Advertisement

Environmental Science and Pollution Research

, Volume 25, Issue 30, pp 30659–30670 | Cite as

Impact of marine and continental sources on aerosol characteristics using an on-board SPAMS over southeast sea, China

  • Jinpei YanEmail author
  • Liqi Chen
  • Shuhui Zhao
  • Miming Zhang
  • Qi Lin
  • Lei Li
Research Article
  • 77 Downloads

Abstract

The chemical composition of atmospheric aerosols was characterized using an on-board single particle aerosol mass spectrometer (SPAMS) over the Southeast China Sea. High-time-resolution observation of marine aerosols was carried out to clarify the source of aerosols and the interaction of marine and continental aerosols. Atmospheric aerosols were determined by the interaction of continental and marine sources over coastal area. Aerosols from continental sources flux into sea surfaces through deposition or diffusion, which results in the rapid decrease of continental aerosols. Five main subtypes of carbonaceous particles are identified as C_Al-Si, C_V-Ni, C_S, C_K, and C_secondary to clarify the impact of marine and continental sources on atmospheric aerosols. High fraction of C_Al-Si and C_secondary is present over XA (Xiamen anchorage), accounting for 23.8% and 18.6% of total carbonaceous particles. Contrarily, the relative percentage of C_S increases as the distance from land to sea increases. The influence of continental aerosols declines, while the contribution of marine aerosols increases as the distance from land to sea increases. Air masses in XA, LSA (land to sea area), SLA (sea to land area), and SA (sea area) were all from ocean during the observation period, resulting in low relative fraction of continental aerosols in SLA, SA, and LSA. High-time-resolution measurement is useful to understand aerosol source types and the impact of marine and continental sources on marine atmosphere aerosols.

Keywords

Marine aerosol Chemical compositions Size distribution On-board observation Single particle aerosol mass spectrometer (SPAMS) 

Introduction

Marine aerosols have a significant contribution to atmospheric aerosols, which affect the atmospheric radiative budget by dictating the cloud condensation nuclei (CCN) population. (Woods et al. 2010; Gkikas et al. 2016; Decesari et al. 2011). Large quantities of aerosols were inevitably emitted into the atmosphere with the rapid development of economic and energy consumption (Yao et al. 2009; Ohlstrom et al. 2000; Yang et al. 2010; Wang 2018). Consequently, marine atmosphere is significantly impacted by the continental pollutant emissions, especially in the coastal zone adjacent to urban or industrial areas (Fu et al. 2015; Yan et al. 2015; Masalaite et al. 2017). More and more continental aerosols flux into the upper ocean through air-sea exchange with the increase of aerosol pollution emissions, causing the change of nutrient supply, which affected marine primary productivity and marine ecological environment (Zhang et al. 2013; Murphy et al. 2009).

The Southeastern China Sea (SECS) is a marginal zone adjacent to the southeast China with a subtropical climate under the influence of Asian monsoon. As a transitional area, northeasterly wind is the dominant wind direction in cold seasons. Air masses mainly come from northern China during wintertime, which makes aerosols transport from continent to SECS easily (Yan et al. 2015; Zhang et al. 2012). In this area, complex mixed aerosols are formed by the interaction of marine aerosols and anthropogenic particles (Falkowska and Lewandowska 2004; Chalbot et al. 2013). Studies on marine aerosols have been carried out, such as aerosol chemical characteristics, size distribution, and air-sea deposition fluxes, in recent years. (Kim et al. 2009; Nakamura et al. 2006; Chen et al. 2010). However, offline sampling methods were conducted in previous studies. Sampling time for 24 h or even several days was required to accommodate the detection limits (Zhang et al. 2007). However, atmospheric aerosols change rapidly due to the oxidation or reaction occurred on particle surfaces. Long sampling interval of aerosol is difficult to estimate with the atmospheric process of marine aerosols, which makes it hard to clarify the aging process of aerosols and the interaction between different aerosols using offline sampling methods (Yan et al. 2016). Therefore, high-time-resolution observation is important to investigate the rapid change of aerosol properties in marine atmosphere.

Online aerosol mass spectrometry has been widely used in atmospheric aerosol observation, since it is available to provide size and chemical information of aerosols (Sullivan et al. 2007; Spencer et al. 2007). Recently, two types of aerosol mass spectrometers (such as aerosol time-of-flight mass spectrometer (ATOFMS) and quadrupole aerosol mass spectrometer (AMS)) were used in marine aerosol observation to characterize chemical compositions, sources, and size distribution of marine aerosols (Fu et al. 2015; Phinney et al. 2006; Hoomaert et al. 2003; Gaston et al. 2011; O’Dowd et al. 2004). Phinney used an on-board AMS to investigate characteristics of marine aerosols over the North Atlantic Ocean. AMS only detects the volatile and semi-volatile components (such as SO42−, NO3, NH4+, Cl, and organics) and provides the mass concentration of these species (Jimenez et al. 2003). It is useful to investigate the formation of secondary aerosols and classify the organic species. However, it is hard to clarify the aerosol sources due to the undetectable refractory species by AMS. While ATOFMS can provide particle size distribution and chemical compositions (such as heavy metals, elemental carbon, organic carbon, SO42−, NO3, NH4+, K+, Na+, Mg+) (Gaston et al. 2011), which makes it possible to investigate mixed state and source of aerosols. But ATOFMS just provide the particle number count without the mass concentration. Recently, a newly developed single particle aerosol mass spectrometer (SPAMS) was proverbially used in urban atmospheric aerosol monitoring to investigate their sources and chemical compositions (Murphy et al. 2006; Li et al. 2014; Ma et al. 2016; Bi et al. 2011). However, rare studies of marine aerosols using SPAMS were present, especially with an on-board SPAMS observation. Single particle aerosol mass spectrometer provides particle size and chemical compositions of single particle. In this study, an on-board SPAMS was deployed on “Xiang Yang Hong 03” vessel to investigate aerosol chemical components and particle sources in different sea areas. This study aims to understand the rapid variation of atmospheric aerosol chemical compositions and the influence of continental and marine sources on marine atmospheric aerosols over the SECS. The backward trajectory results were also present to evaluate the pollutant sources during the observation period. It provides an insight into the interaction between marine aerosols and continental aerosols over the offshore marine atmosphere.

