LIU Haoran, XIE Zhouqing,*,YE Peipei, KANG Hui & XU Siqi

1 Institute of Polar Environment, School of Earth and Space Sciences, University of Science and Technology of China,Hefei 230026, China;

2 School of Life Sciences, University of Science and Technology of China, Hefei 230026, China;

3 Shenzhen Futian Environmental Monitoring Center, Shenzhen 518048, China

1 Introduction

Oceans cover 70% of the earth’s surface area and have considerable effects on the earth’s climate system. The effects of sulfur on the biogeochemical cycle in the atmosphere have certainly influenced climate change[1]. In addition to atmospheric sulfur, airborne sulfur at the ocean surface mainly comes from dimethyl sulfide (DMS) produced by microbes at the sea surface, such as phytoplankton[2-4]. It has been estimated that half of the total biological sulfur comes from DMS produced by marine phytoplankton[5]. When DMS is released into the air from the ocean/atmosphere boundary, it is oxidized by OH radicals, generating nonseasalt sulfate (nss-SO42-) and methanesulphonic acid (MSA)[6].Oceanic sulfur emissions containing MSA acting as cloud condensation nuclei (CCN) are believed to play an important role in the earth’s radiation balance[7-8]. This may be caused by a negative feedback for climatic perturbations through the effects of MSA on cloud cover and the earth’s albedo[9]. MSA is considered to be a stable oxidized form of DMS in the marine atmosphere. Measures of MSA concentrations have therefore been frequently used to estimate the productivity of marine organisms[10-11], and serve as a useful tracer to distinguish marine biogenic sources of atmospheric sulfur from other sources[12-13].

The Arctic and Antarctic are the world’s source of cold currents in both the atmosphere and oceans, and play a significant role in global climate change. General atmospheric circulation may transmit anthropogenic pollutants to remote oceans; these could form a thick layer of smog over the Arctic every winter and spring[14]. Therefore, climatic research in the Arctic must be an important component in the study of global climate change. In the Arctic, the primary productivity of phytoplankton plays an important role in DMS emissions[15].Moreover, phytoplankton is a significant source of iodohydrocarbon[16-18]. We will therefore discuss the potential relationship between iodine and DMS.

In the past several decades, many researchers have attempted to analyze the characteristics of MSA in different oceanic regions, including the northwestern Pacific in different oceanic regions. Saltzman et al. found high concentrations of MSA corresponding to high primary productivity in the vicinity of Shemya of the Aleutian Islands in the northwest Pacific Ocean[19]. Li et al. reported on seasonal changes in the atmospheric MSA concentration along the west coast of Canada, and also mentioned that the concentration varied depending on seasonal changes with higher values in August[20]. Gao et al. found that most secondary MSA aerosols were created from the anthropogenic release of SO2in the China Sea region in the northwestern Paci fi c[21]. Mihalopoulos et al. argued that a high level of air pollution may result in the production of greater amounts of MSA. Based on the above analyses, we can conclude that many factors can influence the atmospheric concentration of MSA, such as phytoplankton,temperature, solar radiation, and air pollutants[22].

In summary, many types of research related to MSA have been conducted in the Arctic, in both the open ocean and in coastal regions. However, reports related to the spatial and temporal distributions and the interannual variations of MSA in the Arctic are lacking. To fill these gaps, we conducted a study using R/VXUE LONGicebreaker as a monitoring platform, and collected and analyzed samples from the 3rd Chinese National Arctic Research Expedition(CHINARE2008). The new data were compared with those from the 1st and 5th cruises (CHINARE1999 and CHINARE2012, respectively). Furthermore, we compared the spatial and temporal distributions, and the interannual variations of MSA in the Arctic, northwest Pacific Ocean, and offshore China. The overall goal of this study was to gain a thorough understanding of the way in which biogenic sulfur aerosol concentrations respond to global climate change.

