Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/ doi:10.5194/acp-17-1-2017 Author(s) 2017. CC Attribution 3.0 License.
Michael R. Giordano1,a, Lars E. Kalnajs2, Anita Avery1, J. Douglas Goetz1, Sean M. Davis3,4, and Peter F. DeCarlo1,5
1Department of Civil, Architectural, and Environmental Engineering, Drexel University, Philadelphia, Pennsylvania, USA
2Laboratory for Atmospheric and Space Physics, University of Colorado at Boulder, Boulder, Colorado, USA
3Chemical Sciences Division, NOAA Earth System Research Laboratory, Boulder, Colorado, USA
4Cooperative Institute for Research in Environmental Sciences, University of Colorado at Boulder, Boulder, Colorado, USA
5Department of Chemistry, Drexel University, Philadelphia, Pennsylvania, USA
anow at: AAAS Science and Technology Fellowship, Hosted at US Environmental Protection Agency, Washington DC, USA Correspondence to: Peter F. DeCarlo ([email protected])
Received: 10 July 2016 Published in Atmos. Chem. Phys. Discuss.: 14 July 2016
Revised: 3 November 2016 Accepted: 4 November 2016 Published: 2 January 2017
Abstract. Understanding the sources and evolution of aerosols is crucial for constraining the impacts that aerosols have on a global scale. An unanswered question in atmospheric science is the source and evolution of the Antarctic aerosol population. Previous work over the continent has primarily utilized low temporal resolution aerosol lters to answer questions about the chemical composition of Antarctic aerosols. Bulk aerosol sampling has been useful in identifying seasonal cycles in the aerosol populations, especially in populations that have been attributed to Southern Ocean phytoplankton emissions. However, real-time, high-resolution chemical composition data are necessary to identify the mechanisms and exact timing of changes in the Antarctic aerosol. The recent 2ODIAC (2-Season Ozone Depletion and Interaction with Aerosols Campaign) eld campaign saw the rst ever deployment of a real-time, high-resolution aerosol mass spectrometer (SP-AMS soot particle aerosol mass spectrometer or AMS) to the continent. Data obtained from the AMS, and a suite of other aerosol, gas-phase, and meteorological instruments, are presented here. In particular, this paper focuses on the aerosol population over coastal Antarctica and the evolution of that population in austral spring. Results indicate that there exists a sulfate mode in Antarctica that is externally mixed with a mass mode vacuum aerodynamic diameter of 250 nm. Springtime increases in sulfate aerosol are observed and attributed to biogenic sources, in agreement with previous research identifying phytoplankton activity as the source of the aerosol. Furthermore, the to-
tal Antarctic aerosol population is shown to undergo three distinct phases during the winter to summer transition. The rst phase is dominated by highly aged sulfate particles comprising the majority of the aerosol mass at low wind speed. The second phase, previously unidentied, is the generation of a sub-250 nm aerosol population of unknown composition. The second phase appears as a transitional phase during the extended polar sunrise. The third phase is marked by an increased importance of biogenically derived sulfate to the total aerosol population (photolysis of dimethyl sulfate and methanesulfonic acid (DMS and MSA)). The increased importance of MSA is identied both through the direct, real-time measurement of aerosol MSA and through the use of positive matrix factorization on the sulfur-containing ions in the high-resolution mass-spectral data. Given the importance of sub-250 nm particles, the aforementioned second phase suggests that early austral spring is the season where new particle formation mechanisms are likely to have the largest contribution to the aerosol population in Antarctica.
1 Introduction
Although the present aerosol burden and processes are still relatively poorly understood, aerosol particles contained in ice cores obtained from the Antarctic ice shelves are used as proxies for many properties of the paleo-atmosphere. The abundance of mineral dust, sea salt aerosol (Fischer et al.,
Published by Copernicus Publications on behalf of the European Geosciences Union.
A missing source of aerosols in Antarctica beyond long-range transport, phytoplankton, and photochemistry
2 M. R. Giordano et al.: A missing source of aerosols in Antarctica
2007), and sulfur compounds (Legrand et al., 1988; Mulvaney et al., 1992) in Antarctic ice cores has been used to provide information on glaciation, sea level, cloudiness, and volcanic activity over the past millennia. Aerosols also play a major role in Earths current climate due to their impact on the global radiative balance and cloud microphysics.Unfortunately, aerosols are still the least understood and constrained aspects of the climate system (Boucher et al., 2013). The uncertainty of aerosols climate impacts arises from the fact that how an aerosol affects the radiative balance is a function of both an aerosols chemical composition and physical properties (e.g., size, shape). Both the chemical and physical properties of aerosols are functions of emission sources, atmospheric processing, and lifetime in the atmosphere. Over recent decades, much work has been done to characterize aerosol emission sources and background aerosol across much of the globe (Boucher et al., 2013), but there are difculties in assessments of preindustrial to present-day forcing. Measurements in Antarctica, provide insight into one of the more pristine environments and can be useful in the understanding of preindustrial background aerosol (e.g., Hamilton et al., 2014). However, the ability to sample pristine aerosols is directly related to an areas inaccessibility. Because of the difculty in performing science in Antarctica, the Antarctic aerosol mass and number population (particularly its sources and evolution) is still a subject of many open questions in atmospheric science. Improving our understanding of the processes that govern aerosol formation and evolution in Antarctica is important not only to our understanding of present-day Antarctica, but also to understanding the broader climate history.
Besides a few research stations, bases, and minor tourism activities, anthropogenic emission sources of aerosols and trace gases are rare in and around Antarctica (Shirsat and Graf, 2009). Much of the work produced over the years examining Antarctic aerosols has been done precisely because of the lack of direct anthropogenic atmospheric inuences. Multiple investigations of the concentration, size distribution, spatial distribution, and composition of Antarctic aerosols have been performed over various parts of the continent (e.g., Shaw, 1979; Lechner et al., 1989; Harwey et al., 1991; Savoie et al., 1993; Hara et al., 1996; Minikin et al., 1998, Koponen et al., 2003). Low time-resolution bulk aerosol analysis (generally lter-integrated) dominates the chemical composition measurements over the continent (e.g., Prospero et al., 1991; Wagenbach, 1996; Minikin et al., 1998;Preunkurt et al., 2007) though real-time measurement techniques have been deployed in recent years (e.g., Belosi et al., 2012).
The results of aerosol measurements over Antarctica over the past decades have been generally consistent on two major points: that a persistent, low concentration, aerosol population exists over the entirety of the continent and that sulfate is a major component of that aerosol, especially in the austral spring and summer (Shaw, 1979; Parungo et al., 1981;
Wagenbach et al., 1988). On a global scale, anthropogenic sources of sulfur dominate the sulfate aerosol mass which has long been known to be a major component of aerosol-induced climate forcing (Charlson et al., 1987, 1990; Kulmala et al., 2002). However, in the Southern Hemisphere, and particularly in Antarctica, biogenic sources are thought to be the largest source of sulfur to the atmosphere (Bates et al., 1992; Carslaw et al., 2010, and references therein). Hence, the question of the origin of the Antarctic aerosol sulfate arises. Globally, volcanic activity provides a major source of atmospheric sulfur so the presence of Mt Erebus (78 S), the southernmost active volcano, should be noted. However, isotopic studies have shown that, in Antarctica, sulfate of volcanic origin is of minimal importance although infrequent, short-lived exceptions due to eruptions have occurred (Patris et al., 2000). The descent of stratospheric sulfur into the lower atmosphere is another potential sulfur source but has also been shown to be a minor source of sulfate around coastal Antarctica (Legrand and Wagenbach, 1998). The two most likely remaining origins of sulfur are ocean sea-spray and non-sea-spray marine sources.
Marine sulfur appears in both primary and secondary aerosols. Primary marine aerosols, generally referred to as sea-spray aerosols, are produced through mechanical actions such as wave breaking and bubble bursting. Sea-spray aerosols are mostly super-micron in size and production is a strong function of wind speed (Lewis and Schwartz, 2004).The super-micron sea-spray aerosol is dominated by sea salt, of which sulfate is 8 % by weight in ocean water. In Antarctica, previous measurements have placed sea salt as 5080 % by mass of the (sub-10 m) aerosol population depending on the time of year (Weller et al., 2008). Marine secondary aerosols, however, are driven by biological emissions of volatile organic compounds. Of these compounds, dimethyl sulde (DMS) and its oxidation product methanesulfonic acid (MSA) are the most common and account for 75 % of the global biogeochemical sulfur cycle (Chasteen and Bentley, 2004). In Antarctica, enhancements in sulfate-to-sodium ratios (over the seawater ratio) in aerosols have been attributed to DMS and MSA. Off-line measurement and calculation of non-sea-spray (nss-) sulfate have been used to identify seasonal cycles of aerosol sulfate fractions over Antarctica (e.g., Prospero et al., 1991; Wagenbach, 1996; Minikin et al., 1998; Preunkurt et al., 2007). Spring and summertime enhancements in phytoplankton activity in the Southern Ocean provide an excellent explanation for the spring sulfate aerosol enhancements seen over Antarctica (Gibson et al., 1990; Arimoto et al., 2001; von Glasow and Crutzen, 2004; Preunkert et al., 2007; Read et al., 2008; Weller et al., 2011a). Unfortunately, non-sea-spray sulfate has generally been calculated using aerosol sodium concentrations. Because sodium is nonconservative in Antarctic aerosols due to mirabilite (Na2SO4 [notdef] 10H2O) precipitation in sea ice microstructures,
nss-sulfate calculations can be biased (Rankin et al., 2000;
Rankin and Wolff, 2003). Direct measurements of gas-phase
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
sea ice prevented the collocation of the eld sites over both eld seasons. However, the 2015 site was chosen such that the wind fetch was similar to the fetch from the 2014 site. In 2014, the eld site was located approximately 16 km away from the sea ice edge as compared to approximately 8 km away in 2015.
In both eld seasons, the eld sites consisted of 2 structures (sh huts). One structure housed the atmospheric sampling instrumentation, while the other housed a 5 kW diesel generator. The generator hut was placed 75 m to the southwest of the instrumentation hut because at both eld sites the wind blew primarily from the northwest and southeast. The distance between and orientation of the huts ensured that minimal self-sampling occurred over both eld seasons. During low-wind periods and certain wind shifts, self-sampling and sampling of McMurdo did occur, but these periods were identied and removed from the data via careful observation of NOx and aerosol mass spectrometer (AMS) data. The thresholds in regard to self-contamination were applied if (a) a rapid increase in NOx and associated decrease in ozone and (b) the organic signal in the AMS at combustion-relevant m/z (e.g., m/z 57 or 55) increased by greater than 20 %.
2.2 Instrumentation
Atmospheric sampling instrumentation for both seasons consisted of a suite of aerosol and gas-phase instruments. The gas-phase suite was composed of an ozone monitor (1 s time resolution; Thermo Environmental model 49C) and an NOx monitor (1 s time resolution; Thermo Environmental model 42C). The aerosol suite consisted of aerosol sizing instruments, an aerosol concentration counter, aerosol composition measurements, and aerosol collectors. In 2014, aerosol sizing was carried out via a scanning electrical mobility spectrometer ( 9 to 850 nm; 2 min time resolution; Brechtel Man.
Inc.; SEMS) and an ultra-high sensitivity aerosol spectrometer ( 55 to > 1000 nm; 1 s time resolution; Droplet Meas.