Experimental methods and data

Sampling site

The observation was directed by “Xiang Yang Hong 03” of Third Institute of Oceanography. On-board SPAMS monitoring system was deployed in the vessel, and the cruise covered the Southeast China Sea (N 20o54′26″-24o24′35″, E 117o15′55″-118o30′36″). The vessel tracks are given in Fig. S1. The SPAMS instrument was equipped to a stabilized platform and held on a mobile laboratory located on the front deck of the vessel. The stabilized platform can adjust the swing range when the vessel jolted to ensure the safety and steady run of the instrument. Note that there is a potential influence of the self-contamination from the vessel on the measurement. To minimize the impact of the sampling vessel emission, the sampling inlet connecting to the SPAMS was fixed in the foremast, which was about 20 m above the sea level (seen in Fig. S2). Pollution emissions from the vessel are located at the downwind of the sampling inlet, when the vessel is running. A continuous observation was carried out from 11 to 24 June 2016. In this study, a total of 347,590 particles were detected by SPAMS.

On-board SPAMS and data analysis

The method for aerosol detection and operation procedure for the on-board SPAMS is similar with other single particle mass spectrometer, which has been described in detail by Li et al. (2014). Sampling gases are dried with a Nafion dryer before entering into the SPAMS. A PM2.5 collector was deployed to remove particles larger than 2.5 μm. Fine particles were drawn into the vacuum system through a critical orifice, which were accelerated and focused to form a particle beam. Particle with special velocity then passed through two continuous diode Nd:YAG lasers (532 nm). The aerodynamic diameter of single particle was calculated based on the particle velocity. Individual particle was then ionized by an Nd:YAG laser (266 nm) to produce positive and negative fragment ions. The power density of ionization laser was kept at 1.55 × 108 w.cm−2. Fragment ions were detected by a bipolar time-of-flight mass spectrometer. Polystyrene latex spheres (PSL Nanosphere Size Standards, Duke Scientific Corp., Palo Alto) with different diameter of 0.2, 0.3, 0.5, 0.72, 1.0, and 2.0 μm were used to calibrate the particle size for SPAMS. PbNO3 particles with diameter of 0.35 μm generated by an aerosol generation and monitoring system (AGM-1500, MSP Corporation, USA) were used to clarify the mass spectrum.

In this study, the integration time employed for the analysis was 30 min. The particle size and mass spectra information were analyzed using YAADA software toolkit (http://www.yaada.org/) (Allen 2005). An adaptive resonance theory using neural network algorithm (ART-2a) was applied to cluster individual particles into separate groups based on the presence and intensity of ion peaks in single particle mass spectrum (Song et al. 1999), with a vigilance factor of 0.65, learning rate of 0.05, and a maximum of 20 iterations.

Identification of carbonaceous types

Carbonaceous material includes organic carbon (OC) and elemental carbon (EC). Fragment ions of CxHyOz+, CxHyOz, CxHy+, and CxHy were used to classify the organic carbon, and Cn and Cn+ were used to classify the element carbon. Carbonaceous particles were extracted from the total particles firstly. The rules used to classify different carbonaceous subtypes are as follows: (1) particles with ions of Al+ (m/z 27), Si+ (m/z 28), and SiO2 (m/z − 60) were classified as C_Al-Si particles; (2) particles containing V+ (m/z 51), VO+ (m/z 67), and Ni+ (m/z 58) ions in the positive spectrum, element carbon chain ions, and fewer organic ions in the positive and negative spectrum were classified as C_V-Ni particles; (3) particles with ions of MSA (m/z − 95), HSO4 (m/z − 97) in the negative spectrum, and Mg+ (m/z 24) in the positive spectrum were classified as C_S particles; (4) particles with clear ions of K+ (m/z 39) and organic species were classified as C_K particles; and (5) particles with abundant secondary ions (m/z − 80 [SO3], − 97 [HSO4], − 46 [NO2], − 62 [NO3], 18 [NH4]+) were classified as C_secondary particles. Carbonaceous particles other than the above types accounted only about 3% of total carbonaceous. Hence, these particles were not discussed in this study.

Meteorological parameter and back trajectory (BT) analysis

Wind speed and direction were monitored by the weather station equipped in the vessel during the cruise. Air mass backward trajectory analysis was calculated by HYSPLIT_4 model to clarify the effect of source regions on marine atmospheric aerosols. The meteorological data used for the computation of the trajectories was obtained from the GDAS archive maintained by the Aerological Research Laboratory (ARL) (available at http://ready.arl.noaa.gov/gdasl.php).

Results and discussion

Spatial distribution of aerosol chemical characteristics

High temporal and spatial resolution data of chemical compositions provide a useful clue to track the variation process of aerosols from land to sea. Figure 1 shows the spatial distribution of six major aerosol species (SO42−, NO3, NH4+, Na+, Element carbon (EC), and organic carbon (OC)) in SECS. The covered observation area can be classified as four main regions, Xiamen anchorage (XA), land to sea area (LSA), sea area (SA), and sea to land area (SLA). Sulfate, nitrate, sodium, and carbonaceous (including OC and EC) were dominant aerosol species in XA, accounting for 25%, 21%, 19%, and 31%, respectively. Generally, high concentration of SO42−, NO3, and NH4+ produced by the transformation of their precursors of SO2, NO2, and NH3 was concerned with high secondary aerosol pollution (Wang et al. 2006,). EC was derived from incomplete combustion of carbon-containing materials, such as coal combustion and vehicle emissions (Murphy et al. 2009), while OC was from the primary particle emission or secondary reaction (Sitaras and Siskos 2008). Na+ was mainly contributed by sea salt aerosols. Therefore, atmospheric aerosols were impacted by anthropogenic emissions (coal combustion, vehicle, residential, and ship emission, etc.) and marine sources (sea salt particles) over XA. Similar observation results were present in previous studies (Zhao et al. 2011; Yan et al. 2013).
Fig. 1

a Spatial distribution of different aerosol species including OC, EC, NH4+, SO42−, NO3, and Na+ over the Southeastern China Sea. b Average percentage of the six major aerosol species in XA, LSA, SA, and SLA