2 Experimental methods

2.1 Two cruises

In this study, samples were collected from two different routes. The first and major cruise of CHINARE2008,conducted from July to September 2008 by R/VXUE LONGicebreaker, sailed from Shanghai, China, to the Arctic Ocean.The ship returned to Shanghai through the western and northern Pacific Ocean and the Bering and Chukchi seas(31.1°N-85.18°N, 122.48°W-146.18°W; Figure 1).

The second cruise was the 1st China Offshore Research Expedition (CORE2006) organized by the Institute of Oceanology of the Chinese Academy of Sciences. The route was divided into three segments: (1) Qingdao to Zhoushan,(2) Zhoushan to Xiamen, and (3) Xiamen to Sanya. The trip began in September 2006 in Qingdao using Research VesselScience III, and reached Xiamen two weeks later. We collected samples from the first two segments in this study(24.49°N-36.07°N, 118.98°E-123.55°W).

In addition, this present study also employed and compared data from 1999 and 2012 CHINAREs[23-24].However, the routes of these expeditions were different. For example, CHINARE2012 started in the Chukchi Sea on 18 July and returned on 8 September by crossing the central Arctic Ocean.

2.2 Sampling and experimental methods

We analyzed a total of 22 suspended particle (TSP) samples and four field control samples, which were collected during CHINARE2008. A high volume air sampler (Tianhong Instruments, Wuhan, China) was placed on the uppermost deck of the vessel. Wind direction was monitored to ensure the samples were not influenced by the vessel’s own emissions. The TSP samples were collected at a flow rate of～1.0 m3.min-1by glass micro fiber filters (Whatman epm2000,Maidstone, UK, 20×25 cm2) for 24 h. The filtered samples were removed from the sampler at the end of the sample period and placed in individual clean plastic bags, which were then sealed and stored at 4°C prior to laboratory analysis.

Eight TSP samples had been collected during CORE2006; four filter samples in the daytime, and four at night. To meet the demands of marine operations, the ship stopped every 3-6 h during the day. When the ship was stationary, sampling also ceased to avoid possible contamination. After the ship-related operations were completed and sailing was again underway, sampling was resumed. Samples were acquired continuously at night.

A 2×2 cm2section was taken from each filter and analyzed. Every sample was sonicated twice for 40 min each time with 20 mL ultra-pure water (≥18 MΩ). Samples were placed in an ice-water bath to control the water temperature.Samples were passed through a clean 0.45 μm filter to extract particles from the samples. Then, a Dionex DX-2100 ion chromatograph (IC) system (Sunnyvale, CA, USA) with an AG11-HC guard column and an AS11-HC analytical column was used to analyze MSA from the samples. Four field filter blanks were analyzed as controls; the MSA concentrations were subtracted from the value of the fi lter blanks to calibrate the measurements.

3 Results and discussion

Figure 2 shows the cruise paths of CHINARE2008 and CORE2006, including the locations where samples had been collected to observe aerosol MSA concentrations.Detailed information for each sample is listed in Table A1.During CHINARE2008, the MSA concentrations varied considerably from non-detectable to 319 ng?m-3(mean,66±84 ng?m-3; median, 35 ng?m-3). This finding was similar to that of Xie et al. who reported an average MSA concentration of 71 ± 60 ng?m-3over the Arctic Ocean during CHINARE-1999 (69°N-75°N, 133°W-165°W)[23]. During CORE2006,the concentration of MSA in the China Sea and vicinity averaged 28 ± 14 ng?m-3(median, 30 ng?m-3).

3.1 Regional characteristics

To improve the analysis of the two routes, the sea area was divided into four parts: (1) CORE2006 in offshore China, (2)CHINARE2008 in the northwestern Pacific Ocean including the Bering Sea, (3) the Chukchi Sea, and (4) the Arctic Ocean including the Beaufort Sea. The last three regions belong to the CHINARE study area.

Figures 2 and 3 show that the Arctic Ocean has the highest MSA concentration among the four regions listed above. In addition, the concentration of MSA in the Arctic Ocean exhibited a large magnitude of variation, peaking in the Beaufort Sea region near the edge of the ice shelf(319 ng?m-3; Table 1). In this region, the ice would melt in summer creating an ice-free sea. The minimum MSA concentration (2 ng?m-3) was observed in an area with pack ice.