Tech.; UHSAS). In 2015, aerosol sizing was carried out via a scanning mobility particle sizer ( 10 to 420 nm; 2 min
time resolution; TSI, 3080/3081 SMPS and 3787 CPC) and an optical particle counter (0.3 to 25 m; 1 s time resolution; OPC Lighthouse Remote 3104). In both years, total aerosol counts were measured with a water-based condensation particle counter (7 nm to > 3 m; TSI, 3783 environmental particle counter EPC), and aerosol composition was measured with a soot particle aerosol mass spectrometer (Aerodyne Research Inc. Billerica, MA; SP-AMS;DeCarlo et al., 2006; Onasch et al. 2012). The SP-AMS is a combination of the Aerodyne High-Resolution Time-of-Flight aerosol mass spectrometer (HR-ToF-AMS) and a soot vaporizing laser (from Droplet Meas. Tech.). The soot vaporizing laser was not used in this study and the AMS operated on a 2 min time resolution. The detection limit of sulfate in the AMS was calculated to be 10 and 12 ng m3 for the 2014
www.atmos-chem-phys.net/17/1/2017/ Atmos. Chem. Phys., 17, 120, 2017
M. R. Giordano et al.: A missing source of aerosols in Antarctica 3
DMS have been conducted over inland Antarctica (Concordia Station), but without real-time aerosol measurements, it is impossible to determine the actual impact of marine biota on Antarctic sulfate (Preunkert et al., 2008).
Beyond sulfate and chemical composition information in general, knowledge of aerosol physical properties is necessary to constrain their climate effects. For example, quantifying the ability of aerosols to form cloud droplets is one of the key challenges in determining the overall climate effects of aerosols. This could be of particular importance in the Antarctic where cloud formation may be limited by the low concentration of aerosols available to act as cloud condensation nuclei (CCN). Determining the CCN spectrum of a given aerosol population is possible once the size distribution, size-resolved composition, and mixing state of the aerosol population is known (Petters and Kreidenweis, 2007; Wang et al., 2010). Mixing state is one of the more difcult to measure properties of the aerosol spectrum. The extremes of mixing state are termed internal mixtures (all particles in a population, regardless of size, have identical chemical compositions) or external mixtures (there are multiple particle populations with distinct and differing compositions) though most real aerosol populations are somewhere in between (Textor et al., 2006). The mixing state of an aerosol population affects cloud forming predictions when the aerosol population is comprised of components of signicantly differing hygroscopicities (Wex et al., 2010). Because Antarctic aerosols seem to primarily be composed of sulfates and salts, the effect of the mixing state on cloud forming predictions may be minimized over the continent itself but overestimated as continental air masses ow out over the Southern Ocean and gain organic components.
The recent eld campaign 2ODIAC 2 Season Ozone Depletion and Interaction with Aerosols Campaign deployed a set of instruments that can begin to answer some of the outstanding questions about Antarctic aerosols, in particular, the questions surrounding high time-resolution chemical speciation to determine the sources of the aerosol mass population.This is the rst paper from the campaign and focuses on the sources of Antarctic (coastal) aerosols and the physical properties of the sulfate aerosol mass population.
2 Methods
2.1 Field site
The 2ODIAC campaign took place over 2 years with measurements occurring during the austral springsummer of 2014 (OctoberDecember) and the austral winterspring of 2015 (AugustOctober). A eld site was set up on the sea ice in McMurdo Sound approximately 20 km from Mc-Murdo Station, Antarctica. The 2014 eld site was located at 77 41[prime]40[prime][prime] S, 166 11[prime]58[prime][prime] E, while the 2015 site was 5 km away at 77 42[prime] 58[prime][prime] S, 166 24[prime]30[prime][prime] E. The dynamic nature of
4 M. R. Giordano et al.: A missing source of aerosols in Antarctica
and 2015 deployments, respectively (Jayne et al., 2000). The inlet for the aerosol sampling line was covered and heated to prevent sampling of windblown snow and to prevent riming, respectively. At the ow conditions and geometry of the sampling inlet, transmission of < 1 m particles to the AMS was > 95 %. Overall, the aerosol inlet system had a 50 % transmission efciency at 5 m and 0 % transmission efciency of particles > 9 m. All transmission efciency values are as calculated by a particle loss calculator using the specic geometry of the setup (von der Weiden et al., 2009). This work focuses primarily on results from the aerosol size, concentration, and composition instruments. Future work will discuss results from other parts of the instrumentation suite, including the particle lters and snow samples.
Meteorological data were recorded by a colocated weather station (Davis Vantage Pro2). The anemometer (rated 0 200 km h1), temperature and relative humidity probes (3 C and 3 % accuracy, respectively), and solar radiation sensors (spectral response 4001100 nm, 1 Wm2 resolution, [notdef]5 %
accuracy at full scale) were mounted on a pole 50 m NE of the sampling hut. It should be noted that, due to the sensitivity and accuracy of the radiometers, solar irradiance is presented in this paper as an approximate value. These values are useful to separate different temporal periods in the data but should not be taken as indicative of actinic ux values.Data from the sensors was wirelessly transmitted to the control unit located in the instrumentation hut.
During both eld seasons, all of the instruments were routinely calibrated and, where applicable, veried against each other. For 1 h each day, at different times every day, the inlet was switched to a HEPA lter for all of the particle instruments. This daily period allowed for a background signal to be calculated for the AMS and ensured no part of the inlet system was leaking (via monitoring of the particle counters/sizers). In depth calibrations and checks were conducted weekly. For the AMS, the weekly calibration included an ionization efciency calibration using ammonium nitrate (atomized and size selected via the SEMS or SMPS at 300 nm) and a particle time-of-ight (PToF) size calibration using PSL (polystyrene latex; 60, 100, 300, 600, and 800 nm). Across both eld seasons, the AMS calibrations changed by less than 10 % while in the eld. To check the particle counters against each other, ammonium nitrate was size selected by the SEMS or SMPS and the output measured by the associated CPCs and EPC. Weekly zeros and spans were also conducted for the gas analyzers.
3 Results and discussion
Figure 1 shows the records for wind speed, wind direction, number concentration from the EPC, and sulfate concentration from the AMS for both 2ODIAC eld seasons. The wind speed, wind direction, and particle number concentration traces were collected every second but have been aver-
aged to 2 min records to match the AMS data recording rate.
The time series is displayed with the 2015 eld season as the leftmost part of the x axis to emphasize inter-seasonal transitions from winter to spring (2015) and spring to summer (2014).
Figure 1a shows the wind direction (degrees) as a function of time and colored as a function of wind speed (m s1), for both seasons. The 2015 eld season (winterspring) was dominated by winds coming from the ESE with high wind speeds. Over 80 % of the entire 2015 eld season had wind coming from the ESE. Additionally, more than 60 % of the 2015 eld season had wind speeds recorded at over 8 m s1.
By contrast, the 2014 eld season (springsummer) wind fetch had a more bimodal wind direction distribution and an opposite wind speed probability distribution as compared to 2015. In 2014, the wind direction distribution was 60/40 %ESE/NW and was above 8 m s1 for only 20 % of the season. These meteorological patterns and seasonal differences are not unusual for this region (Seefeldt et al., 2003).
Figure 1b shows the number concentration from the EPC over both eld seasons. The gure shows the 2 min average as well as a 1 h average. With total condensation nuclei (CN) number concentrations ranging between 501000 cm3, 2ODIACs observations are consistent with other coastal and interior Antarctic eld measurements (e.g., Jaenicke et al., 1992; Gras, 1993; Koponen, et al., 2003; Belosi et al., 2012). Figure 1b shows a major facet of Antarctic aerosols: that there is a steady-state aerosol concentration during calm and low-wind periods. The steady-state concentrations here are dened as the concentrations at which 99 % of the recorded data is over. During the 2015 eld season aerosol number concentrations were at a minimum of near 50 cm3 at the start of the campaign (early September). As the season progressed and the sun began to rise in the austral spring, the background concentration of aerosols rose to 125 cm3. The 2014 eld season experienced a similar minimum of approx. 75 cm3 but did not see the same rise in the background aerosol number population. Dotted lines at these values are included on Fig. 1b to aid the eye. A lack of combustion-derived organic aerosol signal in the AMS and the persistence of these particles independent of wind direction suggests that the background aerosol is neither a result of local pollution (e.g., our own generators or vehicles) nor transported pollution from McMurdo.
Figure 1c shows the aerosol sulfate concentration measured by the AMS over the course of both eld seasons.Aerosol sulfate positively tracks with the total aerosol counts both in the wind speed dependence and the trends in background concentrations. In 2015, total aerosol sulfate begins the season (in early September) at 20 ng m3 and rises to approx. 40 ng m3 as solar irradiance begins increasing at the end of September. In 2014, total aerosol sulfate increases from approx. 30 ng m3 (in October) to 60 ng m3 (in December) over the course of the campaign. The trend of an increasing sulfate aerosol burden in springtime Antarctica is
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
M. R. Giordano et al.: A missing source of aerosols in Antarctica 5
(a)
(b)
(c)
Figure 1. For both the 2014 and 2015 eld seasons, with 2015 on the left: (a) wind direction record colored as a function of wind speed, displayed as a 2 min average record (standard deviation 0.2 m s1 and 2 ); (b) 2 min (light blue; standard deviation of 11 and 18 cm3 for 2015 and 2014, respectively) and 1 h (black) records of particle number concentration from the EPC; (c) 2 min records of sulfate concentration from the aerosol mass spectrometer. Dotted lines in (b) indicate the minimums in particle number concentrations (99th percentile) measured over the eld seasons.
well documented. The values reported here are 23 times less than previous aerosol sulfate measurements in coastalAntarctica (Minikin et al., 1998; Rankin and Wolff, 2003).However, the previous measurements are lter-integrated assessments from Antarctic bases other than McMurdo (e.g.,Halley, Neumayer, Dumont dUrville) with differing size ranges and sampling techniques. Without coincident measurements across the continent, it is difcult to determine if spatial or temporal factors are responsible for the differences in concentrations between literature values of aerosol sulfate and the presented values.
While aerosol sulfate is the main focus of this paper, it is not the only aerosol component and the relative amount of sulfate measured by the AMS should be contextualized.Over both eld seasons, sulfate generally makes up more than 50 % of the total mass of the traditionally reported non-refractory species (organics, sulfate, nitrate, and ammonium). Both the absolute amount and relative percentage of total non-refractory mass of sulfate are higher in 2014 than 2015. Ammonium, organics, and nitrate, in that order, make up the rest of the non-refractory species measured by theAMS. When adding measurements of refractory Na and Cl to the non-refractory species (as discussed in a forthcoming paper, Giordano et al., 2016), sulfate is the third most abundant species at 530 % of the total submicron aerosol mass.
3.1 Sulfate aerosols as an external mixture
Taken together, the fact that neither total aerosol counts nor aerosol sulfate fall below a minimum value suggests that the background Antarctic aerosol number population may be primarily composed of sulfate. The relatively constant mass concentration of sulfate, independent of wind speed, wind direction, and air mass origin, would indicate that this background aerosol is relatively temporally and geographically invariant. Other chemical species showed strong dependence on meteorological conditions in contrast to sulfate. If the background aerosol is primarily composed of sulfate species, then any enhancements in total aerosol counts would have to be a separate aerosol number population. An independent, externally mixed sulfate mode would therefore be expected in the AMS PToF mode. The PToF mode in the AMS is a mode of operation in which the size-resolved aerosol composition is determined by averaging many particles during thePToF operation time (2 min in this study). The PToF mode does not represent single particles but rather the size distribution of particles in which ions, such as sulfate ions, are present. A large-scale, externally mixed sulfate mode would be expected to maintain a consistent size, during variations in wind speed and direction. In both 2014 and 2015, the (inorganic) sulfate species in the AMS exhibited such a wind-independent mass distribution. Figure 2 shows the sulfate
www.atmos-chem-phys.net/17/1/2017/ Atmos. Chem. Phys., 17, 120, 2017
Wind direction
(deg)
25
20
15
10
5
0
Wind speed (m s )
-1
Particle number
concentration (no. cm-3 )
1000 800 600 400 200
0
100
80 60 40 20
0
Sulfate concentration
(ng m-3 )
300
200
100
0
9/11 9/21 10/1 10/11Austral spring (2015)
10/31 11/10 11/20 11/30Austral summer (2014)
6 M. R. Giordano et al.: A missing source of aerosols in Antarctica
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
ESE, low wind
ESE, med. wind
ESE, high wind
NW, low wind
NW, med. wind
NW, high wind
-3 )
3 cm
Arbitrary units
dV / dlogD p (um
2 3 4 5 6 7 8 9
2 3 4 5 6 7 8 9
1000
2
5 6 7 8 9
100
2 3 4 5 6 7
100 PToF size (nm)
Volume equivalent diameter (nm)
Figure 3. Average dV / dlogDp as a function of volume equivalent diameter for low and medium wind speeds of 2014. Results of the particle time of ight of the sulfate species from the AMS (red) are shown with the SEMS (black) and UHSAS (blue).