Compared with aerosol chemical compositions over XA, a significant change of aerosol species fraction was present over LSA. The species of NO3 and SO42− decreased dramatically, accounting for only 2% and 15%. While the mean relative percentage of EC, OC, and Na+ increased to 36%, 16%, and 23%, respectively. Note that NO3 and SO42− were mainly from anthropogenic emission. The decrease of NO3 and SO42− containing particles meant that the impact of continental aerosols declined. As mentioned above, EC was mainly from vehicle or ship emission, high fraction of EC containing particles over LSA was contributed by the ship emission. This can be further demonstrated by the mass spectra of carbonaceous clusters and the spatial distribution of carbonaceous materials (seen in Fig. 4 and Fig. 5). Atmospheric aerosols were governed by marine sources over SA. Sodium, representing the sea salt particles (Udisti et al. 2012), was the only dominated species over SA, accounting for 57% of the total particles. Contrarily, the relative fraction of SO42−, NO3, and EC was extremely low, accounting for 2%, 2%, and 9%, respectively. That means continental sources have few impacts on atmospheric aerosols over SA, since most of continental aerosols deposit during the transport process from land to sea. Generally, SO42− aerosols were from marine sources or anthropogenic sources (Yan et al. 2013). The relative percentage of SO42− was 25%, 15%, and 2% over XA, LSA, and SA, respectively, suggesting that the contribution of SO42− species decreased as the distance from land to sea increased. That means SO42− aerosols were mainly from anthropogenic emission in the observation area. Contrarily, high relative percentage of OC (31%) was present over SA, comparing with the value over LSA (16%) and XA (11%). Organic species was used to investigate the contribution of different emission sources, such as fossil combustion, biomass emission, and marine biogenic source (Yin et al. 2012). Since continental sources have few influence on marine atmospheric aerosols over SA, high relative fraction of OC indicated that OC was mainly derived from the marine biogenic source in this area.

High relative fraction of EC, OC, and Na+ containing particles, accounting for 25%, 37%, and 29% with low relative fraction of NO3 and SO42− (3% and 4%) was present over SLA. The average fraction of NH4+ and SO42− was about 2% and 4% over LSA, which was lower than the value (8% of NH4+ and 15% of SO42−) obtained over LSA. Compared with aerosol chemical compositions over LSA, higher contribution of marine aerosols was present over LSA, though SLA is closer to shore area. Generally, the impact of continental sources on atmospheric aerosols increased when the location was closed to the shore. However, the contribution of continental aerosols over SLA was lower than over LSA in this study. Note that observations over LSA and SLA were not performed simultaneously. Higher wind speed was present over SLA during the observation episode, which contributed to higher fraction of marine aerosols, as the production of marine aerosols from whitecaps was enhanced with the increase of wind speed (Norris et al. 2013).

Variation of particle number from land to sea

Figure 2a shows the spatial distribution of total particle number count measured by the on-board SPAMS over SECS. High particle number count was found over XA, since the atmospheric aerosols were dominated by both continental sources and marine sources. High percentage of SO42−, NO3, and NH4+ species (Fig. 1) over XA implied that high particle number count was determined by anthropogenic pollutants in this area. Aerosols from continental sources fluxed into the sea surface through deposition or diffusion in the transport process from land to sea (Shi et al. 2013), which reduced the continental aerosol number concentration. Meanwhile, the contribution of marine aerosols was enhanced. Hence, the particle number was determined by sea salt particles over the sea area, where the impact of continental aerosols was low. Note that the generation of sea salt aerosols correlated well with the whitecaps and wind speed, indicating that the sea salt particle number did not always increase with the distance from land to sea. As seen in Fig. 2b, the number of Na+ particles was not a monotonic tendency as the linear distance from land to sea increased. The contribution of marine sources was enhanced when the observation location was far from the continent, but the sea salt particle number did not always increase.
Fig. 2

a Spatial distribution of total particle number count detected by SPAMS with diameter of 0.2 to 2.5 μm. b Average particle number of different aerosol species (OC, EC, NH4+, SO42−, NO3, and Na+) varies with the linear distances from Xiamen

To further clarify the variation of continental and marine aerosols from land to sea, the average aerosol particle number of different species (OC, EC, NH4+, SO42−, NO3, and Na+) varies with the linear distance from Xiamen is given in Fig. 2b. The particle number of OC, EC, SO42−, and NO3 decreased rapidly during the first 100 km. The particle number of NH4+, SO42−, and NO3 was low and showed few changes when the linear distance from Xiamen exceeded 150 km. As seen in Fig. 2b, the SO42− and NO3 particle number showed a dramatically drop in the first 50 km, varying from 96 to 22 and 82 to 6, respectively, as the linear distance increased from 0 to 50 km. Previous studies have shown that anthropogenic nitric acid can react with sea salt to form coarse particles of sodium nitrate (NaNO3), increasing the deposition velocity of particles in the coastal area. Moreover, surface reactions also occurred between sulfuric acid and sea salt to produce sodium sulfate (Aller et al. 2005). Abundant sea salt particles in the atmospheric during this sea area accelerated the deposition of SO42− and NO3. As mentioned above, low particle numbers of SO42− and NO3 suggested a weak impact of continental sources on atmospheric aerosols in these sea areas, which was in accordance with the observation results obtained in Fig. 1. SO42−, NO3, and NH4+ are considered to be secondary aerosol species (Wang et al. 2006). However, the variation of NH4+ was different from that of SO42− and NO3. Low particle number count of NH4+ was present in this study.