The MSA concentration of the northwestern Pacific Ocean was also high and changed dramatically over time(Table 1); the highest two values were observed on the outbound trip and return trip in Japanese waters (Sample A02, 14 July, 208 ng?m-3; Sample A21, 15 September,225 ng?m-3). This shows that the concentration changes in the Sea of Japan over time. MSA was not detected in the middle of the northwestern Pacific at Sample A06 (chlorophyll appears black here) during July. In Figure 3, the black areas indicate that there is no valid Chlorophyll-adata. We assumed that the Chlorophyll-ain these areas is similar to the value at the edge of black areas. Based on this assumption, the concentration of Chlorophyll-ain July was lower than in August in the middle of the northwestern Pacific Ocean. This may be the reason for the relatively low MSA observed in July in this region.

From Table 1 we can see that the MSA concentration in the Chukchi Sea was similar to the area offshore China and that it remained stable. Changes in the distribution of MSA in offshore China may be affected by terrigenous MSA sources.

Table 1 Comparison of MSA concentrations from different data sources

3.1.1 The Arctic Ocean

The MSA concentration of the Arctic Ocean (mean 114 ng?m-3) was highest among the values documented for different regions during CHINARE2008. In addition, Xie et al. reported the average MSA concentration over the Arctic Ocean during CHINARE1999 of 71 ng?m-3, slightly lower than that of the present study[23]. Leck and Persson reported that the average MSA concentration in the Arctic Ocean was 134 ng?m-3in August, and then sharply declined to 5 ng?m-3in October[25]. The MSA data of CHINARE2012,which was much lower than that of CHINARE1999 and CHINARE2008, is discussed below (Section 3.4).

Leck and Persson pointed out that during August the concentration of organic sulfur in ice was always the highest year round at the edge of the Arctic[25]. As mentioned above,marine phytoplankton produce DMS, which comprise over 90% of the sulfur flux in oceans[26]. Chlorophyll-ais well known to be a good proxy for the primary productivity of phytoplankton[5,27]. Chlorophyll-aconcentrations were much higher in August than in July throughout the route in the Arctic and northwestern Pacific Ocean (Figures 3a and 3b).However, almost all of the CHINARE2008 samples from the Arctic were collected in August, while those of other regions were collected in July.

Sample A14 had the highest MSA concentration found in samples from the Arctic. This was also the highest found over the entire route to the Beaufort Sea (319 ng?m-3)in August. Sample A14 was also collected at the highest air temperature among all Arctic samples. In general, air temperature appears to peak in August in the Northern Hemisphere. In the Arctic, high August temperatures could accelerate the melting of sea ice, and greatly enhance the exchange of MSA between the ocean surface and the air[28].In addition, all these factors increased the availability of nutrients that would accelerate the growth of phytoplankton.August is more suitable for the growth of phytoplankton than July or any other time (Figure 3). In summary, August may have conditions that lead to the highest MSA concentrations.

Based on satellite data from the National Snow and Ice Data Center, we found that the average ice coverage in the Arctic Ocean in August 2008 was the second lowest on record (based on data from 1979 to 2008). The warm southern air flow in August 2008 may have been partially responsible for this major loss of sea ice[29]. The data from NASA’s Ice,Cloud, and land Elevation Satellite revealed a clear trend in the decline in the thickness of Arctic Ocean ice in 2008[30].This resulted in a large area of algal growth and additional air-sea exchange[28]. All the factors listed above may therefore contribute to the high concentration of MSA observed in the Arctic during August 2008.

3.1.2 Chukchi Sea

Chukchi Sea was specifically selected from the Arctic Ocean for further discussion to compare it with the data observations from CHINARE2012. The MSA concentration of the Chukchi Sea (25 ± 13 ng?m-3; median, 29 ng?m-3) was lower than that of the other regions in this study (Table 1),but was higher than the MSA concentration of CHINARE-2012 (12±16 ng?m-3). The samples of CHINARE2012 were collected in July whereas the samples of CHINARE 2008 were collected in August (peak primary productivity),partially explaining why concentrations of MSA in samples were higher in CHINARE2008 than in CHINARE2012.