Figure 2. AMS PToF results of AMS sulfate for 2014 as functions of wind speed (solid lines (> 8 m s1) vs. dotted lines (28 m s1 and < 2 m s1)) and wind direction (black (northwest) vs. red (east
southeast)). The PToF mass has been arbitrarily scaled to enhance readability of the gure. The PToF results for 2015 are identical in both distribution shape and peak location for all wind regimes.
vacuum aerodynamic diameter distributions from the AMS for the 2014 eld season as a function of wind speed and direction. Regardless of where an air mass originated, the sulfate aerosol as measured by the AMS exhibited a well-distributed mode centered at a vacuum aerodynamic diameter of approximately 250 nm. This result is consistent with previous ocean-based size measurements of subantarctic MSA-containing aerosol (250 and 370 nm from Zorn et al., 2008, and Schmale et al., 2013, respectively) as well as off-line lter-integrated measurements in coastal Antarctica (200 350 nm aerodynamic diameter impactor stage; Jourdain and Legrand, 2001). None of the other species measured in the AMS showed a well-dened size mode.
Similar to the AMS, the size distributions from the other aerosol sizing instruments were consistent across wind direction regimes as long as conditions conducive to blowing snow events (here dened as wind speeds greater than 8 m s1) are excluded. Figure 3 shows the average size distributions from the UHSAS, SEMS, and AMS from the 2014 eld season when blowing snow events are excluded. For Fig. 3, all three measurements have been converted to a volume equivalent diameter to allow intercomparisons of the three different methods of aerosol sizing by applying the following assumptions: (1) a density of 1.8 g cm3 for the sul-fate aerosol population; (2) sphericity of all aerosols in the population; (3) equality of the optical diameter from the UHSAS and volume equivalent diameter (DeCarlo et al., 2004). As evidenced by the sizing instruments, the sulfate distribution is well encompassed by both the UHSAS and SEMS.
100x10-3
80
60
40
20
0
For 2015 (not shown), the result is the same with the SMPS distribution compared to the AMS PToF. Without the size-resolved sulfate distribution from the AMS, it would not be possible to identify the externally mixed mode that is present.
The presence of an external mixture has implications both for estimating the bulk direct radiative forcing of Antarctic aerosol and for predicting CCN number concentrations (NCCN) in the Antarctic troposphere. Many atmospheric radiation and global climate models assume external mixtures of aerosols (Koch et al., 2006; Bauer et al., 2007). Incorrectly assuming an externally mixed aerosol can result in incorrectly estimating radiative forcing by a factor of 3 or more (Bauer et al., 2007; Kim et al., 2008). Recent improvements to radiative forcing models have focused on incorporating internally mixed aerosol populations (Boucher et al., 2013).The results presented here, although limited in seasonal coverage and duration of sampling, suggest that radiative forcing models for Antarctica should continue to treat the sulfate mass population as an external mixture. This work does support the assumptions of older estimates of radiative forcing for sulfate aerosols over Antarctica of approx. 0.1 Wm2
(Myhre et al., 1998). It should be noted that the appearance of sulfate primarily at the lower end of the size distributions of Antarctic aerosol does not preclude the presence of sul-fate in larger or smaller particles. The aerodynamic lens of the AMS inlet system precludes measurements of particles larger than 1 .
With regard to NCCN, though size-resolved composition has been used and is preferable, bulk composition data are often used to predict CCN number concentrations (Zhang et
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
M. R. Giordano et al.: A missing source of aerosols in Antarctica 7
al., 2007). A simple volume-weighted mixing rule such as the ZdanovskiiStokesRobinson relationship (ZSR, Stokes and Robinson, 1966) or the equivalent Kappa formulation (Petters and Kreidenweis, 2007) is used to determine particle hygroscopicity. If non-size-resolved data from the AMS are used, the mixing rule implicitly assumes an internally mixed aerosol mass population and can bias calculated hygroscopicity high or low. In general, internal mixtures dominate aerosol populations as the distance from the emission source increases, and the mixing rule is an appropriate assumption, regardless of the individual emission sources (e.g., Juranyi et al., 2010; Garbariene et al., 2012). However, when close to emission sources (especially anthropogenic sources), external mixtures are generally observed (Kander and Schultz, 2007; Swietlicki et al., 2008; Wex et al., 2010). The results here, however, are an externally mixed aerosol number population. The results presented here suggest that the Antarctic aerosol population may be a special case: despite a large distance between open water (presumably the major source of breaking waves near Antarctica) and the coast, an externally mixed aerosol population is persistent over the continent.Generally, there exists in the atmosphere enough mass to condense on pre-existing particles, which reduces the impact of external mixtures downwind of the sources. Alternatively, these results may suggest that open water is unnecessary for large oceanatmosphere aerosol uxes and that sea ice or other sources could be a dominant source of coastal Antarctic aerosol number population. The results presented here are limited by the fact that 2ODIAC occurred only on the sea ice, but the consistency of these results in context of the literature suggests that, at the least, they apply over a wider area of coastal Antarctica. It should be noted that previous studies in Antarctica that have measured NCCN have reported externally mixed CCN populations. Unfortunately, previous work has been conducted at Palmer Station or around the Antarctic Peninsula areas which have a strong and consistent marine inuence (DeFelice, 1996; ODowd et al., 1997). This previous work has been unable to remove the marine inuence from the continental air masses. An externally mixed sulfate mode, as this work demonstrates, appears to be a unique and ubiquitous feature of Antarctic aerosols. Calculating NCCN is not within the scope of this paper but will be presented in future publications on 2ODIAC.
3.2 Sulfate as a predominant component of the background Antarctic aerosol number population
If the externally mixed sulfate mode observed in the AMS is, in fact, the primary component of the background aerosol concentration, then some degree of closure between the EPC number concentration and the AMS PToF data should be possible. Again assuming particle sphericity and an average material density of 1.8 g cm3 for sulfate (an average value between sulfuric acid and ammonium sulfate), the total number of sulfate particles can be calculated from the size-
resolved sulfate mass distribution in the AMS PToF mode measured over the entirety of the campaign. This process assumes sphericity of particles, takes the mass distribution measured in vacuum aerodynamic diameter space, and using the density value transforms it to volume vs. volume equivalent diameter for that sulfate mode, which can then be converted into a number distribution and integrated to nd the total number of particles from the measured sulfate mass distribution. Figure 4 shows the ratio of the total number of sulfate particles calculated from the measured AMS sulfate size distribution (average PToF size distribution scaled to the high-resolution sulfate mass concentrations) to the total number of particles measured in the EPC. This number distribution calculation is assuming no counter ion to the measured sulfate,i.e., the sulfate is present as sulfuric acid, which is consistent with previous Southern Ocean data (Zorn et al., 2008). As in Fig. 1a, the 2014 and 2015 eld seasons have been combined and the data trace is colored by wind speed in meters per second.
As discussed in the forthcoming Kalnajs et al. (2016) publication, higher wind speeds increase the non-sulfate aerosol counts. In discussing the background aerosol number population, any occurrence of blowing snow (implemented in this paper roughly as wind speeds above 8 m s1, as suggested by Li and Pomeroy, 1997) should therefore be ignored. For Fig. 4, this roughly translates to ignoring non-blue/green sections of the record. The total record is produced here for completeness, but Fig. 4 also includes a smoothed trace (boxcar smoothing) that excludes high wind speed events (since high wind speeds and particle counts are strongly correlated, to be discussed in a forthcoming paper, Giordano et al., 2016).
Sulfate modes can be described as three distinct phases as shown in Fig. 4 (indicated by numbers and vertical lines).First, in the late austral winterearly spring, sulfate comprises the majority of the total aerosol counts often over 60 % (1). Second, the percent of the total aerosol number population that is made up of sulfate, then takes a sharp decline (2). Third, the percent sulfate gradually climbs again as the austral spring transitions to summer (3). The spring summer increase is primarily evident in the 2014 data. While the fraction of particles that are primarily sulfate has a distinct cycle during these three phases, it should be emphasized that the sulfate mass is steadily increasing from winter to summer (Fig. 1c).
Phases 1 and 3 are both in line with previous studies performed in and around Antarctica. Phase 1 is indicative of the low aerosol loadings and the persistent aerosol sulfate component (total, i.e., not nss) seen multiple times over the continent in the winter (e.g., Gras et al., 1985; Wagenbach et al., 1988; Savoie et al., 1993). Phase 3 is likewise a measurement of the well-known springtime buildup of aerosol sulfate mass in and around Antarctica. Both direct measurements and model results have suggested that the sulfate buildup is due to phytoplankton activity in the Southern Ocean (e.g., Minikin et al., 1998, Khan et al., 2015). Phase 2, however,
www.atmos-chem-phys.net/17/1/2017/ Atmos. Chem. Phys., 17, 120, 2017
8 M. R. Giordano et al.: A missing source of aerosols in Antarctica
1.0
0.8
0.6
0.4
0.2
0.0
Figure 4. Sulfate number ratio as calculated from the AMS PToF mode divided by the total number concentration from the EPC over both 2014 (right) and 2015 (left). Both eld seasons data are colored by recorded wind speeds. A smoothed trace of only wind speeds < 8 m s1 is overlaid in black. Rough timing for the phases discussed in Sect. 3.2 are noted.
has not previously been identied as a major aspect of the annual evolution of the Antarctic aerosol number population.We note that this transitional season has only been observed in this dataset, and future observational datasets measured during this period are important for understanding the consistency of this period.
Phase 2 is consistent with measuring newly formed particles that have been transported to our measurement location during a transitional period during the extended Antarctic sunrise. Throughout this paper, all instances of new particle formation refer to a regional, not local, growth of previously unobserved aerosol mass unless otherwise noted. The particles that are formed during this transitional period make up a signicant fraction of the total aerosol number concentration at the very start of the austral spring. The new aerosol populations contribution to the total Antarctic number population, however, decreases as biogenic emissions in the Southern Ocean increase.
The exact cause of Phase 2 is difcult to determine from the instrumentation deployed during 2ODIAC, but potential explanations can be narrowed down. From Fig. 1 duringPhase 2, both the total counts on the EPC and the sulfate mass in the AMS trend upward. However, total counts increases faster than the mass captured in the AMS. Three possible explanations therefore exist for the trends seen in Fig. 4 duringPhase 2:
(1)
(2)
(3)
Sulfate particle number ratio
20
15
10
5
0
Wind speed (m s )
-1
9/11 9/16 9/21 9/26 10/1 10/6
Austral spring (2015)
10/31 11/5 11/10 11/15 11/20 11/25 11/30
Austral summer (2014)
1. The particles are refractory and would not be measured by the AMS regardless of size,
2. the particles are being counted in the AMS but are not sulfate (e.g., organics, nitrate), or
3. there are particles measured by the EPC that are either not producing a measurable size distribution signal or outside of the AMS measurement range at either the low (less than 40 nm) or the high end (greater than 1 ).
These three explanations for the transitional aerosol number population observed in Phase 2 must be examined before any conclusions on the source(s), impacts, or fate of the new number population can be drawn. A thorough discussion of these three possibilities is found in Appendix B, but the main points are discussed here.