Compared with the species of SO42− and NO3, the variations of OC and EC were more complicated. The OC particle number count was lower than the particle number count of EC and SO42− in the linear distance range from 0 to 150 km, while the particle number count of OC exceeded the other species as the linear distance was over 150 km. As mentioned above, except the fossil combustion source, OC was also derived from marine biogenic source. Biogenic aerosols were primarily emitted into the atmosphere through bubble bursting and had an important contribution to marine aerosols (Després et al. 2012 and O’Dowd et al. 2004). Significant positive correlation between OC and Na+ was present during the observation over XA, LSA, and SA, suggesting that Na+ and OC may have the same sources from marine bubble bursting in these areas. This can be certified by the variations of OC and Na+ fraction in Fig. 1. The fraction of OC was lower than that of Na+ and displayed the same variation trend in SA, LSA, and XA. But the relative fraction of OC was higher than the fraction of Na+ in SLA and became the most dominant component in atmospheric aerosols.

Particle size distribution in different areas

The temporal size distribution of aerosol is illustrated in Fig. 3. The highest average particle number count was present over XA with mean diameter of 510 nm. An accumulation mode of aerosol number size distribution with diameter range of 0.2–0.6 μm was present over XA, SLA, and LSA. While a coarse mode (0.6–2.0 μm) with mean diameter of 0.8 μm was found over SA. Previous studies have shown that fine particles with diameter of 0.2–0.6 μm were the major long range transport aerosols over offshore areas (Lin et al. 2007). Note that the residence time for accumulation mode particles was about 5–10 days (Fu et al. 2015; Shan 2010). In this case, the anthropogenic accumulation mode particles can be transported from land to the observation sea areas, indicating that marine atmospheric aerosols over XA, SLA, and LSA were influenced by the long range transport sources. Compare with the other seas, the lowest average particle number count was present over SA, but the mean diameter of aerosol was largest in this area. As seen in Fig. 1, atmospheric aerosols were governed by sea salt particles, accounting for more than 57% over SA. Marine aerosols were generated by whitecaps and bursting bubbles, which were in the coarse mode generally. Similar observation results were given in the previous studies (Fu et al. 2015; Yan et al. 2013). Furthermore, the influence of long range transport continental sources on atmospheric aerosols was weak in this area, resulting in low accumulation mode aerosol particle number. Hence, the coarse particles over SA were due to the contribution of marine aerosols and the lack of long range transport aerosols. This was consistent in the spatial distribution of marine aerosol chemical compositions in Fig. 1.
Fig. 3

Temporal size distribution of aerosol particles during the observation period

Carbonaceous particle signature and variation characteristics

Mass spectra characteristics of carbonaceous types

Carbonaceous material, including organic carbon (OC) and elemental carbon (EC), was one of the major dominant aerosol components in atmospheric aerosols in this study, accounting for 31%, 52%, 40%, and 62% over XA, LSA, SA, and SLA, respectively. Moreover, the source and variation of carbonaceous species in the observation areas were more complex than other aerosol components. OC particles were mainly from the fossil fuel combustion, biomass burning, and marine biogenic emissions. EC particles were associated with the incomplete combustion of carbon-containing materials, such as vehicle exhaust, coal combustion, ship emission, and biomass burning (Liu and Shao 2007). Five main subtypes of carbonaceous particles were identified as C_Al-Si, C_V-Ni, C_S, C_K, and C_secondary based on the mass spectral similarity. The average positive and negative mass spectra of major carbonaceous clusters are illustrated in Fig. 4.
Fig. 4

Average positive and negative mass spectra of major carbonaceous particle types. a C_Al-Si, b C_V-Ni, c C_S, d C_K, and e C_secondary

The type of C_Al-Si particles was dominated by the progressions of CnHm+ ions in the mass spectra, as seen in Fig. 4a. A high intensity of Al+ signature peaking at m/z 27 was present in positive spectrum, while an evident silicon signature of SiO2− and SiO3 peaking at m/z − 60 and m/z − 76 was present in negative spectrum. Strong ion peaks of HSO4 with low peaks of NO2, NO3, and SO32− were observed in the negative spectrum. Al, Si, and O were the main components of coal combustion fine particles (Yan et al. 2016). Intense peaks of Al+, SiO2, and SiO3 in the C_ Al-Si particles indicated that those particles were from coal combustion. SO42− was also an important composition of fine particles from coal-fired power plant emission, since WFGD system was installed to reduce the emission of SO2 from coal-fired power plant (Yang et al. 2010; Bao et al. 2013). High peak of SO42− in C_Al-Si particles also confirmed that this type of carbonaceous particles was derived from coal combustion emission. As seen in Fig. 5, the relative percentage of C_Al-Si particles was about 25% over XA, but the value decreased rapidly with the distance from land to sea increased, suggesting that the type of C_Al-Si particles was from continental sources.
Fig. 5

Average fractions of different types of carbonaceous particles (C_Al-Si, C_V-Ni, C_S, C_K, and C_secondary) and the total carbonaceous particle number vary with the linear distance from Xiamen

The average spectra of C_V-Ni particles are shown in Fig. 4b, which were dominated by Cn+ and Cn peaks in positive and negative spectra. An evident intense of V+ peak with weak peaks of VO+, Ni+, and Na+ was also present in positive spectrum. Note that V+ and Ni+ containing particles were regarded as a suitable tracer for ship emission. The presence of V+, VO+, and Ni+ in the C_V-Ni particles indicated that those particles were from ship emission. The lack of NO2, NO3, and SO42− fragment ions in C_V-Ni particles indicated that this type of C_V-Ni particles has not been aged in the atmosphere. NO2, NO3, and SO42− were considered to be the secondary species in atmospheric aerosols, and the uptake of these species on EC particles was often observed in aged aerosols (Zhang et al. 2007).

C_S was characterized by peaks associated with K+, Na+, Mg+, NaCl+, and oxygenated organics (CxHyOz) in positive spectrum, as seen in Fig. 4c. Strong peaks for methanesulfonic acid (MSA) and HSO4 with weak peaks of CNO and CHNO were present in negative spectrum. MSA was the oxidation product of dimethylsulfide (DMS) in the atmosphere (Zhang et al. 2015), which was an important marine biogenic gas oxidized by OH, BrO, and NO3 radicals by addition and abstraction reaction routs (Von Glasow and Crutzen 2004; Legrand and Pasteur 1998). Intense peaks of MSA and HSO4 in C_S particles indicated that this type of particles was mainly from the secondary conversion of DMS, deriving from the marine biological activity.