3.1.3 Northwestern Paci fi c Ocean

The average aerosol MSA concentrations in the northwestern Pacific Ocean (54±77 ng?m-3) were the second highest for the entire CHINARE2008 route (Table 1). These values were similar to the average MSA concentrations (97±98 ng?m-3)in northwestern Pacific in the vicinity of Shemya (52°N,174°E), which were collected from May 1981 to April 1982[31]. These were also similar to the MSA concentration(69 ng?m-3) from Ireland’s Mace Head (53.32°N, 9.85°W)in August 1989 to August 1990[10]. For the present study,the aerosol MSA samples were collected in July (lower primary production from phytoplankton than August) and the MSA values were lower than those of CHINARE1999(263 ng?m-3)[23].

The northwestern Pacific Ocean region exhibited two extremely high MSA values (Sample A02, 208 ng?m-3;Sample A21, 225 ng?m-3), both collected during CHINARE 2008,and both from the Sea of Japan. From Figure A1, it can be seen that samples A02 and A21 were clearly affected by air masses originating from areas near Japan, Korea, and China. Pollutants from inland air may have caused the high MSA values in these two samples. Mukai and Yokouchi had observed similar high MSA concentrations from 1983 to 1992[32].

The Chlorophyll-aconcentration in the Bering Sea (25±17 ng?m-3) was higher than that of the Sea of Japan (Figures 2 and 3). However, the MSA concentration was the opposite.Turner et al. pointed out that sometimes the concentration of DMS in seawater would not necessarily be associated with the chlorophyll concentration, since various factors, such as weather conditions (including precipitation), temperature, and algal growth, may have a combined effect[33].

The concentrations of nitrogen oxides, such as nitrates,peak in the inland areas of Japan during summer[34]. This, in turn, affects the efficiency of oxidation of DMS and thereby creates significantly higher concentrations of MSA in the atmosphere[35].

To better analyze the data, we divided the northwestern Pacific Ocean into two parts: the east coast of Asia (based on the Sea of Japan, 33°N-45°N, 126°E-147°E) and the other areas of the northwestern Pacific Ocean (46°N-64°N,148°E-168°W). Then, we analyzed the NO3-concentrations of these regions during CHINARE2008. The aerosol NO3-concentration (2.01 μg?m-3) on the east coast of Asia was far greater than that of other areas (Figure 4). In addition, the mean MSA value (107.4 ng?m-3) was also much higher than that of the northwestern Pacific Ocean (20.3 ng?m-3) and the Chukchi Sea (25.3 ng?m-3). This may have been caused by the long distance transport of pollutants from Japan and other east Asian inland areas[34]. Furthermore, the high nitrogen oxide content of the pollutants may have improved the oxidation efficiency of DMS.

Yuan et al. reported that MSA was detected from most aerosol samples collected in the Beijing area, and in concentrations that were significantly higher than those in coastal Chinese cities and oceanic areas[36]. However,no obvious sea-salt source from the aerosol was observed in Beijing. The release of DMS in the inland areas and dimethyl sulfoxide (DMSO) from industrial wastes may be two precursors to MSA. The CHINARE2008 vessel passed though the Sea of Japan, which is near the inland areas of China. Also, anthropogenic pollution on land may play an important role in the formation of MSA in this area.

3.1.4 Offshore China

The MSA concentration over areas offshore China was 28±14 ng?m-3and the distribution of MSA was relatively stable (Table 1). This mean MSA value was comparable to previous studies related to the MSA content of air in coastal areas worldwide, and our findings were similar to results from these areas. Nakamura et al. reported that the MSA content of air in the East China Sea was 27±20 ng?m-3(26°N-34°N, 122°E-130°E). Gao et al. found that the average MSA concentrations in Qingdao, Xiamen, and on an East China Sea voyage ranged from 29 to 69 ng?m-3. These results are close to the MSA levels in the current study[21,37].