Explanation 1 (refractory particles) is unlikely to account for the transitional-period aerosol due to the fact that the main source of refractory aerosols would be sea spray. Enhancements in sea spray would also cause enhancements in the mass and volume loadings due to the large contribution of sea spray to the super-micron particle distribution. None of these enhancements were observed in other instrumentation running during 2ODIAC. Explanation 2 (the transitional aerosol being non-sulfate) is unlikely since the ratio of sul-fate to total non-refractory mass is steady when wind speed is accounted for (Fig. A1 in Appendix). Explanation 3 (the transitional aerosol being a size not measurable by the AMS)
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
with solar irradiance levels at 300 Wm2 (Saiz-Lopez et
al., 2007). Solar photooxidation of frozen iodine-containing solutions has been shown to increase gas-phase iodine concentrations as well (Kim et al., 2016). Halogens have long been linked to ozone depletion events (ODEs), and ODEs are linked to aerosol enhancements (e.g., Kalnajs et al., 2013).Unfortunately, the data are still unclear on whether the spatially and temporally invariant aerosol increase is due to ODEs as they have been classically observed in the polar regions. Another possible reservoir species in the Antarctic is mercury. Recent work has suggested that mercury can participate in new particle formation in the Antarctic atmosphere (Humphries et al., 2015). If reservoir species exist, then questions beyond that of their origin begin to arise. The mass of the reservoir, rates of source depletion and aerosol production, and aerosol formation mechanisms are all questions that would require additional measurements to answer.
Regardless of whether the non-sulfate aerosol is related to a mechanism such as ODEs or the existence of a reservoir species, two observations are clear: rst, a (likely) sub-40 nm mode of aerosols is somehow produced over coastal Antarctica in late winterearly spring, and, second, the importance of the sub-40 nm mode is reduced as sulfate mass in the larger aerosol modes increases in the late spring and early summer.
3.3 The source of Antarctic sulfate
Since the externally mixed sulfate mode observed in 2ODIAC makes up such a dominant fraction of the total non-refractory aerosol number population, it is important to determine the source of the sulfate aerosol. As discussed, various other investigators have apportioned aerosol sulfate to either sea-spray or non-sea-spray sources usually through the use of sodium as a sea-spray marker. In addition to the problems of using sodium due to mirabilite fractionation, sodium is generally a refractory species in the AMS and not well captured at the standard operating conditions of the instrument.Sodium can be well measured by the ion chromatography of lters, but the record produced has long time integrations (24 h in the case of 2ODIAC) and does not separate out wind and aerosol effects. Fortunately, the AMS is able to measure carbonsulfur compounds in real time, which have previously been tied to MSA concentrations in the marine boundary layer (Phinney et al., 2006; Zorn et al., 2008; Schmale et al., 2013).
MSA (CH3HSO3) is observed in the AMS as a marker ion at the mass-to-charge ratio (m/z) of 78.99 (CH3SO+2).
Unfortunately, due to fragmentation of the molecule on the vaporizer and in the ionization region of the AMS, the signal of m/z 78.99 alone is not sufcient to determine the total MSA concentration (see MSA average mass spectrum, Fig. S1 in the Supplement or Zorn et al., 2008). The total MSA signal can be reconstructed from the total aerosol signal recorded in the eld, however, by using the fragmentation pattern of MSA aerosolized directly into the AMS (Zorn et
www.atmos-chem-phys.net/17/1/2017/ Atmos. Chem. Phys., 17, 120, 2017
M. R. Giordano et al.: A missing source of aerosols in Antarctica 9
is impossible to rule out. The reader is again referred to Appendix B for a complete discussion, but the measurements from the sizing instruments (SMPS and SEMS) and AMS suggest that the transitional aerosol is equal to or less than 250 nm in (vacuum aerodynamic) diameter. Unfortunately, given the instrumentation deployed, it is impossible to determine the composition of these very small particles (e.g., less than 100 nm) due in part to the low mass loadings inherent in Antarctica and the effects that low mass loadings have on AMS PToF data.
The potential sources of the transitional aerosol population
Phases 2 and 3 suggest that the photochemical processes affecting the Antarctic and Southern Ocean aerosol populations undergo at least two distinct phases. Unfortunately, determining the driving force behind either phase is not possible with the data available nor are there much data in the literature to explain the existence of phases 2 and 3. For example, though the springtime MSA-derived aerosol increase has long been measured in Antarctica, data regarding what threshold of sunlight is necessary for the sulfate increase has been lacking. Data on Southern Ocean chlorophyll concentrations, light requirements for phytoplankton photosynthesis, and even solar irradiance measurements during Antarctic aerosol campaigns are sparse. During 2ODIAC, however, solar irradiance at the eld site was measured for both eld seasons though the values presented here should, again, be seen as approximate and indicative of a transitional period in the aerosol number population. The percentage of sulfate particles declines to a minimum of 20 % when solar irradiance averages between 100 and 200 Wm2 (i.e., most of the austral spring after the sun had (partly) risen over McMurdo Sound).In the late springearly summer transition, where daily peak solar irradiance averages well over 300 Wm2, the percentage of sulfate aerosol begins to increase again.
Taken in conjunction with the sulfate data presented in Fig. 1 and the < 40 nm/total data in the Supplement, Fig. 4 shows that during some transitionary periods, measured during 2ODIAC as being between 100 and 300 Wm2, newly formed particles (and transport to the measurement site) may generate additional particles in the Antarctic number population. This Phase 2 to Phase 3 evolution of Antarctic aerosol has not previously been observed. Given the particle composition measurements, it is possible that signicant contributions from non-sulfate (e.g., non-DMS related) aerosol formation mechanisms contribute to the Phase 2 aerosol. A likely explanation for the source of the new particles is the existence of some reservoir species that is photoactive during the early austral spring. One likely reservoir species candidate for aerosol enhancement are halogens. Much work has recently been performed examining the importance of halogens and solar irradiance in coastal Antarctica. Gas-phase concentrations of IO and BrO have been shown to track well
10 M. R. Giordano et al.: A missing source of aerosols in Antarctica
80x10-3
60
40
20
0
Figure 5. MSA vs. sulfate as measured by the AMS for both eld seasons, colored by the daily high solar irradiance. 2ODIAC data are presented with results from Zorn et al. (2008) and Schmale et al. (2013) for context. Inset is a zoomed in area showing the 2ODIAC data in higher detail.
al., 2008). Figure 5 shows the reconstructed MSA concentration vs. the total measured sulfate signal from the AMS, plotted for both eld seasons. Also included in Fig. 5 are the data from two other eld campaigns that measured Antarctic air masses with an AMS. The rst is a reconstruction of total aerosol MSA (using an MSA fragmentation pattern) from aSouthern Ocean cruise, adapted from Zorn et al. (2008). The second is an MSA-attributed factor reconstruction (via positive matrix factorization, see below) from a stationary deployment of an AMS to Bird Island, adapted from Schmale et al. (2013). The 2ODIAC data in Fig. 5 are colored as a function of the daily high solar irradiance measured at the eld sites. As the sun reaches its approximate zenith, the MSA concentration in the AMS increases to almost 10 % of the total sulfate signal in the AMS. During the late austral winter and into the early austral spring, the MSA fragment makes up less than 1 % of the total sulfate. The absolute concentrations of MSA and sulfate of 2ODIAC are lower than those previously reported by both Zorn et al. (2008) and Schmale et al. (2013) though the ratios of MSA to sulfate are similar across the three studies. Both of the previous measurements took place in regions where the origin of particulate sulfate was dominated by open-ocean source regions and took place in the austral summer and fall where more active DMS chemistry is likely to occur. Lower MSA and sulfate concentrations during 2ODIAC are therefore not surprising given the differences in season and location as compared to the previous studies. Higher MSA concentrations are likely to appear over McMurdo Sound as the sea ice recedes around the continent in the austral summer (Jourdain and Legrand, 2001).
600
500
400
300
200
100
0
Daily high solar irradiation (W m)
-2
Zorn et al.
Schmale et al.
This study
MSA
MSA
5x10-3
4 3 2 1 0
0 10 20 30 40 50 60x10-3
SO4
0.0
0.6
0.1
0.2
0.3
SO4 (g m )
0.4
0.5
-3
Positive matrix factorization
One advantage of the AMS over gas or ion chromatography in source apportionment is the ability to capture a full mass-spectral signature of an aerosol mass population. By assuming that the full spectrum is a combination of discreet components, the component spectra and mass loading of various sources can be determined using positive matrix factorization (PMF; Paatero and Tapper, 1994; Lanz et al., 2007). PMF application to AMS data has been described in detail elsewhere (Ulbrich et al., 2009; Ng et al., 2011; Zhang et al., 2011).For the results from 2ODIAC, PMF was applied to the high-resolution data of the AMS. PMF analysis uses as input any number of mass spectra as a function of time, measured in theAMS, to pick out patterns in both temporal and mass-spectral phases of the data. PMF determines factors that contribute to the time series and mass-spectral series by minimizing a quality of t parameter (the sum of the residual of a given input not t by the algorithm and scaled by estimated error in the time and mass-spectral series). The PMF algorithm is run over multiple starting points in the rotations of the matrices and limited to given numbers of factors. Q/Qexp is used as a metric to determine how well the error model assumed inPMF is represented by the dataset. In general, the lowest values of Q/Qexp are the best solution. This metric is described in detail in Ulbrich et al. (2009). All of the measured sulfur-containing m/z ratios (excluding isotopes) were used as input to PMF to determine the sources of sulfur in Antarctic aerosols. PMF is generally performed on the organic species measured in the AMS; using only the sulfur compounds is
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
M. R. Giordano et al.: A missing source of aerosols in Antarctica 11
Figure 6. PMF results for the sulfur-containing species observed over both eld seasons. (a) Mass concentration reconstructions in nanogram per cubic meter for each factor and (b) the mass-spectral ngerprints for each of the three factors.
a novel use of the method. PMF was run to explore solutions between 1 and 5 factors (sources of sulfur-containing aerosols) and numerous rotations of the data matrices (fpeak between 8 and 8 at steps of 0.1). The major diagnostic plots
and metrics are included in the Supplement.
Over all of the permutations of the input data, the nal chosen solution resulted in three factors. It should be noted that the numbering of the factors is preserved from the PMF solution and is not indicative of any valuation on the factors themselves. Figure 6 shows both the factor mass spectra and their time series reconstructions. The day where the daily high solar irradiance reached 300 Wm2 is also noted on the gure. Factor 3, the sulfate factor, is comprised of the traditional AMS sulfate peaks: SO+ (47.97), SO+2 (63.962),
SO+3 (79.96), HSO+3 (80.96), and H2SO+4 (97.97). The MSA marker peak (m/z 78.99) and CH2SO+2 (77.98) are split between the biogenic/MSA and aged biogenic factors 1 and 2, respectively. The remainder of the carbonsulfur complexed ions (e.g., CHS+ (44.98), CH3S+ (47.00), C2H3S+ (59.00),
CH2SO+2 (77.98), CH3SO+2 (78.99), and CH4SO+3 (95.99)) are predominantly found in factor 2. Besides CH3SO+2 and
CH2SO+2, factor 1 is primarily composed of SO+ (47.97),
HSO+ (48.97), SO+2 (63.96), HSO+2 (64.96),SO+3 (79.96),
HSO+3 (80.96), and H2SO+4 (97.97) (Fig. 6a). Factors 1 and 2 are differentiated according to both their percent contributions of the CxHySzOp ions as well as the ratios of major sulfate peaks (e.g., SO+ / SO+2).