C_K particles were characterized by a high intense peak of K+ with weak intense Na+ and Ca+ ion peaks in positive spectrum, seen in Fig. 4d. While high intense peaks of CN, CNO, CHNO, and C2H3O2 with weak peaks of SO42−, O, and OH were observed in negative spectrum. Generally, K+ was identified as a suitable tracer for biomass burning particles (Spencer et al. 2007; Guazzotti et al. 2003.). However, biomass burning was not the only source of K particles. Previous study showed that particles from coal combustion or biological activity emission also contributed to the source of K particles (Wang et al. 2013; Ma et al. 2016). As seen in Fig. 4d, the presence of Na+ and Ca+ and the lack of carbonaceous species suggested that this type of C_K particles was influenced by the emission of marine biological activity. Studies have found that K containing particles can also be associated with marine sources by the bubble bursting (Yan et al. 2013).

C_secondary particles were dominated by intense CxHy+ and CxHyOz+ peaks with weak peak of Na+ in positive spectrum, as seen in Fig. 4e. While high ion peaks of NO2, NO3, and SO42−, C2H2+ with weak peaks of O, OH, CN, CNO, and SO32− were present in negative spectrum. SO42−, NO3, and NH4+ were believed as derived mainly from secondary pollution particles produced by the transformation of their precursors of SO2, NO2, and NH3 (Wang et al. 2006). Generally, the heterogeneous reaction uptake of secondary species on the surfaces of primary aerosol particles was much easier than homogeneous reaction in the atmosphere (Yu and Turco 2011). Hence carbonaceous species often mixed with secondary sulfate and nitrate species. High percentage of C_secondary particles was observed over XA, as seen in Fig. 5. The contribution of this type of particles decreased rapidly with the increase of linear distance between the observation location and land. Relatively abundant gas precursors and high aerosol concentration with high value of RH were observed over XA, which was prone to the formation of secondary particles in this area (Foltescu et al. 1996). However, the aerosol number and gas precursors declined as the distance between observation location and continent increased, resulting in the decrease of secondary particles.

Variation characteristics of carbonaceous particles

Figure 5 shows the average fraction of different carbonaceous particle types, and the total carbonaceous particle number varies with the linear distance from Xiamen. The total carbonaceous particle number displayed a decreasing tendency as the linear distance from land to sea increased, suggesting that the carbonaceous aerosols were strongly influenced by the continental sources. However, note that the carbonaceous particle number increased as the linear distance increased from 300 to 350 km. Compared with the average fraction given in 300 km, the percentage of C_V-Ni type particles increased from 16.5 to 32.1%. As mentioned above, C_V-Ni particles were mainly from ship emission. Hence, the increase of carbonaceous particles was determined by the ship emission aerosols in this area.

As seen in Fig. 5, high fraction of C_Al-Si particles and C_secondary particles was observed over coastal area, accounting for 23.8% and 18.6%, individually. But the fraction of both type of particles decreased as the linear distance from land to sea increased. Contrarily, the relative percentage of C_S showed an increasing tendency as the linear distance from land to sea increased. Therefore, C_Al-Si and C_secondary particles were determined by the continental sources, while the C_S particles were strongly affected by marine sources, which were accord with the particle cluster analyses obtained in Fig. 4. However, the variation of C_V-Ni particles was more complex in this study. This type of particles was mainly from the primary ship emission. Hence, the type of C_V-Ni particles was concerned with the vessel activity in this area but not the distance from land.

Characteristics of particle sources in different sea areas

Particles were merged into six major classes (sea salt particles, secondary aerosols, K-rich particles, heavy metal particles, EC, and OC particles) covering over 95% of the total particles using the ART-2a algorithm in this study. The temporal distribution of six particle types during the whole observation period is shown in Fig. 6. Aerosols were dominated by the interaction of continental and marine sources over XA. High relative fraction of secondary aerosols was present in this area, which was similar with the previous studies (Zhao et al. 2011). However, the percentage of secondary aerosols was extremely low over LSA, SA, and SLA. High fraction of EC and K-rich was observed over LSA. While high percentage of sea salt particles and OC particles was found over SLA, and the K-rich particles were extremely low in this area. Compared with other periods, sea salt particles and OC particles were the two dominant components over SA, indicating that atmospheric aerosols were governed by marine sources in this area.
Fig. 6

Temporal distribution of particle sources during the whole observation period

To further determine the possible influence of air masses on atmospheric aerosol sources through long range transport. The average air masses of 72-h back trajectories and different classes of aerosols are illustrated in Fig. 7. Similar air mass types over XA, SLA, and LSA were present with most of the trajectories from the ocean south or southwest of the observation locations. Most of the trajectories were from the surrounding area or the southeast of the observation location over SA. High percentage of sea salt particles was present in Fig. 7a, indicating that the air masses from ocean enhanced the atmospheric sea salt aerosols over XA. Similar results over coastal area of Xiamen were obtained in previous studies (Yan et al. 2016). As mentioned above, aerosols from continental sources decreased rapidly in the transport process from land to sea. This may be impacted by air masses transport during the observation period, since air masses over LSA, SLA, and SA were all from ocean, which enhanced marine aerosol sources and diluted the air masses from continent. As seen in Fig. 7b, c, and d, the fraction of continental aerosol sources (secondary aerosol and heavy metal containing particles) was extremely low over LSA, SLA, and SA. The continental sources would play a more important role in these observation sea areas, if air masses were from the continent. The variations of OC and EC containing particles were more complex, high percentage of EC containing particles, accounting for more than 40%, was observed in LSA, seen in Fig. 7b, while high OC fractions (35% and 17%) were present in SLA and SA, because OC and EC containing particles were derived not only from continental sources but also from ship emissions, biological emission, and marine sources (Yin et al. 2012).
Fig. 7