The route of CORE2006 also documented the spatial distribution of MSA concentrations. The East China Sea,offshore China (32 ng?m-3, 25°N-33°N) had higher MSA concentrations than the other part of China’s offshore region(21 ng?m-3, 34°N-36°N). The diverse areas of the main region of China Sea have different levels of marine productivity. The regions near the Yellow and Yangtze rivers, both of which flow into the East China Sea, have higher marine primary productivity. The concentration of atmosphere NOx, which plays an important role in the process of oxidation of DMS into MSA, also influences the MSA concentrations. High concentrations of NOxpollutant gases will cause atmospheric concentrations of MSA to become very high[22,36,38]. The eastern coastal areas of China are industrial with well developed manufacturing infrastructure. Therefore, a high concentration of anthropogenic NOxcan be transmitted to the East China Sea from the inland areas, causing an increase in the atmospheric concentration of MSA.

3.2 Link with iodine

Marine phytoplankton are well known as producers of DMS, which comprises more than 90% of the sulfur flux in oceans[2]. Moreover, phytoplankton are a significant source of various halocarbons (including iodocarbons)[16-18]. In the Arctic, diatoms are the most frequently documented phytoplankton taxa that may produce significant amounts of iodocarbons[39-40]. By coincidence, diatoms may also produce DMS in the Arctic Ocean[41-42].

The data from CHINARE2008 show that Sample A14 had the highest MSA content of all samples from the Arctic Ocean and the entire cruise. Similarly, the highest concentration of TI found during this cruise was also found in Sample A14[43]. Therefore, by combining our data with the TI data from Xu et al.[43], we analyzed the correlation between MSA and TI in the Arctic Ocean (including the Chukchi Sea).

The concentrations of MSA and TI show a strong correlation (R2= 0.82,P＜0.01, Figure 5). When the same environmental factors are considered, DMS and TI may be derived from similar sources in the Arctic marine boundary.

3.3 Comparing previous cruises in the Arctic Ocean

Previous studies have shown that the highest concentration of organic sulfur was observed at the edge of the Arctic Ocean ice in August[44]when warm humid weather corresponds with the high concentration of MSA in the atmosphere[45]. In the Arctic, the highest concentration of atmospheric MSA occurs in summer. In addition, the atmospheric concentration of MSA at the edge of the area between the ocean and sea ice is far higher than that in areas with thick ice. High temperatures accelerate the melting sea ice, which is a significant cause of the high MSA concentrations observed in summer. Ice cover plays an important role in air-sea exchange, and influences the nature of the marine community and climate[46]. Sea ice may limit the DMS air-sea exchange and reduce the content of MSA in the atmosphere[25,44].

The extent of ice coverage in the Arctic Ocean has tended to decrease in recent years, as a result of great changes in the Arctic environment[47]. In addition, the overall downtrend may be continuing throughout the entire year,during both summer and winter.

Compared with CHINARE1999, the concentration of MSA throughout the Arctic segment in CHINARE2008(114±107 ng?m-3) was slightly higher than that in CHINARE 1999 (71±60 ng?m-3). The average ice cover in the Arctic Ocean during August 2008 was much lower than that in 1999[29]. This maybe the most important cause of the higher MSA concentrations observed during CHINARE2008.However, CHINARE2012 documented much lower MSA concentrations (1.2 ng?m-3) than previous research trips in the Arctic, especially when compared to CHINARE2008. The extremely low concentrations may be the result of the specific route. The cruise of CHINARE2012 began in the Chukchi Sea and passed through the entire central Arctic Ocean.Furthermore, a majority of the samples collected during this cruise were from the central Arctic Ocean, which has extensive ice cover.