The naming of the factors, sulfate, biogenic/MSA, and aged biogenic, for factors 3, 2, and 1, respectively, was chosen based on two main aspects of the resultant PMF spectra.The sulfate factor was named due to its lack of organosulfate complexes and the fact that it bears some resemblance to the mass spectra obtained by atomizing diluted sulfuric acid into the AMS. The biogenic and aged biogenic factors were distinguished primarily by the predominance of the CHS com-
plexes appearing in the biogenic factor while not being completely absent in the aged biogenic factor. The aged biogenic factor is so named based on the assumption that, in much the same way that aging organics eventually reduce to oxidized forms, organosulfur complexes may reduce to pure sulfate forms. If this assumption is not correct then the naming scheme of the last factor may need to be adjusted in the future. Again, the three-factor solution was chosen due to it minimizing the Q/Qexp of the PMF solution without creating obviously extraneous factors (see Fig. S4).
The sulfate factor makes up the majority, 5060 %, of the total (reconstructed) mass over both eld seasons (Fig. 6b).The aged biogenic factor (factor 2) only has an appreciable contribution (1020 %) to the total non-refractory mass at the beginning of the austral winterspring eld season in 2015 and ceases to contribute appreciable mass after 23 September, when solar irradiance regularly and consistently reaches above 100 Wm2. The aged biogenic factor has almost no inuence on the austral summer measurements. A possible source for the aged biogenic factor is long-range transport from areas of the Southern Ocean that experience more sunlight earlier in the year than the Antarctic coast. The biogenic/MSA factor, however, signicantly contributes (40 50 %) to the reconstructed mass during the austral summer and, with the exception of 25 October 2015, has a steady contribution to the early austral spring mass. Despite the minimal contribution of the aged biogenic factor, the three-factor solution was chosen over the two-factor solution for two reasons. The primary reason is the inadequacy of the two-factorPMF solution with regard to MSA, which based on previous measurements in the Southern Ocean and the presence of a marker ion (CH3SO+2 at m/z 79) should make up some of the sulfur contribution. Two-factor PMF results either apportioned m/z 79 to two factors with 48/64 ratios that did not resemble any known substance (e.g., things tested in a lab setting included ammonium sulfate, pure MSA, diluted
www.atmos-chem-phys.net/17/1/2017/ Atmos. Chem. Phys., 17, 120, 2017
12 M. R. Giordano et al.: A missing source of aerosols in Antarctica
H2SO4, and Southern Ocean sea water) or apportioned the majority of m/z 79 to a factor that was not temporally consistent with the CH3SO+2 fragment in the dataset. The secondary reason for choosing the three-factor solution is that three factors were consistently the number where diminishing returns in Q/Qexp began to occur. The attribution of the
MSA marker ion to the aged biogenic and biogenic/MSA factor indicates that both factors are likely representative of either MSA directly or of biologically inuenced aerosols.Comparison to direct atomization of MSA into the AMS (see Supplement) suggests that the biogenic factor is made up of more than just MSA contributions since PMF did not nd a pure MSA factor mass spectrum for this dataset. The ratios of CH3SO+2 to the major sulfate peaks (SO+, SO+2, HSO+3,
SO+4) in the two biogenic factors differ from pure MSA measured by the AMS in the laboratory.
Since using only the sulfur-containing ions in PMF analysis is novel, it is difcult to compare these PMF results to previously published results. The most closely related study is Schmale et al. (2013), which measured Antarctic/Southern Ocean air masses. In both the results presented here and in Schmale et al. (2013), the percent contribution of MSA to the total aerosol burden increases as sunlight (phytoplankton activity) increases over the Southern Ocean. Additionally, the MSA-associated factor in that study is postulated to contribute signicantly to the total sulfate signal, although it is not measured explicitly, which agrees with the results here.
One major conclusion evident from the PMF solution is the relative lack of carbonsulfur (CS) complexes, other than CH3SO+2, in the measured total sulfur (submicron, non-refractory) aerosol budget. The small amount of carbon
sulfur complexes present at the very end of the austral winter are an interesting observation, and the low concentration of CS complexes during the austral spring and early summer (aged biogenic, Fig. 6b) indicates that either (a) the aging and oxidation of DMS/MSA may not occur over coastal Antarctica in the early summer or (b) that the aging products of DMS/MSA complexes are present but not detectable by the AMS (see, e.g., Sect. 3.2.3).
As for MSA itself, the fact that an MSA-containing fragment increases (in both absolute and relative terms) during the austral spring and summer supports previous work that has suggested that phytoplankton-derived sulfate is a major source of Antarctic sulfate aerosol. The PMF analysis supports the hypothesis that DMS and MSA are a primary component of sulfate aerosol over Antarctica in the austral (late) spring and (early ) summer. The PMF analysis also supports the observations of Fig. 4 with a marked increase in MSA-derived signal in the AMS after a threshold of solar irradiance is observed.
Combining Figs. 4, 5, and 6, three distinct phases of the aerosol number population over Antarctica become evident.The rst is a highly aged sulfate population that makes up the majority of the total aerosol counts over the continent.
The second phase is the increase of a new particle number population that occurs in the austral spring. The third phase is the increasing importance of non-sea-spray sulfate (likely phytoplankton derived) with regard to the total aerosol number population. It remains to be seen if the trend reverses itself over the austral autumn as has been suggested in the measurement of MSA over Antarctica (Weller et al., 2011a,b).
4 Conclusions
The 2ODIAC campaign successfully deployed a suite of aerosol and gas-phase instruments, including the rst ever deployment of an AMS to Antarctica, over two eld seasons. The data from the AMS and particle counting instruments has shed new light on the aerosol mass and number population over the continent. First, a sulfate aerosol mode is shown to comprise the majority of the background aerosol population over two eld seasons. Second, the background sulfate mass population is shown to be externally mixed, with the distribution peaking at a vacuum aerodynamic diameter of 250 nm. The 250 nm mode was consistent throughout both seasons over which 2ODIAC was conducted, regardless of local meteorology. Both of these conclusions together suggest that sulfate may be a relatively temporally and geographically invariant feature of the Antarctic aerosol population. Third, the 250 nm sulfate population is shown to undergo three distinct phases with the austral seasons. In the austral winter, the 250 nm mode dominates the total aerosol number population. In the late austral winterearly spring a secondary aerosol mode of unknown composition comprises the majority of the aerosol number population. As austral spring progresses into summer, the importance of the 250 nm sulfate mode begins to recover to the austral winter levels of dominance of the aerosol number population. The austral spring summer buildup of sulfur is shown to be increasingly composed of MSA-derived aerosols, matching previous measurements over Antarctica. The unknown austral spring aerosol number population is shown to be likely comprised of particles less than 250 nm in size, possibly formed from halogen photochemistry or mercury-catalyzed new particle formation over the Southern Ocean. This work further underscores the need to closely examine new particle formation and newly formed particles over Antarctica, and the Southern Ocean, in the early austral spring.
5 Data availability
Data are available upon request to the correspondence author Peter DeCarlo ([email protected]).
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
M. R. Giordano et al.: A missing source of aerosols in Antarctica 13
Appendix A
1.0
0.8
0.6
0.4
0.2
0.0
Percent sulfate of total AMS mass
25
20
15
10
5
Wind speed (m s )
9/11 9/21 10/1 10/11
Austral spring (2015)
10/31 11/10 11/20 11/30
Austral summer (2014)
Figure A1. Mass of sulfate as a percent of total non-refractory mass measured in the AMS, colored as a function of wind speed, for both eld seasons.
1.0
0.8
0.6
0.4
0.2
0.0
Particles < 40 nm / total particles
25
20
15
10
5
0
Wind speed (m s )
-1
9/11 9/21 10/1 10/11
Austral spring (2015)
10/31 11/10 11/20 11/30
Austral summer (2014)
Figure A2. Fraction of particles less than 40 nm of the total counts from the particle sizing instruments for the 2015 (SMPS) and 2014 (SEMS) eld seasons. Both records are colored by the wind speed (m s1) records.
www.atmos-chem-phys.net/17/1/2017/ Atmos. Chem. Phys., 17, 120, 2017
14 M. R. Giordano et al.: A missing source of aerosols in Antarctica
Figure A3. dN / dlogDp image plots from the particle sizing instruments for the 2015 (a, SMPS) and 2014 (b, SEMS) eld seasons. Included are the wind speed and direction records. Note that small (2030 nm) particle enhancements only happen during calm-wind periods that coincide with elevated m/z 55 concentrations in the AMS (not shown), indicating generator contamination.
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
M. R. Giordano et al.: A missing source of aerosols in Antarctica 15
Appendix B
B1 The transitional aerosol as a refractory number population
Completely ruling out that the excess particle counts are due to refractory particles (explanation 1) is impossible with the instrumentation available during 2ODIAC. However, sea spray, the volcano Mt Erebus, and the dry valleys are the only major sources of refractory particles in the region. Since there were no eruptions or seasonally dependent change in volcanic activity during 2ODIAC, Mt Erebus can likely be ruled out as the source of the particle counts. Sea spray is more difcult to rule out, especially considering the relatively closer ice edge in the spring eld season. However, sea spray is predominantly super micron size and signicant enhancements in the super micron number distribution would lead to signicant enhancements in the mass and volume loadings. Such enhancements, however, were not observed in other instrumentation (e.g., Lighthouse OPC). Further, sea spray source strength is largely determined by wind speed (Madry et al., 2011), and the slow increase observed is unlikely to come from highly variable wind encountered during this time.
B2 The transitional aerosol as a non-sulfate number population
The possibility that the AMS is measuring the particles associated with the increased counts of the EPC but that the particles are not sulfate (explanation 2) should also be explored. If non-sulfate particles are the same size or larger than the sulfate particles, then the ratio of sulfate mass to total non-refractory mass measured in the AMS would likely exhibit opposing trends to Fig. 4. The non-sulfate particles would have to be the same size or larger than the sulfate particles or there could be no observed change in measured total non-refractory mass. In fact, the ratio of sulfate to total non-refractory mass (both as measured by high resolution and unit mass resolution in the AMS) is steady for both eld seasons when wind speeds are accounted for (Fig. A1). Therefore if the particles are of a similar size to the sulfate mode, the observed mass composition is not enough to explain the trends in Figs. 1 and 4.
B3 The transitional aerosol as a population outside the bounds of the AMS or not producing a measurable size distribution signal
The possibility of particles larger than the AMS size cutoff (1 m aerodynamic diameter, explanation 3) explaining Phase 2 is unlikely due to the inlet geometry. However, the existence of small particles, either signicantly smaller than the 250 nm mode of sulfate particles or particles smaller
than the AMS cutoff (40 nm aerodynamic diameter), explaining the trends is possible. If the particles are sulfate, and
measurable by the AMS bulk composition, it is not necessarily true that they will produce a measurable signal in the size distribution. Since the AMS is sensitive to mass and not number, small diameter particles do not produce as much signal as a large particle (as mass signal is proportional to d3).
Additionally, the use of a 2 % chopper to make the sizing measurements cuts total signal in the sizing mode by a factor of 25 compared to the mass spectra mode signal, which switches between total particle signal and background signal each for half of the measurement time. Finally, the signal in the sizing mode is spread out over multiple size bins making detection above the instrument baseline noise much more difcult in sizing mode and especially challenging in a pristine environment such as Antarctica. Consequently small particles at a very low concentration are unlikely to produce a size-resolved signal above the noise of the instrument.
Though measuring 40250 nm particle enhancements in the AMS is not possible, we can examine the number fraction of particles less than 40 nm to see if these particles contribute to the additional particle number measured by the EPC. Using the particle sizing instruments (SEMS and SMPS) counts between 7 and 40 nm, the importance of particles less than
40 nm can be examined. For a spherical particle of unit density (1.0 g cm3) or a pure sulfate particle with a density of1.8 g cm3, the 40 nm mobility diameter cutoff size would be equivalent to a vacuum aerodynamic diameter of 40 nm (unit density) or 72 nm (density 1.8 g cm3). If the ratio of the total number of particles below 40 nm to the total number of particles increases, then more small particles would be counted by the EPC but not by the AMS. However, during both eld seasons, the ratio of < 40 nm to total particles is steady. In 2014, the ratio of < 40 nm to total counts from the SEMS averages at 0.23 [notdef] 0.003 (condence interval: 0.05; SD: 0.15). In
2015, the ratio of < 40 nm to total counts from the SMPS averages at 0.16 [notdef] 0.002 (condence interval: 0.05; SD: 0.09).