Average air mass back trajectories (72 h) and different classes of aerosols (sea salt particles, secondary aerosols, K-rich particles, heavy metal, EC, and OC particles). a XA. b LSA. c SLA. d SA

Conclusions

A high-time-resolution on-board single particle aerosol mass spectrometer (SPAMS) was deployed on “Xiang Yang Hong 03” vessel to measure size and chemical composition of single marine aerosol simultaneously over Southeastern China Sea to provide the impact of continental and marine sources on atmospheric aerosols over the coastal areas. The observation areas were divided into four main parts, Xiamen anchorage (XA), land to sea area (LSA), sea area (SA), and sea to land area (SLA). Sulfate, nitrate, Na+, and carbonaceous (including OC and EC) were the dominant aerosol species over XA, accounting for 25%, 21%, 19%, and 32%, representing the coastal urban area. LSA and SLA represented the interaction of continental and marine sources, as abundance of Na+, NO3, SO42−, and OC were present in these areas. Atmospheric aerosols were governed by marine sources over SA with 57% of Na+ containing particles of the total particles. The particle number of OC, EC, SO42−, and NO3 decreased rapidly during the first 100 km from land to sea. The particle number of NH4+, SO42−, and NO3 containing particles was low and showed few changes when the linear distance from Xiamen exceeded 150 km.

Carbonaceous material, including organic carbon (OC) and elemental carbon (EC), was the major dominant aerosol components in this study, accounting for 31%, 52%, 40%, and 62% over XA, LSA, SA, and SLA, respectively. The source and variation of carbonaceous species in the observation areas were more complex than other aerosol species. Five major subtypes of carbonaceous particles were identified as C_Al-Si, C_V-Ni, C_S, C_K, and C_secondary based on the mass spectral similarity. High fraction of C_Al-Si particles and C_secondary particles was observed in coastal area, accounting for 23.8% and 18.6%. But the fraction of both types decreased, as the distance from land to sea increased. Contrarily, the relative percentage of C_S showed an increasing tendency, as the distance from land to sea increased. The variation of C_V-Ni containing particles was strongly influenced by the vessel activity in the observation areas.

In this study, six major classes were identified based on the mass spectra from SPAMS and clustered using ART-2a algorithm, including sea salt particles, secondary aerosols, K-rich particles, heavy metal particles, EC, and OC particles. The atmospheric aerosols were dominated by the interaction of continental and marine sources over XA. High fraction of secondary aerosols was observed over XA, while the value was extremely low in other sea areas. This was caused by air masses transport during the observation period, since most of the trajectories were from the ocean during the observation period.

Notes

Acknowledgements

The authors gratefully acknowledge Guangzhou Hexin Analytical Instrument Company Limited for the SPAMS data analysis and on-board observation technical assistance. The authors gratefully acknowledge NOAA Air Resources laboratory (ARL) for the provision of the HYSPLIT_4 transport model used in this publication.

Funding information

This study is financially supported by Qingdao National Laboratory for marine science and technology (No. QNLM2016ORP0109), the Natural Science Foundation of Fujian Province, China (No. 2015J05024), the National Natural Science Foundation of China (No. 21106018 and No. 41305133), the Scientific Research Foundation of Third Institute of Oceanography, SOA. (No. 2014027), and the Special Fund for Marine Researches in the Public Interest (No. 2004DIB5J178).

Supplementary material

11356_2018_2902_MOESM1_ESM.doc (14.6 mb)
ESM 1 (DOC 14938 kb)