The distribution of sea ice in the high Arctic (77°N-86°N), measured by helicopter aerial photography from R/VXUE LONGicebreaker, the CHINARE2008 vessel[48], can be divided into three parts: (1) areas with pack ice (83°N-86°N,the pack ice covered more than 70% of the sea), (2) areas with floating sea ice (77°N-83°N), and (3) areas of open sea(＜77°N). The samples collected during CHINARE 2008 were nearly all from areas of open sea, and areas with floating sea ice. Only one sample was collected in an area of pack ice(Sample A17, 2 ng?m-3). The MSA value here was similar to the samples collected from CHINARE2012, which were collected in areas with pack ice. Thus, the CHINARE2012 samples were mostly collected in areas of pack ice, which may be the primary reason for the much lower MSA values observed during the 2012 cruise.

From the above,we can infer that the MSA concentration is negatively correlated with the spatial extent of sea ice, and a large expansion of sea ice provides an important barrier that limits DMS and MSA concentrations. In the event of continuing loss of sea ice in the future, MSA concentrations will be expected to increase.

4 Summary

We investigated MSA concentrations from aerosol samples collected during CHINARE2008 from July to September. We found that the MSA concentrations varied considerably from non-detectable to 319 ng?m-3(mean, 66±84 ng?m-3; median,35 ng?m-3). In coastal regions, such as the Sea of Japan, air pollution (especially nitric oxide) from inland areas could also have strong effects on the concentration of aerosol MSA. In remote areas of the Pacific and Arctic oceans,MSA generated by phytoplankton is the dominant factor influencing the airborne MSA concentration. Moreover, we found a fairly similar variation trend between TI and MSA in the Arctic during CHINARE2008, indicating that iodine and MSA may come from similar sources in Arctic. High MSA concentrations were always observed during August throughout CHINARE2008, similar to many previous reports from the Northern Hemisphere. The combination of increased warmth and nutrients creates more suitable conditions for the growth of phytoplankton in August than at any other time. Finally, when comparing the CHINARE2008 data with the data from CHINARE1999 and CHINARE2012 cruises, we found that sea ice is an extremely significant factor that influences the aerosol concentration of MSA in the Arctic. Global warming will tend to accelerate the melting of sea ice, leading to higher MSA concentrations in the future.

AcknowledgementsThis research was supported by grants from the National Natural Science Foundation of China (Grant nos. 41176170,41025020), and the Chinese Polar Environment Comprehensive Investigation & Assessment Programs (Grant no. CHINARE 2011-2015)

1 Andreae M O, Raemdonck H. Dimethyl sulfide in the surface ocean and the marine atmosphere: a global view. Science, 1983, 221(4612):744-747

2 Ayers G P, Ivey J P, Gillett R W. Coherence between seasonal cycles of dimethyl sulphide, methanesulphonate and sulphate in marine air.Nature, 1991, 349(6308): 404-406

3 Bates T S, Cline J D, Gammon R H, et al. Regional and seasonal variations in the flux of oceanic dimethylsul fide to the atmosphere. J Geophys Res, 1987, 92(C3): 2930-2938

4 Koch D, Jacob D, Tegen I, et al. Tropospheric sulfur simulation and sulfate direct radiative forcing in the Goddard Institute for Space Studies general circulation model. J Geophys Res, 1999, 104(D19):23799-23822

5 Buat-Ménard P. The role of air-sea exchange in geochemical cycling.New York: Springer Science & Business Media, 2012, 185

6 O’Dowd C D, Smith M H, Consterdine I E, et al. Marine aerosol, seasalt, and the marine sulphur cycle: A short review. Atmos Environ,1997, 31(1): 73-80

7 Shaw G E. Bio-controlled thermostasis involving the sulfur cycle.Clim Change, 1983, 5(3): 297-303

8 Clarke A D, Davis D, Kapustin V N, et al. Particle nucleation in the tropical boundary layer and its coupling to marine sulfur sources.Science, 1998, 282(5386): 89-92

9 Charlson R J, Lovelock J E, Andreae M O, et al. Oceanic phytoplankton, atmospheric sulphur, cloud albedo and climate.Nature, 1987, 326(6114): 655-661

10 Savoie D L, Arimoto R, Keene W C, et al. Marine biogenic and anthropogenic contributions to non-sea-salt sulfate in the marine boundary layer over the North Atlantic Ocean. J Geophys Res, 2002,107(D18): 4356