In fact, neither eld season exhibits a strong dependence on wind speed (Fig. A2). Removing the high-wind events has negligible impacts on both the averages and standard deviations. This again suggests that the background sulfate mass population is relatively temporally and geographically invariant.
However, the possibility of correlation between the ratio of the total counts to the AMS sulfate counts (Fig. 4) and the < 40 nm ratio (Fig. A2) should also be examined as it suggests particles are being measured by the AMS but are possibly not sulfate. A Pearson correlation value of 0.4 ex
ists between the two ratios for the late austral winterearly austral spring eld season of 2015. For the late springearly summer eld season of 2014; this value is 0.04. The 2014 lack of correlation implies that changes in the two ratios are unrelated. The 2015 dataset exhibits a slight anticorrelation suggesting that as the AMS/EPC ratio goes down, the number of sub-40 nm particles increases. Even when wind speeds above 8 m s1 are excluded from the correlation (since high wind speeds and particle counts are strongly correlated), the
www.atmos-chem-phys.net/17/1/2017/ Atmos. Chem. Phys., 17, 120, 2017
16 M. R. Giordano et al.: A missing source of aerosols in Antarctica
correlation between AMS/EPC and < 40 nm/total still stands at 0.37. Additionally, the above correlations can be calcu
lated for larger electrical mobility diameters (50 and 60 nm) to account for the unknown electrical mobility/vacuum aerodynamic cutoff of these small particles. The correlation values change less than 15 % with these higher electrical mobility cutoffs. These correlation values have two implications: rst, the change in the fraction of sulfate particles in the total aerosol number population in the summer may not be due to small particles, and, second, changes in the percentage of sul-fate in the early spring may be due in part to small (< 40 nm mobility diameter) particles.
Figure 4 therefore demonstrates that small particles, either relative to the sulfate mode or below the AMS cutoff diameter, form in the early austral spring (Phase 2). Unfortunately, given the instrumentation deployed, these particles remain of an unknown composition. No obvious new particle formation events were captured during 2ODIAC (Fig. A3), so the increase in small particles is likely due to differing source regions as the air masses that owed over the eld camp changed.
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
M. R. Giordano et al.: A missing source of aerosols in Antarctica 17
The Supplement related to this article is available online at http://dx.doi.org/10.5194/acp-17-1-2017-supplement
Web End =doi:10.5194/acp-17-1-2017-supplement .
Acknowledgements. The authors of this paper would like to thank the National Science Foundation for funding this work through grant numbers 1341628 and 1341492. Additionally, the authors extend many thanks to all of the support staff at McMurdo Station, especially Tony Buchanan; without their support none of this work would have been possible. The authors would also like to thank Terry Deshler, Andrew Slater, and Anondo Mukherjee for their direct help with measurements in the eld and Erin Frolli for her assistance in satellite image retrieval and creation. The authors would also like to thank and acknowledge Soeren Zorn and Julia Schmale for their permission to reproduce their data in Fig. 5. Any opinions, ndings, and conclusions expressed in this material are those of the authors and do not necessarily reect the views of the NSF.
Edited by: M. ShiraiwaReviewed by: two anonymous referees
References
Arimoto, R., Nottingham, A. S., Webb, J., Schloesslin, C. A., and Davis, D. D.: Non-sea salt sulfate and other aerosol constituents at the South Pole during ISCAT, Geophys. Res. Lett., 28, 3645 3648, 2001.
Bates, T. S., Lamb, B. K., Geunther, A., Dignon, J., and Stoiber, R.E.: Sulfur Emissions to the Atmosphere from Natural Sources, J. Atmos. Chem., 14, 315337, 1992.
Bauer, S. E., Mishchenko, M. I., Lacis, A. A., Zhang, S., Perlwitz, J., and Metzger, J. M.: Do sulfate and nitrate coatings on mineral dust have important effects on radiative properties and climate modeling?, J. Geophys. Res., 112, D06307, doi:http://dx.doi.org/10.1029/2005JD006977
Web End =10.1029/2005JD006977 http://dx.doi.org/10.1029/2005JD006977
Web End = , 2007.
Belosi, F., Contini, D., Donateo, A., Santachiara, G., and Prodi, F.: Aerosol size distribution at Nansen Ice Sheet Antarctica, Atmos. Res., 107, 4250, 2012.
Boucher, O., Randall, D., Artaxo, P., Bretherton, C., Feingold, G., Forster, P., Kerminen, V. M., Kondo, Y., Liao, H., Lohmann, U., Rasch, P., Satheesh, S., Sherwood, S., Stevens, B., and Zhang,X.: Clouds and aerosols, in: Climate change 2013: the physical science basis. Contribution of working group I to the fth assessment report of the intergovernmental panel on climate change, edited by: Stocker, T. F., Qin, D., Plattner, G. K., Tignor, M., Allen, S., Boschung, J., Nauels, A., Xia, Y., Bex, V., and Midgley, P., chap. 7, Cambridge Universtiy Press, Cambridge, 2013. Carslaw, K. S., Boucher, O., Spracklen, D. V., Mann, G. W., Rae,J. G. L., Woodward, S., and Kulmala, M.: A review of natural aerosol interactions and feedbacks within the Earth system, Atmos. Chem. Phys., 10, 17011737, doi:http://dx.doi.org/10.5194/acp-10-1701-2010
Web End =10.5194/acp-10-1701- http://dx.doi.org/10.5194/acp-10-1701-2010
Web End =2010 , 2010.
Charlson, R. J., Lovelock, J. E., Andreae, M. O., and Warren, S. G.: Oceanic phytoplankton, atmospheric sulphur, cloud albedo and climate, Nature, 326, 655661, 1987.
Charlson, R. J., Langner, J., and Rodhe, H.: Sulphate aerosol and climate, Nature, 348, doi:http://dx.doi.org/10.1038/348022a0
Web End =10.1038/348022a0 http://dx.doi.org/10.1038/348022a0
Web End = , 1990.
Chasteen, T. G. and Bentley, R.: Volatile Organic Sulfur Compounds of Environmental Interest: Dimethyl Sulde and Methanethiol, An Introductory Overview, J. Chem. Educ., 81, 15241528, 2004.
DeCarlo, P. F., Slowik, J. G., Worsnop, D. R., Davidovits, P., and
Jimenez, J. L.: Particle morphology and density characterization by combined mobility and aerodynamic diameter measurements, Part 1: Theory, Aerosol Sci. Technol., 38, 11851205, 2004. DeCarlo, P. F., Kimmel, J. R., Trimborn, A., Northway, M. J., Jayne,J. T., Aiken, A. C., Gonin, M., Fuhrer, K., Horvath, T., Docherty,K. S., Worsnop D. R., and Jimenez, J. L.: Field-Deployable, High-Resolution, Time-of-Flight Aerosol Mass Spectrometer, Anal. Chem., 78, 82818289, 2006.
DeFelice, T.P.: Variations in cloud condensation nuclei at palmer station Antarctica during February 1994, Atmos. Research, 41, 229248, 1996.
Fischer, H., Siggaard-Andersen, M.-L., Ruth, U., RThlisberger, R., and Wolff, E.: Glacial/interglacial changes in mineral dust and sea-salt records in polar ice cores: Sources, transport, and deposition, Rev. Geophys., 45, RG1002, doi:http://dx.doi.org/10.1029/2005RG000192
Web End =10.1029/2005RG000192 http://dx.doi.org/10.1029/2005RG000192
Web End = , 2007.
Garbariene, I., Kvietkus, K., akalys, J., Ovadnevaite, J., andeburnis D.: Biogenic and anthropogenic organic matter in aerosol over continental Europe: source characterization in the east Baltic region, J. Atmos. Chem., 69, 159174, 2012.
Gibson, J. A. E., Garrick, R. E., Burton, H. R., and McTaggart, A.R.: Dimethylsulde and the alga Phaeocystis pouchetii in antarctic coastal waters, Mar. Biol., 104, 339346, 1990.
Giordano, M. R., Kalnajs, L. E., Avery, A., Goetz, J. D., and De-
Carlo, P. F.: High-resolution, real-time measurements of the composition of Antarctic aerosols: Results from the 2ODIAC eld campaign, in preparation, 2016.
Gras, J. L.: Condensation nucleus size distribution at Mawson,
Antarctica: Microphysics and chemistry, Atmos. Environ. Pt. A, 27, 14171425, 1993.
Gras, J. L., Adriaansen, A., Butler, R., Jarvis, B., Magill, P., and
Lingen B.: Concentration and size variation of condensation nuclei at Mawson, Antarctica, J. Atmos. Chem., 3, 93106, 1985. Hamilton, D. S., Lee, L. A., Pringle, K. J., Reddington, C. L.,
Spracklen, D. V., and Carslaw, K. S.: Occurrence of pristine aerosol environments on a polluted planet, P. Natl. Acad. Sci. USA, 111, 1846618471, 2014.
Hara, K., Kikuchi, T., Furuya, K., Hayashi, M., and Fujii, Y.: Characterization of Antarctic aerosol particles using laser microprobe mass spectrometry, Environ. Sci. Technol. 30, 385391, 1996. Harwey, M. J., Fisher, G. W., Lechner, I. S., Isaac, P., Flower, N. E., and Dick, A. L.: Summertime aerosol measurements in the Ross sea region of Antarctica, Atmos. Environ. A, 25, 569580, 1991. Humphries, R. S., Schoeld, R., Keywood, M. D., Ward, J., Pierce,J. R., Gionfriddo, C. M., Tate, M. T., Krabbenhoft, D. P., Galbally, I. E., Molloy, S. B., Klekociuk, A. R., Johnston, P. V., Kreher, K., Thomas, A. J., Robinson, A. D., Harris, N. R. P., Johnson, R., and Wilson, S. R.: Boundary layer new particle formation over East Antarctic sea ice possible Hg-driven nucleation?, Atmos. Chem. Phys., 15, 1333913364, doi:http://dx.doi.org/10.5194/acp-15-13339-2015
Web End =10.5194/acp- http://dx.doi.org/10.5194/acp-15-13339-2015
Web End =15-13339-2015 , 2015.
www.atmos-chem-phys.net/17/1/2017/ Atmos. Chem. Phys., 17, 120, 2017
18 M. R. Giordano et al.: A missing source of aerosols in Antarctica
Jaenicke, R., Dreiling, V., Lehmann, E., Koutsenoguii, P. K., and Stingl, J.: Condensation nuclei at the German Antarctic station Georg von Neumayer, Tellus B, 44, 311317, 1992.
Jourdain, B. and Legrand, M.: Seasonal variations of dimethyl sul- de, dimethyl sulfoxide, sulfur dioxyde, methanesulfonate, and non-sea-salt sulfate aerosols at Dumont dUrville (December 1998July 1999), J. Geophys. Res., 106, 1439114408, 2001.
Juranyi, Z., Gysel, M., Weingartner, E., DeCarlo, P. F., Kammermann, L., and Baltensperger, U.: Measured and modelled cloud condensation nuclei number concentration at the high alpine site Jungfraujoch, Atmos. Chem. Phys., 10, 78917906, doi:http://dx.doi.org/10.5194/acp-10-7891-2010
Web End =10.5194/acp-10-7891-2010 http://dx.doi.org/10.5194/acp-10-7891-2010
Web End = , 2010.