References

  1. Allen JO (2005) YAADA: software toolkit to analyze single-particle mass spectral data. https://www.arb.ca.gov/research/apr/past/01-338_manual.pdf
  2. Aller JY, Kuznetsova MR, Jahns CJ, Kemp PF (2005) The sea surface microlayer as a source of viral and bacterial enrichment in marine aerosols. J Aerosol Sci 36:801–812CrossRefGoogle Scholar
  3. Bao JJ, Yang LJ, Yan JP, Xiong GL, Lu B, Xin CY (2013) Experimental study of fine particles removal in the desulfurated scrubbed flue gas. Fuel 108:73–79CrossRefGoogle Scholar
  4. Bi XH, Zhang GH, Li L, Wang XM, Li M, Sheng GY (2011) Mixing state of biomass burning particles by single particle aerosol mass spectrometer in the urban area of PRD, China. Atmos Environ 45:3447–3453CrossRefGoogle Scholar
  5. Chalbot MC, Mcelroy B, Kavouras I (2013) Sources, trends and regional impacts of fine particulate matter in southern Mississippi Valley: significance of emissions from sources in the Gulf of Mexico coast. Atmos Chem Phys 13:3721–3732CrossRefGoogle Scholar
  6. Chen Y, Zhuang GS, Guo ZG (2010) Atmospheric deposition of nutrients and trace elements to the coastal oceans: a review. Adv Earth Science 25:682–689Google Scholar
  7. Decesari S, Finessi E, Rinaldi M, Paglione M, Fuzzi S, Stephanou EG, Tziaras T, Spyros A, Ceburnis D, O’Dowd C, Dall’Osto M, Harrison RM, Allan J, Coe H, Facchini MC (2011) Primary and secondary marine aerosols over the North Atlantic Ocean during the MAP experiment. J Geophys Res 116:1–21CrossRefGoogle Scholar
  8. Després VR, Huffman A, Burrows SM, Hoose C, Safatov AS, Buryak G (2012) Primary biological aerosol particles in the atmosphere: a review. Tellus Ser B 64:15598CrossRefGoogle Scholar
  9. Falkowska L, Lewandowska A (2004) Sulphates in particles of different sizes in the marine boundary layer over the southern Baltic Sea. Oceanologia 46:201–215Google Scholar
  10. Foltescu VL, Lindgred ES, Isakson J, Oblad M, Pacyna JM, Benson S (1996) Gas-to-particle conversion of sulphur and nitrogen compounds as studied at marine stations in northern Europe. Atmos Environ 30:3129–3140CrossRefGoogle Scholar
  11. Fu H, Zheng M, Yan C, Li X, Gao H, Yao X, Guo Z, Zhang Y (2015) Sources and characteristics of fine particles over the Yellow Sea and Bohai Sea using online single particle aerosol mass spectrometer. J Environ Sci 29:62–70CrossRefGoogle Scholar
  12. Gaston CJ, Furutani H, Guazzotti SA, Coffee KR, Bates TS, Quinn PK (2011) Unique ocean-derived particles serve as a proxy for changes in ocean chemistry. J Geophys Res 116:597–616CrossRefGoogle Scholar
  13. Gkikas A, Hatzianastassiou N, Mihalopoulos N, Torres O (2016) Characterization of aerosol episodes in the greater Mediterranean Sea area from satellite observations (2000–2007). Atmos Environ 128:286–304CrossRefGoogle Scholar
  14. Guazzotti SA, Suess DT, Coffee KR, Quinn PK, Bates TS, Wisthaler A, Hansel A, Ball WP, Dickerson RR, Neususs C, Crutzen PJ, Prather KA (2003) Characterization of carbonaceous aerosols outflow from India and Arabia: biomass / biofuel burning and fossil fuel combustion. J Geophys Res 108:1211–1222CrossRefGoogle Scholar
  15. Hoomaert S, Godoi RHM, Van Grieken R (2003) Single particle characterization of the aerosol in the marine boundary layer and free troposphere over Tenerife, NE Atlantic, during ACE-2. J Atmos Chem 46:271–293CrossRefGoogle Scholar
  16. Jimenez JL, Jayne JT, Shi Q, Kolb CE, Worsnop DR, Yourshaw I, Seinfeld JH, Flagan RC, Zhang X, Smith KA, Morris JW, Davidovits P (2003) Ambient aerosol sampling with an aerosol mass spectrometer. J Geophys Res 108:8425CrossRefGoogle Scholar
  17. Kim JH, Yum SS, Lee YG, Choi BC (2009) Ship measurements of submicron aerosol size distributions over the Yellow Sea and the East China Sea. Atmos Res 93:700–714CrossRefGoogle Scholar
  18. Legrand M, Pasteur EC (1998) Methane sulfonic acid to non-sea-salt sulphate ration in coastal Antarctic aerosol and surface snow. J Geophys Res 103:10991–11006CrossRefGoogle Scholar
  19. Li L, Li M, Huang ZX, Gao W, Nian HQ, Fu Z, Gao J, Chai FH, Zhou Z (2014) Ambient particle characterization by single particle aerosol mass spectrometry in an urban area of Beijing. Atmos Environ 94:323–331CrossRefGoogle Scholar
  20. Lin P, Hu M, Wu ZJ, Niu YW, Zhu T (2007) Marine aerosol size distributions in the springtime over China adjacent seas. Atmos Environ 41:6784–6796CrossRefGoogle Scholar
  21. Liu Y, Shao M (2007) Estimation amd prediction of black carbon emissions in Beijing city. Chin Sci Bull 52:1274–1281CrossRefGoogle Scholar
  22. Ma L, Li M, Zhang H, Li L, Huang Z, Gao W, Chen D, Fu Z, Nian H, Zou L, Chai F, Zhou Z (2016) Comparative analysis of chemical composition and sources of aerosol particles in urban Beijing during clear, hazy, and dusty days using single particle aerosol mass spectrometry. J Clean Prod 112:1319–1329CrossRefGoogle Scholar
  23. Masalaite A, Holzinger R, Remeikis V, Rockmann T, Dusek U (2017) Characteristics, sources and evolution of fine aerosol (PM1) at urban, coastal and forest background sites in Lithuania. Atmos Environ 148:62–76CrossRefGoogle Scholar
  24. Murphy D, Cziczo D, Froyd K, Hudson P, Matthew B, Middlebrook A, Peltier R, Sullivan A, Thomson D, Weber R (2006) Single-particle mass spectrometry of tropospheric aerosol particles. J Geophys Res Atmos 111(D23):1–15Google Scholar
  25. Murphy SM, Agrawal H, Sorooshian A, Padró LT, Gates H, Hersey S (2009) Comprehensive simultaneous shipboard and airborne characterization of exhaust from a modern container ship at sea. Environ Sci Technol 43:4626–4640CrossRefGoogle Scholar
  26. Nakamura T, Ogawa H, Maripi DK, Uematsu M (2006) Contribution of water soluble organic nitrogen to total nitrogen in marine aerosols over the East China Sea and western North Pacific. Atmos Environ 40:7259–7264CrossRefGoogle Scholar
  27. Norris SJ, Brooks IM, Moat BI, Yelland MJ (2013) Near-surface measurement of sea spray aerosol production over whitecaps in the open ocean. Ocean Sci 9:133–145CrossRefGoogle Scholar