11 Curran M A J, Van Ommen T D, Morgan V I, et al. Ice core evidence for Antarctic sea ice decline since the 1950s. Science, 2003,302(5648): 1203-1206

12 Legrand M, Saigne C F. Methanesulfonic acid in south polar snow layers: A record of strong El Nino? Geophys Res Lett, 1991, 18(2):187-190

13 Davis D, Chen G, Bandy A, et al. Dimethyl sulfide oxidation in the equatorial pacific: comparison of model simulations with field observations for DMS, SO2, H2SO4(g), msa(g), MS and NSS. J Geophys Res, 1999, 104(D5): 5765-5784

14 Sturges W T, Shaw G E. Halogens in aerosols in central Alaska.Atmos Environ Part A, 1993, 27(17-18): 2969-2977

15 Stefels J, Van Boekel W H M. Production of DMS from dissolved DMSP in axenic cultures of the marine phytoplankton speciesPhaeocystissp. Mar Ecol Prog Ser, 1993, 97(1): 11-18

16 Quack B, Wallace D W R. Air-sea flux of bromoform: Controls, rates,and implications. Global Biogeochem Cycles, 2003, 17(1): 1023

17 Salawitch R J. Atmospheric chemistry: biogenic bromine. Nature,2006, 439(7074): 275-277

18 Lovelock J E. Natural halocarbons in the air and in the sea. Nature,1975, 256(5514): 193-194

19 Saltzman E S, Savoie D L, Prospero J M, et al. Methanesulfonic acid and non-sea-salt sulfate in Pacific air: Regional and seasonal variations. J Atmos Chem, 1986, 4(2): 227-240

20 Li S M, Barrie L A. Biogenic sulfur aerosol in the Arctic troposphere:1. Contributions to total sulfate. J Geophys Res, 1993, 98(D11):20613-20622

21 Gao Y, Arimoto R, Duce R A, et al. Atmospheric non-sea-salt sulfate,nitrate and methanesulfonate over the China Sea. J Geophys Res,1996, 101(D7): 12601-12611

22 Mthalopoulos N, Nguyen B C, Boissard C, et al. Field study of dimethylsulfide oxibation in the boundary layer: Variations of dimethylsulfide, methanesulfonic acid, sulfur dioxide, non-sea-salt sulfate and aitken nuclei at a coastal site. J Atmos Chem, 1992, 14(1):459-477

23 Xie Z Q, Sun L G, Jihong C D. Non-sea-salt sulfate in the marine boundary layer and its possible impact on chloride depletion. Acta Oceanol Sin, 2005, 24(6): 162-171

24 Ye P P, Xie Z Q, Yu J, et al. Spatial distribution of methanesulphonic acid in the arctic aerosol collected during the Chinese Arctic Research Expedition. Atmosphere, 2015, 6(5): 699-712, doi: 10.3390/atmos6050699.

25 Leck C, Persson C. Seasonal and short-term variability in dimethyl sulfide, sulfur dioxide and biogenic sulfur and sea salt aerosol particles in the arctic marine boundary layer during summer and autumn. Tellus B, 1996, 48(2): 272-299

26 Jenkins A,Dutrieux P, Jacobs S S, et al. Observations beneath Pine Island Glacier in West Antarctica and implications for its retreat. Nat Geosci, 2010, 3(7): 468-472

27 Barnard W R, Andreae M O, IversonR L. Dimethylsulfide and Phaeocystis poucheti in the southeastern Bering Sea. Cont Shelf Res,1984, 3(2):103-113

28 Arrigo K R, Van Dijken G L. Annual changes in sea-ice, chlorophyll a, and primary production in the Ross Sea, Antarctica. Deep Sea Res Part II: Top Stud Oceanogr, 2004, 51(1-3): 117-138

29 NSIDC. Arctic sea ice news & analysis: Arctic sea ice now secondlo west on record. http://nsidcorg/arcticseaicenews/2008/082608html,2008