Kalnajs, L. E., Avallone, L. M., and Toohey, D. W.: Correlated measurements of ozone and particulates in the Ross Island region, Antarctica, Geophys. Res. Lett., 40, 63196323, doi:http://dx.doi.org/10.1002/2013GL058422
Web End =10.1002/2013GL058422 http://dx.doi.org/10.1002/2013GL058422
Web End = , 2013.
Kandler, K. and Schtz, L.: Climatology of the average water-soluble volume fraction of atmospheric aerosol, Atmos. Res., 83, 7792, 2007.
Khan, M. A. H., Gillespie, S. M. P., Razis, B., Xiao, P., Davies-
Coleman, M. T., Percival, C. J., Derwent, R. G., Dyke, J. M., Ghosh, M. V., Lee, E. P. F., and Shallcross, D. E.: A modelling study of the atmospheric chemistry of DMS using the global model, STOCHEM-CRI, Atmos. Environ., 127, 6979, 2016.Kim, D., Wang, C., Ekman, A. M. L., Barth, M. C., and
Rasch, P. J.: Distribution and direct radiative forcing of carbonaceous and sulfate aerosols in an interactive size-resolving aerosolclimate model, J. Geophys. Res., 113, D16309, doi:http://dx.doi.org/10.1029/2007JD009756
Web End =10.1029/2007JD009756 http://dx.doi.org/10.1029/2007JD009756
Web End = , 2008.
Kim, K., Yabushita, A., Okumura, M., Saiz-Lopez, A., Cuevas, C.A., Blaszczak-Boxe, C. S., Min, D. W., Yoon, H., and Choi,W.: Production of Molecular Iodine and Tri-iodide in the Frozen Solution of Iodide: Implication for Polar Atmosphere, Environ.Sci. Technol., 50, 12801287, 2016.
Koch, D., Schmidt, G., and Field, C.: Sulfur, sea salt, and radionuclide aerosols in the GISS ModelE, J. Geophys. Res., 111, D06206, doi:http://dx.doi.org/10.1029/2004JD005550
Web End =10.1029/2004JD005550 http://dx.doi.org/10.1029/2004JD005550
Web End = , 2006.
Koponen, I. K., Virkkula, A., Hillamo, R., Kerminen, V., and Kulmala, M.: Number size distributions and concentrations of the continental summer aerosols in Queen Maud Land, Antarctica,J. Geophys. Res., 108, D184587, doi:http://dx.doi.org/10.1029/2003JD003614
Web End =10.1029/2003JD003614 http://dx.doi.org/10.1029/2003JD003614
Web End = , 2003.
Kulmala, M., Korhonen, P., Napari, I., Karlsson, A., Berresheim, H., and ODowd, C. D.: Aerosol formation during parforce: Ternary nucleation of H2SO4, NH3, and H2O, J. Geophys. Res.-Atmos., 107, 8111, doi:http://dx.doi.org/10.1029/2001JD000900
Web End =10.1029/2001JD000900 http://dx.doi.org/10.1029/2001JD000900
Web End = , 2002.
Lanz, V. A., Alfarra, M. R., Baltensperger, U., Buchmann, B., Hueglin, C., and Prvt, A. S. H.: Source apportionment of sub-micron organic aerosols at an urban site by factor analytical modelling of aerosol mass spectra, Atmos. Chem. Phys., 7, 1503 1522, doi:http://dx.doi.org/10.5194/acp-7-1503-2007
Web End =10.5194/acp-7-1503-2007 http://dx.doi.org/10.5194/acp-7-1503-2007
Web End = , 2007.
Lechner, I .S., Fisher, G. W., Larsen, H. R., Harwey, M.J., and Knobben, R.A.: Aerosol size distribution in the Southwest Pacic. J. Geophys. Res. 94, 1489314903, 1989.
Legrand, M. R. and Wagenbach, D.: Impact of the Cerro Hudson and Pinatubo volcanic eruptions on the Antarctic air and snow chemistry, J. Geophys. Res., 104, 15811596, 1998.
Legrand, M. R., Delmas, R. J., and Charlson, R. J.: Climate forcing implications from Vostok ice-core sulphate data, Nature, 334, 418420, doi:http://dx.doi.org/10.1038/334418a0
Web End =10.1038/334418a0 http://dx.doi.org/10.1038/334418a0
Web End = , 1988.
Lewis, E. R. and Schwartz, S. E: Sea salt aerosol production. Mechanisms, methods, measurements, and models, in American Geophysical Union 2004 Washington, DC, American Geophysical Union, 2004.
Li, L. and Pomeroy, J. W.: Estimates of Threshold Wind Speeds for
Snow Transport Using Meteorological Data, J. Appl. Meteorol., 36, 205213, 1997.
Madry, W. L., Toon, O. B., and ODowd, C. D.: Modeled optical thickness of sea-salt aerosol, J. Geophys. Res., 116, D08211, doi:http://dx.doi.org/10.1029/2010JD014691
Web End =10.1029/2010JD014691 http://dx.doi.org/10.1029/2010JD014691
Web End = , 2011.
Minikin, A., M. Legrand, J. Hall, D. Wagenbach, C. Kleefeld, E.
Wolff, E. C. Pasteur, and Ducroz, F.: Sulfur-containing species (sulfate and methanesulfonate) in coastal Antarctic aerosol and precipitation, J. Geophys. Res., 103, 1097510990, doi:http://dx.doi.org/10.1029/98JD00249
Web End =10.1029/98JD00249 http://dx.doi.org/10.1029/98JD00249
Web End = , 1998.
Mulvaney, R., Pasteur, E. C., Peel, D. A., Saltzman, E. S., and Whung, P.-Y.: The ratio of MSA to non-sea-salt sulphate in Antarctic Peninsula ice cores, Tellus B, 44, 295303, doi:http://dx.doi.org/10.1034/j.1600-0889.1992.t01-2-00007.x
Web End =10.1034/j.1600-0889.1992.t01-2-00007.x http://dx.doi.org/10.1034/j.1600-0889.1992.t01-2-00007.x
Web End = , 1992.
Myhre, G., Stordal, F., Restad, K., and Isaksen, I. S. A.: Estimation of the direct radiative forcing due to sulfate and soot aerosols, Tellus B, 50, 463477, doi:http://dx.doi.org/10.1034/j.1600-0889.1998.t01-4-00005.x
Web End =10.1034/j.1600-0889.1998.t01-4- http://dx.doi.org/10.1034/j.1600-0889.1998.t01-4-00005.x
Web End =00005.x , 1998.
Ng, N. L., Canagaratna, M. R., Jimenez, J. L., Zhang, Q., Ulbrich, I. M., and Worsnop, D. R.: Real-Time Methods for Estimating Organic Component Mass Concentrations from Aerosol Mass Spectrometer Data, Environ. Sci. Technol., 45, 910916, doi:http://dx.doi.org/10.1021/es102951k
Web End =10.1021/es102951k http://dx.doi.org/10.1021/es102951k
Web End = , 2011.
ODowd, C., Lowe, J. A., Smith, M. H., Davidson, B., Hewitt, C.N., and Harrison, R. M.: Biogenic sulphur emissions and inferred non-sea-salt-sulphate cloud condensation nuclei in and around Antarctica, J. Geophys. Res., 102, 1283912854, 1997.Onasch, T. B., Trimborn, A., Fortner, E. C., Jayne, J. T., Kok, G.L., Williams, L. R., Davidovits, P., and Worsnop D. R.: Soot Particle Aerosol Mass Spectrometer: Development, Validation, and Initial Application, Aerosol Sci. and Tech., 46, 804817, 2012.Paatero, P. and Tapper, U.: Positive matrix factorizaion
A nonnegative factor model with optimal utilization of error-estimates of data values, Environmetrics, 5, 111126, doi:http://dx.doi.org/10.1002/env.3170050203
Web End =10.1002/env.3170050203 http://dx.doi.org/10.1002/env.3170050203
Web End = , 1994.
Parungo, F., Bodhaine, B., and Bortniak, J.: Seasonal Variation in
Antarctic Aerosol, J. Aerosol Sci., 12, 491504, 1981.
Patris, N., Delmas, R. J., and Jouzel, J.: Isotopic signatures of sulfur in shallow Antarctic ice cores, J. Geophys. Res., 105, 70717078, doi:http://dx.doi.org/10.1029/1999JD900974
Web End =10.1029/1999JD900974 http://dx.doi.org/10.1029/1999JD900974
Web End = , 2000.
Petters, M. D. and Kreidenweis, S. M.: A single parameter representation of hygroscopic growth and cloud condensation nucleus activity, Atmos. Chem. Phys., 7, 19611971, doi:http://dx.doi.org/10.5194/acp-7-1961-2007
Web End =10.5194/acp-7- http://dx.doi.org/10.5194/acp-7-1961-2007
Web End =1961-2007 , 2007.
Phinney, L., Leaitch, W. R., Lohmann, U., Boudries, H., Worsnop,D. R., Jayne, J. T., Toom-Sauntry, D., Wadleigh, M., Sharma,S., and Shantz, N.: Characterization of the aerosol over the sub-arctic north east Pacic Ocean, Deep-Sea Res. Pt. II, 53, 2410 2433, 2006.
Preunkert, S., Legrand, M., Jourdain, B., Moulin, C., Belviso, S.,
Kasamatsu, N., Fukuchi, M., and Hirawake, T.: Interannual vari-
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
M. R. Giordano et al.: A missing source of aerosols in Antarctica 19
ability of dimethylsulde in air and seawater and its atmospheric oxidation by-products (methanesulfonate and sulfate) at Dumont dUrville, coastal Antarctica (19992003), J. Geophys. Res., 112, D06306, doi:http://dx.doi.org/10.1029/2006JD0075857
Web End =10.1029/2006JD0075857 http://dx.doi.org/10.1029/2006JD0075857
Web End = , 2007.
Preunkert, S., Jourdain, B., Legrand, M., Udisti, R., Becagli,S., and Cerri, O.: Seasonality of sulfur species (dimethyl sulde, sulfate, and methanesulfonate) in Antarctica: Inland versus coastal regions, J. Geophys. Res. Atmos., 113, D15302, doi:http://dx.doi.org/10.1029/2008JD009937
Web End =10.1029/2008JD009937 http://dx.doi.org/10.1029/2008JD009937
Web End = , 2008.
Prospero, J. M., Savoie, D. L., Saltzman, E. S., and Larsen, R.: Impact of oceanic sources of biogenic sulphur on sulphate aerosol concentrations at Mawson, Antarctica, Nature, 350, 221223, doi:http://dx.doi.org/10.1038/350221a0
Web End =10.1038/350221a0 http://dx.doi.org/10.1038/350221a0
Web End = , 1991.
Rankin, A. M. and Wolff, E. W.: A year-long record of size-segregated aerosol composition at Halley, Antarctica, J. Geophys. Res., 108, D244775, doi:http://dx.doi.org/10.1029/2003JD003993
Web End =10.1029/2003JD003993 http://dx.doi.org/10.1029/2003JD003993
Web End = ,2003.Rankin, A. M., Auld, V., and Wolff, E. W.: Frost ower as a source of fractionated sea salt aerosol in the polar regions, Geophys.Res. Lett., 27, 34693472, 2000.
Read, K. A., Lewis, A. C., Bauguitte, S., Rankin, A. M., Salmon, R.A., Wolff, E. W., Saiz-Lopez, A., Bloss, W. J., Heard, D. E., Lee,J. D., and Plane, J. M. C.: DMS and MSA measurements in the Antarctic Boundary Layer: impact of BrO on MSA production, Atmos. Chem. Phys., 8, 29852997, doi:http://dx.doi.org/10.5194/acp-8- 2985-2008
Web End =10.5194/acp-8- 2985- http://dx.doi.org/10.5194/acp-8- 2985-2008
Web End =2008 , 2008.