  28. O’Dowd CD, Facchini MC, Cavalli F, Ceburnis D, Mircea M, Decesari S (2004) Biogenically driven organic contribution to marine aerosol. Nature 431:676–680CrossRefGoogle Scholar
  29. Ohlstrom MO, Lehtinen KEJ, Moisio M, Jokiniemi JK (2000) Fine-particle emissions of energy production in Finland. Atmos Environ 34:3701–3711CrossRefGoogle Scholar
  30. Phinney L, Leaitch WR, Lohmann U, Boudries H, Worsnop DR, Jayne JT (2006) Characterization of the aerosol over the sub-arctic north East Pacific Ocean. Deep-Sea Res II Top Stud Oceanogr 53:2410–2433CrossRefGoogle Scholar
  31. Shan J (2010) Study on measurement method and its application to determining the residence time of atmospheric aerosol. (PhD thesis). Nanhua University, ShanghaiGoogle Scholar
  32. Shi J, Zhang J, Gao HW, Tan SC, Yao XH, Ren JL (2013) Condentration, solubility and deposition flux of atmospheric particulate nurients over the Yellow Sea. Deep-Sea Res II 97:43–50CrossRefGoogle Scholar
  33. Sitaras IE, Siskos PA (2008) The role of primary and secondary air pollutants in atmospheric pollution: Athens urban area as a case study. Environ Chem Lett 6:59–69CrossRefGoogle Scholar
  34. Song XH, Hopke PK, Fergenson DP, Prather KA (1999) Classification of single particles analyzed by ATOFMS using an artificial neural network, ART-2A. Anal Chem 71:860–865CrossRefGoogle Scholar
  35. Spencer MT, Shields LG, Prather KA (2007) Simultaneous measurement of the effective density and chemical composition of ambient aerosol particles. Environ Sci Technol 41:1303–1309CrossRefGoogle Scholar
  36. Sullivan RC, Guazzotti SA, Sodeman DA, Prather KA (2007) Direct observations of the atmospheric processing of Asian mineral dust. Atmos Chem Phys 7:1213–1236CrossRefGoogle Scholar
  37. Udisti R, Dayan U, Becagli S, Busetto M, Frosini D, Legrand M, Lucarelli F, Preunkert S, Severi M, Traversi R, Vitale V (2012) Sea spray aerosol in central Antarctica. Present atmospheric behaviour and implications for paleoclimatic reconstructions. Atmos Environ 52:109–120CrossRefGoogle Scholar
  38. Von Glasow R, Crutzen P (2004) Model study of multiphase DMS oxidation with a focus on halogens. Atmos Chem Phys 4:589–608CrossRefGoogle Scholar
  39. Wang ZH (2018) Energy and air pollution. Comprehensive Energy Sys 1:909–949CrossRefGoogle Scholar
  40. Wang Y, Zhuang G, Sun Y, Zheng A (2006) The variation of characteristics and formation mechanisms of aerosols in dust, haze, and clear days in Beijing. Atmos Environ 40:6579–6591CrossRefGoogle Scholar
  41. Wang ZB, Hu M, Wu ZJ, Yue DL, He LY, Huang XF, Liu XG, Wiedensohler A (2013) Long-term measurements of particle number size distributions and the relationships with air mass history and source apportionment in the summer of Beijing. Atmos Chem Phys 13:10159–10170CrossRefGoogle Scholar
  42. Woods E, Chung D, Lanney HM, Ashwell BA (2010) Surface morphology and phase transitions in mixed Nacl/MgSO4 aerosol particles. J Phys Chem 114:2837–2844CrossRefGoogle Scholar
  43. Yan JP, Chen LQ, Lin Q, Zhang YH, Li Z (2013) A study of aerosol chemical compositions and size distribution property on the coastal area of southern Xiamen island. J Appl Oceanogr 32:455–460Google Scholar
  44. Yan J, Chen L, Lin Q, Li Z, Chen H, Zhao S (2015) Chemical characteristics of submicron aerosol particles during a long-lasting haze episode in Xiamen, China. Atmos Environ 113:118–126CrossRefGoogle Scholar
  45. Yan JP, Chen LQ, Yang L (2016) Combined effect of acoustic agglomeration and vapor condensation on fine particles removal. Chem Eng J 290:319–327CrossRefGoogle Scholar
  46. Yang L, Bao J, Yan J, Liu J, Song S, Fan F (2010) Removal of fine particles in wet flue gas desulfurization system by heterogeneous condensation. Chem Eng J 156:25–32CrossRefGoogle Scholar
  47. Yao Q, Li S, Xu H, Zhuo J, Song Q (2009) Studies on formation and control of combustion particulate matter in China: a review. Energy 34:1296–1309CrossRefGoogle Scholar
  48. Yin LQ, Niu ZC, Chen XQ, Chen JS, Xu LL, Zhang FW (2012) Chemical compositions of PM2.5 aerosol during haze periods in the mountainous city of Yong’an, China. J Environ Sci 24:1225–1233CrossRefGoogle Scholar
  49. Yu F, Turco RP (2011) The size-dependent charge fraction of sub-3-nm particles as a key diagnostic of competitive nucleation mechanisms under atmospheric conditions. Atmos Chem Phys 11:9451–9463CrossRefGoogle Scholar
  50. Zhang Q, Jimenez JL, Canagaratna MR, Allan JD, Coe H, Ulbrich I, Alfarra MR, Takami A, Middlebrook AM, Sun YL, Dzepina K, Dunlea E, Docherty K, DeCarlo PF, Salcedo D, Onash T, Jayne JT, Miyoshi T, Shimono A, Hatakeyama S, Takegawa N, Kondo Y, Schneider J, Drewnick F, Weimer S, Demerjian K, Williams P, Bower K, Bahreini R, Cottell R, Griffin RJ, Rautiainen J, Sun JY, Zhang YM, Worsnop DR (2007) Ubiquity and dominance of oxygenated species in organic aerosols in anthropogenically-influenced northern hemisphere mid latitudes. Geophys Res Lett 34:L13801Google Scholar
  51. Zhang F, Xu L, Chen J, Zhao J, Yu Y, Niu Z, Yin L (2012) Chemical compositions and extinction coefficients of PM2.5 in peri-urban of Xiamen, China, during June 2009-May 2010. Atmos Res 106:150–158CrossRefGoogle Scholar
  52. Zhang TR, Shi JH, Gao HW, Zhang J, Yao XH (2013) Impact of source and atmospheric processing on Fe solubility in aerosols over the Yellow Sea, China. Atmos Environ 75:249–256CrossRefGoogle Scholar
  53. Zhang MM, Chen LQ, Xu GJ, Liang MY (2015) Linking phytoplankton activity in polynyas and sulfur aerosols over Zhangshan Station, East Antarctica. J Atmos Sci 72:4629–4642CrossRefGoogle Scholar
  54. Zhao J, Zhang F, Xu Y, Chen J (2011) Characterization of water-soluble inorganic ions in size-segregated aerosols in coastal city, Xiamen. Atmos Res 99:546–562CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Third Institute of Oceanography, State Oceanic AdministrationXiamenPeople’s Republic of China
  2. 2.Key Laboratory of Global Change and Marine-Atmospheric ChemistryXiamenChina
  3. 3.Institute of Atmospheric Environment Safety and pollution ControlJinan UniversityGuangzhouChina

Personalised recommendations