30 Kwok R, Rothrock D A. Decline in Arctic sea ice thickness from submarine and ICESat records: 1958-2008. Geophys Res Lett, 2009,36(15):L15501

31 Savoie D L, Prospero J M, Saltzman E S. Non-sea-salt sulfate and nitrate in trade wind aerosols at Barbados: Evidence for long-range transport. J Geophys Res, 1989, 94(D4): 5069-5080

32 Mukai H, Yokouchi Y, Suzuki M. Seasonal variation of methanesulfonic acid in the atmosphere over the Oki Islands in the sea of Japan. Atmos Environ, 1995, 29(14):1637-1648

33 Turner S M, Malin G, Liss P S, et al. The seasonal variation of dimethyl sulfide and dimethylsulfoniopropionate concentrations in nearshore waters. Limnol Oceanogr, 1988, 33(3): 364-375

34 Mukai H, Ambe Y, Shibata K, et al. Long-term variation of chemical composition of atmospheric aerosol on the Oki Islands in the Sea of Japan. Atmos Environ, 1990, 24(6): 1379-1390

35 Jensen N R, Hjorth J, Lohse C, et al. Products and mechanism of the reaction between NO3and dimethylsulphide in air. Atmos Environ,1991, 25(9): 1897-1904

36 Yuan H, Wang Y, Zhuang G S. The MSA concentration of aerosols in Beijing. Chin Sci Bull, 2004, 49(8): 744-749 (in Chinese)

37 Yin F D, Grosjean D, Seinfeld J H. Photooxidation of dimethyl sul fi de and dimethyl disul fi de. I: Mechanism development. J Atmos Chem,1990, 11(4): 309-364

38 Zhang X Y, Zhuang G S, Guo J H, et al. Characterization of aerosol over the Northern South China Sea during two cruises in 2003. Atmos Environ, 2007, 41(36): 7821-7836

39 Sturges W T, Cota G F, Buckley P T. Bromoform emission from Arctic ice algae. Nature, 1992, 358(6388): 660-662

40 Tokarczyk R, Moore R M. Production of volatile organohalogens by phytoplankton cultures. Geophys Res Lett, 1994, 21(4): 285-288

41 Belviso S, Moulin C, Bopp L, et al. Assessment of a global climatology of oceanic dimethylsul fide (DMS) concentrations based on SeaWiFS imagery (1998-2001). Can J Fisheries Aquat Sci, 2004,61(5): 804-816

42 Vogt M, Liss P S. Dimethylsulfide and climate // Le Quéré C, Saltzman E S. Surface ocean-lower atmosphere processes.Geophysical Monograph Series, 2009, 198: 197-232

43 Xu S Q, Xie Z Q, Li B, et al. Iodine speciation in marine aerosols along a 15 000-km round-trip cruise path from Shanghai, China, to the Arctic Ocean. Environ Chem, 2010, 7(5): 406-412

44 Leck C, Persson C. The central Arctic Ocean as a source of dimethyl sulfide Seasonal variability in relation to biological activity. Tellus B,1996, 48(2): 156-177

45 Hara K, Osada K, Hayashi M, et al. Variation of concentrations of sulfate, methanesulfonate and sulfur dioxide at Ny-?lesund in 1995/96 winter. Proc NIPR Symp Polar Meteorol Glaciol, 1997, 11:127-137

46 Clarke D B, Ackley S F. Sea ice structure and biological activity in the Antarctic marginal ice zone. J Geophys Res, 1984, 89(C2): 2087-2095

47 Stroeve J, Frei A, Mc Creight J, et al. Arctic sea-ice variability revisited. Ann Glaciol, 2008, 48(1): 71-81

48 Lu P, Li Z J, Cheng B, et al. Sea ice surface features in Arctic summer 2008: Aerial observations. Remote Sens Environ, 2010, 114(4): 693-699

49 Nakamura T, Matsumoto K, Uematsu M. Chemical characteristics of aerosols transported from Asia to the East China Sea: an evaluation of anthropogenic combined nitrogen deposition in autumn. Atmos Environ, 2005, 39(9): 1749-1758