circum. 17, 110 – 115. Hu, D.X., Han, W.Q., Zhang, S., 2001. Sea and Land Interactions in Yangtze and Pearl River Estuaries. Ocean Press, Beijing. 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. Chin. Mar. Environ. Sci. 16 (3), 1 – 7. Iverson, R.L., Nearhoof, F.L., Andreae, M.O., 1989. Production of dimethylsulfonium propionate and dimethylsulphide by phytoplankton in estuarine and coastal water. Limnol. Oceangr. 34, 53 – 67. Jensen, N.R., Hjorth, J., Lohse, C., Skov, H., Restelli, G., 1991. Products and mechanism of the reaction between NO3 and dimethylsulfide in air. Atmos. Environ. 25A, 1897 – 1904. Kim, K.H., Lee, G.W., Kim, Y.P., 2000. Dimethylsulfide and its oxidation products in coastal atmospheres of Cheju Island. Environ. Poll. 110, 147 – 155. Kouvarakis, G., Mihalopoulos, N., 2002. Seasonal variation of dimethylsulfide in the gas phase and of methanesulfonate and non-sea-salt sulfate in the aerosols phase in the Eastern Mediterranean atmosphere. Atmos. Environ. 36, 929 – 938. Li, X.S., Han, Z.G., Zheng, M., Parungo, E., Zhe, L., 1996. Measurements of DMS and other sulfur compounds in Western Pacific Ocean. Acta Oceanol. Sinica 18 (3), 27 – 35. Liss, P.S., Merlivat, L., 1986. The ocean as a source of atmospheric sulphur compounds. The Role of Air-Sea Exchange in Geochemical Cycling. D. Reidel, Dordrecht, pp. 113 – 127. Liss, P.S., Slater, P.G., 1974. Flux of gases across the air-sea interface. Nature 247, 181 – 184. Millero, F.J., Sohn, M.L., 1992. Chemical Oceanography. CRC Press, Boca Raton. Ohta, S., Okita, T., 1990. A chemical characterization of atmospheric aerosol in Sapporo. Atmos. Environ. 24A, 815 – 822. 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. Putaud, J.P., Davison, B.M., Watts, S.F., Mihalopoulos, N., Nguyen, B.C., Hewitt, C.N., 1999. Dimethylsulfide and its oxidation products in two sites in Brittany (France). Atmos. Environ. 33, 647 – 659. Quinn, P.K., Covert, D.S., Bates, T.S., Kapustin, V.N., Ramseybell, D.C., McInnes, L.M., 1993. Dimethylsulfide/ cloud condensation nuclei/climate system: Relevant sizeresolved measurements of the chemical and physical properties of atmospheric aerosol particles. J. Geophy. Res. 98, 10411 – 10427. Ren, X.R., Shao, K.S., Miao, G.F., Tang, X.Y., 2001. Determination of hydroxyl radical concentration in atmosphere. China Environ. Sci. 21 (2), 115 – 118.

Q. Ma et al. / Journal of Sea Research 53 (2005) 131–145 Saltzman, E.S., King, D.B., Holmen, K., Leck, C., 1993. Experimental determination of the diffusion coefficient of dimethylsulfide in water. J. Geophys. Res. 98 (C9), 16481 – 16486. Savoie, D.L., Prospero, J.M., 1989. Comparison of oceanic and continental sources of non-sea-salt sulfate over the Pacific Ocean. Nature 339, 685 – 687. Savoie, D.L., Prospero, J.M., Larsen, I.M., Saltzman, E.S., 1992. Nitrogen and sulfur species in aerosols at Mawson Antarctica and their relationship to natural radio nuclides. J. Atmos. Chem. 14, 181 – 204. Savoie, D.L., Prospero, J.M., Arimoto, R., Duce, R.A., 1994. Nonsea-salt sulfate and methanesulfonate in the North Pacific Ocean. J. Geophys. Res. 99, 3587. Sciare, J., Baboukas, E., Kanakidou, M., Krischke, U., Belviso, S., Bardouki, H., Mihalopoulos, N., 2000. Spatial and temporal variability of atmospheric sulfur-containing gases and particles during the Albatross campaign. J. Geophys. Res. 105 (D11), 14433 – 14448. Sciare, J., Baboukas, E., Mihalopoulos, N., 2001. Short-term variability of atmospheric DMS and its oxidation products at Amsterdam Island during summer time. J. Atmos. Chem. 39, 281 – 302. Sciare, J., Mihalopoulos, N., Nguyen, B.C., 2002. Spatial and temporal variability of dissolved sulfur compounds in European estuaries. Biogeochemistry 59, 121 – 141. Shenoy, D.M., Patil, J.S., 2003. Temporal variations in dimethylsulphoniopropionate and dimethylsulphide in the Zuari estuary, Geo (India). Mar. Environ. Res. 56, 381 – 402. Simo, R., Grimalt, J.O., Albaiges, J., 1997. Dissolved dimethylsulphide, dimethyl-sulfoniopropionate and dimethylsulfox-


ide in western Mediterranean waters. Deep-Sea Res. II 44, 929 – 950. Slezak, D., Herndl, G.J., 2003. Effects of ultraviolet and visible radiation on the cellular concentrations of dimethylsulfoniopropionate (DMSP) in Emiliania huxleyi (strain L). Mar. Ecol. Prog. Ser. 246, 61 – 71. Turner, S.M., Malin, G., Nightinggale, P.D., Liss, P.S., 1996. Seasonal variation of dimethyl sulfide in the North Sea and an assessment of fluxes to the atmosphere. Mar. Chem. 54, 245 – 262. Uher, G., Schebeske, G., Barlow, R.G., Cummings, D.G., Mantoura, R.F.C., Rapsomanikis, S.R., Andreae, M.O., 2000. Distribution and air-sea gas exchange of dimethylsulfide at the European western continental margin. Mar. Chem. 69, 277 – 300. Uzuka, N., Watanabe, S., Tsunogai, S., 1996. Dimethylsulfide in coastal zone of the East China Sea. J. Oceanogr. 52, 313 – 321. Yang, G.P., 2000. Spatial distributions of dimethylsulfide in the South China Sea. Deep-Sea Res I 47, 177 – 192. Yang, G.P., Zhang, J.W., Li, L., Qi, J.L., 2000. Dimethylsulfide in the surface water of the East China Sea. Cont. Shelf Res. 20, 69 – 82. 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. Est. Coast. Shelf Sci. 49, 385 – 400. Zhao, H.T., 1990. Evolvement of the Pearl River Estuary China. Ocean Press, Beijing (in Chinese).