Saiz-Lopez, A., Mahajan, A. S., Salmon, R. A., Bauguitte, S. J. B.,
Jones, A. E., Roscoe, H. K., and Plane, J. M. C.: Boundary Layer Halogens in Coastal Antarctica, Science, 317, 348351, 2007.Savoie, D., Prospero, J. M., Larsen, R. J., Huang, F., Izaguirre, M.A., Huang, T., Snowdon, T. H., Custals, L., and Sanderson, C.G.: Nitrogen and sulfur species in Antarctic aerosols at Mawson, Palmer Station, and Marsh (King George Island), J. Atmos.Chem., 17, 95122, 1993.
Schmale, J., Schneider, J., Nemitz, E., Tang, Y. S., Dragosits, U., Blackall, T. D., Trathan, P. N., Phillips, G. J., Sutton, M., and Braban, C. F.: Sub-Antarctic marine aerosol: dominant contributions from biogenic sources, Atmos. Chem. Phys., 13, 8669 8694, doi:http://dx.doi.org/10.5194/acp-13-8669-2013
Web End =10.5194/acp-13-8669-2013 http://dx.doi.org/10.5194/acp-13-8669-2013
Web End = , 2013.
Seefeldt, M. W., Tripoli, G. J., and Stearns, C. R.: A high-resolution numerical simulation of the wind ow in the Ross Island region, Antarctica, Mon. Weather Rev., 131, 435458, 2003.
Shaw, G. E.: Considerations on the Origin and Properties of theAntarctic Aerosol, Rev. Geophys., 17, 19831998, 1979.
Shirsat, S. V. and Graf, H. F.: An emission inventory of sulfur from anthropogenic sources in Antarctica, Atmos. Chem. Phys., 9, 33973408, doi:http://dx.doi.org/10.5194/acp-9-3397-2009
Web End =10.5194/acp-9-3397-2009 http://dx.doi.org/10.5194/acp-9-3397-2009
Web End = , 2009.
Stokes, R. H. and Robinson, R. A.: Interactions in Aqueous Nonelectrolyte Solutions, I. Solute-Solvent Equilibria, J. Phys.
Chem., 70, 21262131, doi:http://dx.doi.org/10.1021/j100879a010
Web End =10.1021/j100879a010 http://dx.doi.org/10.1021/j100879a010
Web End = , 1966.Swietlicki, E., Hansson, H., Hmeri, K., Svenningsson, B.,
Massling, A., McFiggans, G., McMurry, P., Petj, T., Tunved,P., Gysel, M., Topping, D., Weingartner, E., Baltensperger, U., Rissler, J., Wiedensohler, A., and Kulmala, M.: Hygroscopic properties of submicrometer atmospheric aerosol particles measured with H-TDMA instruments in various environments: A review, Tellus B, 60, 432469, 2008.
Textor, C., Schulz, M., Guibert, S., Kinne, S., Balkanski, Y., Bauer,S., Berntsen, T., Berglen, T., Boucher, O., Chin, M., Dentener,F., Diehl, T., Easter, R., Feichter, H., Fillmore, D., Ghan, S.,
Ginoux, P., Gong, S., Kristjansson, J. E., Krol, M., Lauer, A., Lamarque, J. F., Liu, X., Montanaro, V., Myhre, G., Penner, J., Pitari, G., Reddy, S., Seland, O., Stier, P., Takemura, T., and Tie,X.: Analysis and quantication of the diversities of aerosol life cycles within AeroCom, Atmos. Chem. Phys., 6, 17771813, doi:http://dx.doi.org/10.5194/acp-6-1777-2006
Web End =10.5194/acp-6-1777-2006 http://dx.doi.org/10.5194/acp-6-1777-2006
Web End = , 2006.
Ulbrich, I. M., Canagaratna, M. R., Zhang, Q.,Worsnop, D. R., and
Jimenez, J. L.: Interpretation of organic components from Positive Matrix Factorization of aerosol mass spectrometric data, Atmos. Chem. Phys., 9, 28912918, doi:http://dx.doi.org/10.5194/acp-9-2891-2009
Web End =10.5194/acp-9-2891-2009 http://dx.doi.org/10.5194/acp-9-2891-2009
Web End = , 2009.von der Weiden, S.-L., Drewnick, F., and Borrmann, S.: Particle
Loss Calculator a new software tool for the assessment of the performance of aerosol inlet systems, Atmos. Meas. Tech., 2, 479494, doi:http://dx.doi.org/10.5194/amt-2-479-2009
Web End =10.5194/amt-2-479-2009 http://dx.doi.org/10.5194/amt-2-479-2009
Web End = , 2009.von Glasow, R. and Crutzen, P. J.: Model study of multiphase DMS oxidation with a focus on halogens, Atmos. Chem. Phys., 4, 589 608, doi:http://dx.doi.org/10.5194/acp-4-589-2004
Web End =10.5194/acp-4-589-2004 http://dx.doi.org/10.5194/acp-4-589-2004
Web End = , 2004.
Wagenbach, D.: Coastal Antarctica: Atmospheric chemical composition and atmospheric transport, in: Chemical Exchange between the Atmosphere and polar snow, edited by: Wolff, E. W. and Bales, R. C., NATO ASI Series, Springer-Verlag Berlin Heidelberg, 43, 173199, 1996.
Wagenbach, D., Gorlach, U., Moser, K., and Munnich, K. O.: Coastal Antarctic aerosol: the seasonal pattern of its chemical composition and radionuclide content, Tellus B, 40, 426436, 1988.
Wang, J., Cubison, M. J., Aiken, A. C., Jimenez, J. L., and Collins,D. R.: The importance of aerosol mixing state and size-resolved composition on CCN concentration and the variation of the importance with atmospheric aging of aerosols, Atmos. Chem. and Phys., 10, 72677283, doi:http://dx.doi.org/10.5194/acp-10-7267-2010
Web End =10.5194/acp-10-7267-2010 http://dx.doi.org/10.5194/acp-10-7267-2010
Web End = , 2010.Weller, R., Wltjen, J., Piel, C., Resenberg, R., Wagenbach, D.,
Knig-Langlo, G., and Kriews, M.: Seasonal variability of crustal and marine trace elements in the aerosol at Neumayer station, Antarctica, Tellus B, 60, 742752, 2008.
Weller, R., Minikin, A., Wagenbach, D., and Dreiling, V.: Characterization of the inter-annual, seasonal, and diurnal variations of condensation particle concentrations at Neumayer, Antarctica, Atmos. Chem. Phys., 11, 1324313257, doi:http://dx.doi.org/10.5194/acp-11-13243-2011
Web End =10.5194/acp- http://dx.doi.org/10.5194/acp-11-13243-2011
Web End =11-13243-2011 , 2011a.
Weller, R., Wagenbach, D., Legrand, M., Elssser, C., Tian-Kunze,X., and Knig-Langlo, G.: Continuous 25-yr aerosol records at coastal Antarctica Part I: inter-annual variability of ionic compounds and links to climate indices, Tellus B, 63, 901919, 2011b.
Wex, H., McFiggins, G., Henning, S., and Stratmann, F.: Inuence of the external mixing state of atmospheric aerosol on derived CCN number concentrations, Geophys. Res. Lett., 37, L10805, doi:http://dx.doi.org/10.1029/2010GL043337
Web End =10.1029/2010GL043337 http://dx.doi.org/10.1029/2010GL043337
Web End = , 2010.
Zhang, Q., Jimenez, J. L., Canagaratna, M. R., Allan, J. D., Coe,H., Ulbrich, I., Alfarra, M. R., Takami, A., Middlebrook, A.M., Sun, Y. L., Dzepina, K., Dunlea, E., Docherty, K., De-Carlo, P. F., Salcedo, D., Onasch, T., Jayne, J. T., Miyoshi,T., Shimono, A., Hatakeyama, S., Takegawa, N., Kondo, Y., Schneider, J., Drewnick, F., Borrmann, S., Weimer, S., Demerjian, K., Williams, P., Bower, K., Bahreini, R., Cottrell, L., Grifn, R. J., Rautiainen, J., Sun, J. Y., Zhang, Y. M., and Worsnop, D. R.: Ubiquity and dominance of oxygenated species
www.atmos-chem-phys.net/17/1/2017/ Atmos. Chem. Phys., 17, 120, 2017
20 M. R. Giordano et al.: A missing source of aerosols in Antarctica
in organic aerosols in anthropogenically-inuenced Northern Hemisphere midlatitudes, Geophys. Res. Lett., 34, L13801, doi:http://dx.doi.org/10.1029/2007GL029979
Web End =10.1029/2007GL029979 http://dx.doi.org/10.1029/2007GL029979
Web End = , 2007.
Zhang, Q., Jimenez, J. L., Canagaratna, M. R., Ulbrich, I. M., Ng, N. L., Worsnop, D. R., and Sun, Y. L.: Understanding atmospheric organic aerosols via factor analysis of aerosol mass spectrometry: a review, Anal. Bioanal. Chem., 401, 30453067, doi:http://dx.doi.org/10.1007/s00216-011-5355-y
Web End =10.1007/s00216-011-5355-y http://dx.doi.org/10.1007/s00216-011-5355-y
Web End = , 2011.
Zorn, S. R., Drewnick, F., Schott, M., Hoffmann, T., and Borrmann, S.: Characterization of the South Atlantic marine boundary layer aerosol using an aerodyne aerosol mass spectrometer, Atmos. Chem. Phys., 8, 47114728, doi:http://dx.doi.org/10.5194/acp-8-4711- 2008
Web End =10.5194/acp-8-4711- http://dx.doi.org/10.5194/acp-8-4711- 2008
Web End =2008 , 2008.
Atmos. Chem. Phys., 17, 120, 2017 www.atmos-chem-phys.net/17/1/2017/
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Copyright Copernicus GmbH 2017
Abstract
Understanding the sources and evolution of aerosols is crucial for constraining the impacts that aerosols have on a global scale. An unanswered question in atmospheric science is the source and evolution of the Antarctic aerosol population. Previous work over the continent has primarily utilized low temporal resolution aerosol filters to answer questions about the chemical composition of Antarctic aerosols. Bulk aerosol sampling has been useful in identifying seasonal cycles in the aerosol populations, especially in populations that have been attributed to Southern Ocean phytoplankton emissions. However, real-time, high-resolution chemical composition data are necessary to identify the mechanisms and exact timing of changes in the Antarctic aerosol. The recent 2ODIAC (2-Season Ozone Depletion and Interaction with Aerosols Campaign) field campaign saw the first ever deployment of a real-time, high-resolution aerosol mass spectrometer (SP-AMS - soot particle aerosol mass spectrometer - or AMS) to the continent. Data obtained from the AMS, and a suite of other aerosol, gas-phase, and meteorological instruments, are presented here. In particular, this paper focuses on the aerosol population over coastal Antarctica and the evolution of that population in austral spring. Results indicate that there exists a sulfate mode in Antarctica that is externally mixed with a mass mode vacuum aerodynamic diameter of 250nm. Springtime increases in sulfate aerosol are observed and attributed to biogenic sources, in agreement with previous research identifying phytoplankton activity as the source of the aerosol. Furthermore, the total Antarctic aerosol population is shown to undergo three distinct phases during the winter to summer transition. The first phase is dominated by highly aged sulfate particles comprising the majority of the aerosol mass at low wind speed. The second phase, previously unidentified, is the generation of a sub-250nm aerosol population of unknown composition. The second phase appears as a transitional phase during the extended polar sunrise. The third phase is marked by an increased importance of biogenically derived sulfate to the total aerosol population (photolysis of dimethyl sulfate and methanesulfonic acid (DMS and MSA)). The increased importance of MSA is identified both through the direct, real-time measurement of aerosol MSA and through the use of positive matrix factorization on the sulfur-containing ions in the high-resolution mass-spectral data. Given the importance of sub-250nm particles, the aforementioned second phase suggests that early austral spring is the season where new particle formation mechanisms are likely to have the largest contribution to the aerosol population in Antarctica.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer