Observations of nighttime new particle formation in the troposphere

Transcription

1 Click Here for Full Article JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 113,, doi: /2007jd009351, 2008 Observations of nighttime new particle formation in the troposphere Shan-Hu Lee, 1 Li-Hao Young, 1,2 David R. Benson, 1 Tanja Suni, 3 Markku Kulmala, 3 Heikki Junninen, 3 Teresa L. Campos, 4 David C. Rogers, 4 and Jorgen Jensen 4 Received 4 September 2007; revised 24 January 2008; accepted 8 February 2008; published 31 May [1] We present atmospheric observations which indicate efficient new particle formation during the nighttime in the troposphere under low condensation sinks, in contrast to the current prevailing assumption that aerosol nucleation takes place only during the daytime and typically from sulfuric acid. High concentrations of ultrafine particles with diameters from 4 to 9 nm (1000 cm 3 ) were measured from the three days of nighttime observations in the upper troposphere during the NSF/NCAR GV Progressive Science Missions. Long-term ground-based observations of charged and neutral clusters and aerosols made in Tumbarumba, Australia, also showed surprisingly high frequency of nighttime new particle formation (30%) with low condensation sinks. Nighttime nucleation can be significant for global aerosol load and cloud condensation nuclei productions and thus needs to be included in global climate models. Future studies are required to understand the nighttime nucleation mechanisms. Citation: Lee, S.-H., L.-H. Young, D. R. Benson, T. Suni, M. Kulmala, H. Junninen, T. L. Campos, D. C. Rogers, and J. Jensen (2008), Observations of nighttime new particle formation in the troposphere, J. Geophys. Res., 113,, doi: /2007jd Introduction [2] Despite the important effects of aerosols in global climate, atmospheric composition and human health, atmospheric aerosol formation mechanisms are poorly understood. There are high uncertainties in the current global climate models and in predictions of the future climate, largely because of our limited understanding of aerosols and clouds [International Panel on Climate Change (IPCC), 2007]. Atmospheric aerosol nucleation (formation of liquid or solid particles directly from gas phases species) [Seinfeld and Pandis, 1998] is a critical atmospheric process to understand, as aerosol particles act as cloud condensation nuclei. Current nucleation theories are also hampered by high uncertainties because they are not rigorously tested by experiments. For example, model-predicted nucleation rates often contain high uncertainties even over many orders of magnitude [Lucas and Akimoto, 2006]. [3] It is generally believed that nucleation is initiated by sulfuric acid (H 2 SO 4 ), because of its low vapor pressure (e.g., <0.001 Torr at 300 K for pure H 2 SO 4 )[Ayers et al., 1980]. H 2 SO 4 forms from SO 2 and OH, via the following 1 Department of Chemistry, Kent State University, Kent, Ohio, USA. 2 Department of Occupational Safety and Health, China Medical University, Taichung, Taiwan. 3 Department of Physical Sciences, University of Helsinki, Helsinki, Finland. 4 Earth Observing Laboratory, National Center for Atmospheric Research, Broomfield, Colorado, USA. Copyright 2008 by the American Geophysical Union /08/2007JD009351$09.00 reactions [Seinfeld and Pandis, 1998; Finlayson-Pitts and Pitts, 1999]: ðr1þ ðr2þ ðr3þ SO 2 þ OH! HSO 3 ðrate limiting stepþ HSO 3 þ O 2! SO 3 þ HO 2 SO 3 þ 2H 2 O! H 2 SO 4 þ H 2 O Because the majority of OH sources are the photochemical production of ozone and water under ultraviolet (UV) radiation, nighttime nucleation from this process (R1 R3) is not possible. Aerosol nucleation events have been observed over a wide range of atmospheric conditions [Kulmala et al., 2000, 2004; Hermann et al., 2003; Lee et al., 2003; McMurry et al., 2005], but discussions of nighttime nucleation have been typically neglected in spite of scattered evidence for their existence. Three-year measurements of aerosol size distributions on a German commercial aircraft showed frequent ultrafine particles at night in a large fraction of aerosol samples (20%) in the subtropical and midlatitude tropopause regions [Hermann et al., 2003]. Mauldin et al. [2003] observed H 2 SO 4 and ultrafine particles at night over the Pacific and suggested some unknown dark oxidation processes involving organic sulfur compounds emitted from the ocean. Observations made near and in orographic clouds also showed nighttime nucleation events, and these events were explained by unidentified multicomponent nucleation processes [Wiedensohler et al., 1997; Meters et al., 2005]. Continuous measurements of aerosol size distributions in 1of7

2 Amazonia also showed the highest concentrations of nuclei mode aerosols during the nighttime (a factor of seven higher than those in the afternoon) [Rissler et al., 2006]. Russell et al. [2007] also reported nighttime new particle formation events observed at Appledore Island, 5 nighttime events out of 24 new particle events. Frequent nighttime new particle formation events were also seen in Southern Spain [M. S. Panero, private communication, 2007]. Nocturnal growth of small clusters in the SMEAR II station in Hyytiälä, southern Finland (H. Junnine et al., Observations on nocturnal growth of atmospheric clusters, submitted to Tellus, Ser. B) and in a native Australian Eucalypt forest [Suni et al., 2008] were also recently discussed. However, these nighttime new particle formation events were usually considered as exceptions, because of the prevailing assumption that nucleation takes place via R1 in the presence of OH. There are also several long-term observations of new particle formation at the ground level [McMurry and Woo, 2002; Birmili et al., 2003; Stanier et al., 2004], but none of these studies have mentioned nighttime events. Such an absence of nighttime nucleation sharply contradicts the ubiquitous existence of ion clusters (<1.5 nm) [Kulmala and Tammet, 2007] and sub-3 nm neutral particles [Kulmala et al., 2007] observed in various atmospheric conditions. Here, we show atmospheric observations (aircraft and ground-based) that indicate new particle formation indeed occurs during the nighttime without UV under low condensation sinks. 2. Material and Methods 2.1. New Particle Measurements in the Upper Troposphere: Instrumentation [5] Studies of new particle formation in the upper troposphere used instrumentations on the NSF/NCAR GV aircraft during the Progressive Science Missions in December 2005 in Broomfield, Colorado [Young et al., 2007]. The GV is also known as HIAPER, the High-performance Instrumented Airborne Platform for Environmental Research. The objectives of these studies were to investigate the photochemistry effects on new particle formation and the latitude and altitude dependence of new particles. Number concentrations of aerosol particles with diameters from 4 to 2000 nm were measured with the University of Denver nuclei mode aerosol size spectrometer (NMASS) and focused cavity aerosol spectrometer (FCAS); these instruments are described in detail elsewhere [Jonsson et al., 1995; Brock et al., 2000; Lee et al., 2003, 2004; Young et al., 2007]. These high-altitude aerosol instruments have been used for several NASA and NSF field studies [Jonsson et al., 1995; Brock et al., 2000; Lee et al., 2003, 2004; Young et al., 2007], including new particle formation studies in the upper troposphere and lower stratosphere [Lee et al., 2003, 2004; Young et al., 2007]. Briefly, the NMASS consists of five condensation nuclei counters that have different supersaturation conditions to measure cumulative number concentrations of particle sizes larger than 4, 8, 15, 30, and 64 nm, respectively. The FCAS uses light scattering to measure particles from 90 to 2000 nm in diameter. Continuous size distributions from 4 to 2000 nm are obtained by combining NMASS and FCAS measurements and by using an inversion algorithm based on Markowski [1987]. The inversion also incorporates corrections for sizedependent anisokinetic sampling of the air sampling inlet, diffusion losses, and instrument counting efficiency. New particle formation events were identified with the following three criteria together: (1) N 4 6 >N 6 9,(2)N 4 9 >1cm 3, and (3) the ratio of N 4 9 over total aerosol number concentrations (N ) > 1/15 [Young et al., 2007]. [6] Previous studies in cirrus clouds with the same NMASS and FCAS have shown that shattering of cloud particles in the sampling inlet has minimal effects on aerosol number concentrations [Lee et al., 2004]. Although no cloud droplet sensors were onboard the GV aircraft, we believe that there were no substantial cloud events during these flights, based on the measured relative humidity (RH). There is also the possibility that commercial aircraft emission plumes produce high number concentrations of particles, and we have examined those events during the Progressive Science Missions [Young et al., 2007]. Results from the GV Terrain-Induced Rotor Experiment (T-REX) in 2006 showed that plume emissions often produced spikes of particle number concentrations at diameters >40 nm and these particles also showed a strong correlation with CO and water vapor. For the new particle formation events shown here, there were high concentrations of ultrafine particles with diameters <10 nm (see Figures 1 and 2). Results using CO and water vapor also indicated that these particle formation events were not affected by plumes. For these reasons, the effects of aircraft plumes on ultrafine particles reported here are likely negligible. [7] There were three days of nighttime flights during the Progressive Science Missions in order to investigate the sun exposure effects on aerosol nucleation. On 2 December 2005, the GV flew from Colorado (40 N latitude) south to latitude 18 N before sunset and returned along the same track to Colorado after sunset by flying through similar longitudes, latitudes, and altitudes. On 12 December, GV flew a similar route before sunrise and returned to Colorado after sunrise. On 19 December, GV flew north to latitude 62 N before sunrise and returned to Colorado after sunrise, returning along a similar track (see Figure 2). These flights were in the upper troposphere region (altitudes 8 to 14 km) for most of the time. All three of these flights were made over 8 h so that each flight had a 4 h of daytime and another 4 h of nighttime measurements. Nighttime aircraft experiments in the upper troposphere are rare and to best of our knowledge, these are the only three days of nighttime observations made in the tropopause region at present Long-Term, Ground-Based Measurements of New Particle Formation: Instrumentation [8] Long-term ground-based observations of new particle formation were made at the Tumbarumba flux station in a tall open Eucalypt forest in south eastern New South Wales, Australia [Suni et al., 2008]. The site was described in detail by Leuning et al. [2005]. Size distributions of air ions and naturally charged particles in the size range of nm were measured continuously from July 2005 to March 2007 with an air ion spectrometer (AIS) (Airel Ltd., Estonia). The AIS consists of two differential mobility analyzers, one for positive and another for negative ions [Mirme et al., 2007]. The AIS was located in a shed on the ground with the inlet at a height of 1.5 m. In addition, total concentrations of aerosol particles down to 14 nm were measured with a 2of7

3 existed at all times, so it is difficult to determine nocturnal formation events based on these clusters.) During weaker events, there were charged particles in the size range from nm. Nocturnal events usually started after midnight and in the early morning, but the strongest events (seen in March and April) sometimes started even in the late afternoon and continued until the next morning. Daytime events could also occur between the end of one nocturnal event and the beginning of the next one (Figure 4 bottom panel). We excluded the data for the periods of rain and fog, because there were usually charged particles in the size range from 2 8 nm during precipitation. 3. Results 3.1. Upper Tropospheric Observations [10] Figure 1 shows the measured particle number concentrations with diameters from 4 9 nm and with diameters Figure 1. The measured particle number concentrations with diameters from 4 9 nm (N 4 9 ) (red) and with diameters from nm (N ) (blue) (a), and altitude (orange), latitude (black) and surface area densities (green) (b), as a function of universal time during the sunrise experiment on 19 December The measurement was made over 8 h. The GV aircraft flew from Colorado (40 N latitude) to northern Canada (62 N) before sunrise and returned to Colorado after sunrise by flying through the same altitudes, latitudes and longitudes as before sunrise (Figure 2). Sunrise occurred when the GV arrived in the northernmost spot so that the entire northern flight was made before sunrise (for 4 h), and the southern flight after sunrise (4 h). NOAA HYSPLIT backward trajectory calculations show that air masses at the same locations of pre- and post-sunrise flights originated from similar regions (Figure 2). The same feature of aerosol size distributions and air masses was also observed during another two days of sunrise and sunset experiments. condensational particle counter (CPC) (TSI 3010) which was located at a height of 70 m on a tower. [9] Frequent nighttime aerosol formation events were observed in Tumbarumba. Whereas daytime nucleation events exhibited a typical banana shape aerosol size distribution (Figure 4 upper panel), these nighttime events did not show such a distinctive growth pattern (Figure 4 middle panel). Instead, the nighttime events were characterized by a sudden appearance of high concentrations of ions and charged particles in all size classes ( nm). (The cluster ions in the size range from nm also Figure 2. (a) Flight track for the sunrise experiment made on 19 December The NOAA HYPSLIT 24 h backward trajectory calculations for this day s measurement are also included here. These results show that air masses originated from similar regions for presunrise (blue trajectories) and post-sunrise (red trajectories) legs. (b) The measured aerosol sizes as a function of universal time during the sunrise experiment on this day. This surface plot shows that sub-10 nm particles existed nearly all the time, indicating high particle growth rates and continuous particle formation even at night. 3of7

4 Figure 3. Normalized median particle size distributions of the measured aerosols in the diameter range from 4 11 nm during the two days of sunrise (blue) and one day of sunset (green) experiments for the NSF/NCAR GV observations. The predicted size distribution after 12 h coagulations (red) is indicative of more aged aerosols, with 20% lower concentrations of particles with diameters from 4 6 nm, compared to those measured aerosols. from nm, altitude, latitude and surface area densities as a function of universal time during the sunrise experiment on 19 December During the nighttime, there were surprisingly high number concentrations of ultrafine particles (with diameters <10 nm), up to 1000 cm 3, comparable to those found during the daytime. The size distributions of particles had three modes, the largest peak <10 nm and two other smaller peaks at 20 nm and at nm (Figure 2). The surface area densities were slightly lower at night (4 mm 2 cm 3 ) than during the daytime (4.3 mm 2 cm 3 ). The measured ultrafine particle concentrations were dependent on latitude and altitude, but not on the presence of sunlight (Figures 1 and 2). These high concentrations of nighttime ultrafine particles were also observed during two other sunrise and sunset experiments with a large spatial scale (up to several hundreds of kilometers) (Figures 1 and 2). [11] The size distributions from our aircraft measurements made on 19 December 2005, as a function of time are given as a surface plot in Figure 2. This surface plot shows that sub-10 nm particles existed practically all the time, suggesting a source of these small aerosols both during the day and nighttime. The same feature was also seen during another two days of sunrise and sunset experiments. We also calculated aerosol growth rates from the surface plot based on Kulmala et al. [2001]. Our calculations show that the particle growth rate for sub-10 nm sizes could be up to 30 nm/h. Assuming that growth rates were the same even for <4 nm particles (and small clusters), particles will grow from 1 2 nm to 4 nm in 200 s. However, the growth could be also due to spatial variations of aerosol size and concentration. In these calculations, we also assumed that air masses did not change within the flight time (8 9 h), and NOAA HYSPLIT [Draxler and Rolph, 2003; Rolph, 2003] backward trajectory calculations confirm that air masses indeed originated from very similar regions for both legs of pre- and post-sunrise (or sunrise) flights (e.g., Figure 2), during the three days of sun exposure experiments. [12] Ultrafine particles are highly mobile and can easily coagulate with preexisting particles and without a source of aerosol precursors the nighttime aerosol sizes are expected to be more aged than the daytime aerosol sizes. Using our three days of nighttime (and daytime) measurements of aerosol size distributions, we also did coagulation calculations by assuming there is no supply of sulfuric acid at night, and compared the calculated nighttime aerosol sizes with those measured (Figure 3). The starting condition is the measured aerosol size distributions during the daytime (note, our observations showed no difference between day and nighttime aerosol sizes [Figures 1 and 2]). We wanted to see how aerosols sizes in the diameter range from 4 6 nm differ from the observations after 12 h of coagulation (typical nighttime hours are >12 h under the GV flight regions in December). We normalized the measured median aerosol concentrations in the size range from 4 11 nm (since a majority of the measured particles were in this range [Figure 2]), so that they have the same normalized concentrations at 6 nm. The measured aerosol concentrations from 4 6 nm were 20% higher than that predicted after 12 h coagulation at night. The shape of the calculated size distribution also indicates more aged aerosols, different from those measured. Our coagulation calculations were made in a very conservative way to avoid an over-prediction of aerosol number concentrations. The high number concentrations of ultrafine particles up to 1000 cm 3 (Figure 3), after coagulation is taken into account, furthermore demonstrate that the measured nighttime new particles are not transitory and some nucleation sources must exist at night to sustain the high numbers of ultrafine particles Long-Term Ground-Based Measurements [13] Figure 4 shows the measured AIS ion cluster size distributions indicative of the typical daytime and nighttime events seen from the long-term ground-based measurements of aerosol size distributions in Tumbarumba, Australia. The Tumbarumba measurements showed high frequencies of nighttime events, 30% of the nights over 15 months [Suni et al., 2008]. The majority of the nighttime events occurred from January through June (64% of the observed nights); from January to May, the frequency of nighttime events was a factor of 2.5 higher than that of the daytime events. High concentrations of small ion clusters in the size range from nm ( cm 3 ) also appeared at all times, both during the day and nighttime (Figure 4), consistent with observations made in other locations [Kulmala and Tammet, 2007]. However, for most of the time these small clusters will unlikely grow to larger sizes, under the conditions with high condensation sinks. The ground-based measurements of new particle formation in the continental US were mostly conducted in polluted urban environments (e.g., EPA SuperSites including Atlanta and Pittsburgh) [McMurry and Woo, 2002; Stanier et al., 2004] where high condensation sinks can easily inhibit the growth of small clusters; for example, the total background aerosol concentrations (>3 nm) are two to three orders of magnitude higher in Atlanta and Pittsburgh (up to cm 3 )[McMurry 4of7

5 Figure 4. Ion cluster and aerosol size distributions from 0.34 nm to 40 nm measured with AIS in Tumbarumba, Australia. Three days of measurements are shown here with distinct examples of only daytime event (21 July 2005) (upper panel), only nighttime nucleation (16 February 2006) (middle), and nighttime and daytime events together (10 March 2006) (lower). The average day and nighttime concentrations of particles from 0.34 to 40 nm were 2300 cm 3 and 3500 cm 3 (21 July), 3800 cm 3 and 5800 cm 3 (16 February), and 3400 cm 3 and 4700 cm 3 (10 March), respectively. Note that high concentrations of small clusters (<1.5 nm) (2000 cm 3 ) appeared at all times, consistent with Kulmala and Tammet [2007]. and Woo, 2002; Stanier et al., 2004] than in Tumbarumba ( cm 3 ). This is the likely reason that nighttime new particle formation was seen more frequently in remote areas (e.g., Tumbarumba) with low background aerosol concentrations, than in these EPA SuperSites where no nighttime events were reported [McMurry and Woo, 2002; Stanier et al., 2004]. Such a different level of the condensation sink also explains why overwhelmingly high concentrations of nighttime new particles were seen in the upper troposphere [where the total particle concentrations are often cm 3 (Figure 1)]. Note that in addition to low condensation sinks, to explain this extraordinary high frequency of nighttime nucleation in Tumbarumba, there may be also additional factors that affect aerosol precursor sources at this site, including radon, sulfuric acid, and organic compounds, although at present we cannot identify these chemical sources. 4. Discussion and Conclusions [14] The GV aircraft measurements showed efficient new particle formation during the nighttime in the upper troposphere with large temporal and spatial scales, consistent with other aircraft observations which also showed high 5of7

6 concentrations of ultrafine particles at night [Wiedensohler et al., 1997; Hermann et al., 2003; Mauldin et al., 2003; Meters et al., 2005], but differ from other aircraft observations that showed new particle formation is associated with sun exposure [Lee et al., 2003]. There are several factors that affect nighttime particle formation in the upper troposphere. Convection can bring higher concentrations of aerosol precursors and insoluble organic compounds [Kulmala et al., 2006] from lower altitudes and contribute to new particle formation in the upper troposphere. In fact, our case studies during the GV Progressive Science Missions showed that vertical motion is important for enhanced new particle formation observed in this region [Benson et al., 2007]. This is because convection can bring higher concentrations of aerosol precursors from the lower altitudes to the higher altitudes where temperatures and surface areas are lower; these factors together can create an ideal condition for nucleation, higher concentrations of aerosol precursors and lower surface areas and temperatures. SO 2 conversion by nighttime oxidants (e.g., NO 3 )[Brown et al., 2006] to H 2 SO 4 is also possible, although these processes are more important in polluted plumes where sulfur and nitrogen oxides are abundant. There is also the possibility that the marineemitted dimethylsulfide (DMS) reacts with NO 3 to produce H 2 SO 4 at night, but this reaction requires very high NO 3 concentrations [Mauldin et al., 2003; Lucas and Prinn, 2005]. In addition, many studies have shown that SO 2 uptake by cloud and fog droplets and the subsequent oxidation (e.g., by H 2 O 2 ) in the aqueous phase produce sulfate aerosols [Seinfeld and Pandis, 1998; Finlayson-Pitts and Pitts, 1999]. However, the question here is whether or not similar heterogeneous oxidation processes also occur with nanometer size particles or smaller clusters in the atmosphere. Intensive new particle formation events were seen in cirrus clouds [Lee et al., 2004], and near or in orographic clouds during the nighttime [Wiedensohler et al., 1997; Meters et al., 2005]. With the conventional gas phase homogeneous nucleation scheme (involving R1), nucleation theories cannot predict these nucleation events in clouds because of the extremely high surface areas of preexisting aerosols (e.g., >100 mm 2 cm 3 ; a factor of >20 higher in clouds than out of clouds [Lee et al., 2004]), implying that these clouds in fact may contribute to new particle formation via heterogeneous processes [Lee et al., 2004]. [15] The continuous ground-based measurements in Tumbarumba, Australia also indicate nighttime new particle formation in the clean environment. There is also other evidence that nighttime new particle formation takes place in a wide range of geographical locations [Rissler et al., 2006; Russell et al., 2007; M. S. Panero, private communication, 2007]. Our nighttime new particle formation is consistent with the ubiquitous existence of sub-3 nm neutral clusters [Kulmala et al., 2007] and it is likely that those small clusters more efficiently grow to larger sizes under the low condensation sinks rather than in the polluted environment. For new particle formation at the ground level, there is the possibility that low volatility condensable organic compounds (e.g., oxidation products of biogenic sourceemitted organics) may produce particles even without H 2 SO 4 [O Dowd et al., 2002; Bonn and Moortgat, 2003]; these organic compounds can also form during the nighttime via ozone or NO 3 in the absence of UV and OH. Observations also have shown OH production during the nighttime from hydrocarbon ozonolysis reactions with ozone [Paulson et al., 1997, 1999; Donahue et al., 1998], and such dark OH chemical sources may be very important for nighttime new particle formation. [16] We have shown aircraft and ground-based observations of high concentrations of ultrafine particles, suggesting nighttime nucleation in the troposphere under conditions with low condensation sinks. At present, we cannot conclusively identify the mechanisms for such unconventional nighttime new particle formation. We speculate that heterogeneous oxidation of SO 2 is a possible route for nighttime formation of ultrafine particles in the troposphere. Also, ammonia can contribute to aerosol nucleation in the troposphere [Lucas and Akimoto, 2006] and ammonia effects can be significant especially for those low concentrations of sulfuric acid (e.g., 10 6 cm 3 [Mauldin et al., 2003]) during the nighttime. The effects of organics on aerosol nucleation and growth [O Dowd et al., 2002; Zhang et al., 2004; Tunved et al., 2006] can also be important for nighttime aerosol formation, in the regions where high concentrations of condensable organics exist. Nighttime oxidants (NO 3 or OH production via hydrocarbon ozonolysis) and vertical convection can also play important roles. Because aerosol sizes and number concentrations, rather than the aerosol chemical composition, can play decisive roles in the cloudnucleating ability of aerosols [Dusek et al., 2006], it is crucial to understand the aerosol formation mechanisms to correctly estimate the effects of aerosol and cloud on climate. Our results raise important questions about atmospheric aerosol formation mechanisms, and future laboratory, field, and modeling studies will be required to fully address these questions. [17] Acknowledgments. This study was supported by NSF grants awarded to KSU (ATM and CAREER ATM ). NCAR is sponsored by NSF but any opinions expressed here do not reflect those of NSF. We thank Chuck Wilson for providing NMASS and FCAS for the NSF/NCAR HIAPER GV Progressive Science Missions; scientists, engineers and pilots involved in the Progressive Science Missions; Helen Cleugh and Ray Leuning at CSIRO, Australia for the assistance on the AIS measurements in Tumbarumba; Donald Lucas, Renyi Zhang, and Kari Lehtinen for helpful comments on the manuscript. References Ayers, G. P., R. W. Gillett, and J. L. Gras (1980), On the vapor pressure of sulfuric acid, Geophys. Res. Lett., 7, Benson, D. R., L.-H. Young, S.-H. Lee, T. L. Campos, D. C. Rogers, and J. Jensen (2007), When does new particle formation not occur in the upper troposphere?, Atmos. Chem. Phys. Discuss., 7, 14,209 14,232. Birmili, W., H. Berresheim, C. Plass-Dülmer, T. Elste, S. Gilge, A. Wiedensohler, and U. Uhrner (2003), The Hohenpeissenberg aerosol formation experiments (HAFEX): A long-term study including size-resolved aerosol, H 2 SO 4, OH, and monoterpenenes measurements, Atmos. Chem. Phys., 3, Bonn, B., and G. K. Moortgat (2003), Sesquiterpene ozonolysis: Origin of atmospheric new particle formation from biogenic hydrocarbons, Geophys. Res. Lett., 30(11), 1585, doi: /2003gl Brock, C. A., F. Schrörder, B. Kärcher, A. Petzold, R. Busen, and M. Fiebig (2000), Ultrafine particle size distributions measured in aircraft exhaust plumes, J. Geophys. Res., 105, 26,555 26,567. Brown, S. S., et al. (2006), Variability in nocturnal nitrogen oxide processing and its role in regional air quality, Science, 311, Donahue, N. M., J. H. Kroll, J. G. Anderson, and K. Demerjian (1998), Direct observation of OH production from the ozonolysis of olefins, Geophys. Res. Lett., 25, Draxler, R. R., and G. D. Rolph (2003), HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory), Model access via NOAA ARL 6of7

7 READY Website, NOAA Air Resources Laboratory, Silver Spring, MD. Dusek, U., et al. (2006), Size matters more than chemistry for cloud-nucleating ability of aerosol particles, Science, 312, Finlayson-Pitts, B. J., and J. N. Pitts Jr. (1999), Chemistry of Upper and Lower Atmosphere: Theory, Experiments and Applications, pp , Academic Press, San Diego, Calif. Hermann, M., J. Heintzenberg, A. Wiedensohler, A. Zahn, G. Heinrich, and C. A. M. Brenninkmeijer (2003), Meridional distributions of aerosol particle number concentrations in the upper troposphere and lower stratosphere obtained by Civil Aircraft for Regular Investigation of the Atmosphere Based on an Instrument Container (CARIBIC) flights, J. Geophys. Res., 108(D3), 4114, doi: /2001jd International Panel on Climate Change (IPCC) (2007), Climate Change 2007: The Physical Science Basis, Paris, France. Jonsson, H. H., et al. (1995), Performance of a focused cavity aerosol spectrometer for measurements in the stratosphere of particle size in the mm diameterrange,j. Atmos. Oceanic Technol., 12, Kulmala, M., and H. Tammet (2007), Finnish Estonian air ion and aerosol workshops, Boreal Environ. Res., 12, Kulmala, M., L. Pirjola, and J. Mäkelä (2000), Stable sulphate clusters as a source of new atmospheric particles, Nature, 404, Kulmala, M., M. Dal Maso, J. M. Mäkelä, L. Pirjola, M. Väkevä, P. Aalto, P. Miikkulainen, K. Hämeri, and C. D. O Dowd (2001), On the formation, growth and composition of nucleation mode particles, Tellus, Ser. A and Ser. B, 53B, Kulmala, M., H. Vëkamäki, T. Petajda, M. Dal Maso, A. Lauri, V. M. Kerminen, W. Birmili, and P. H. McMurray (2004), Formation and growth rates of ultrafine atmospheric particles: A review of observations, J. Aerosol Sci., 35, Kulmala, M., A. Reissell, M. Sipila, B. Bonn, T. M. Ruuskanen, K. E. J. Lehtinen, V. M. Kerminen, and J. Strom (2006), Deep convective clouds as aerosol production engines: Role of insoluble organics, J. Geophys. Res., 111, D17202, doi: /2005jd Kulmala, M., et al. (2007), Towards direct measurement of atmospheric nucleation, Science, 317, Lee, S.-H., J. M. Reeves, J. C. Wilson, D. E. Hunton, A. A. Viggiano, T. M. Miller, J. O. Ballenthin, and L. R. Lait (2003), Particle formation by ion nucleation in the upper troposphere and lower stratosphere, Science, 301, Lee, S.-H., et al. (2004), New particle formation observed in the tropical/ subtropical cirrus clouds, J. Geophys. Res., 109, D20209, doi: / 2004JD Leuning, R., H. A. Cleugh, S. J. Zegelin, and D. Hughes (2005), Carbon and water fluxes over a temperature eucalyptus forest and a tropical wet/ dry savanna in Australia: Measurements and comparison with MODIS remote sensing estimated, Agric. For. Meteorol., 129, Lucas, D. D., and H. Akimoto (2006), Evaluating aerosol nucleation parameterization in a global atmospheric model, Geophys. Res. Lett., 33, L10808, doi: /2006gl Lucas, D. D., and R. G. Prinn (2005), Parametric sensitivity and uncertainty analysis of dimethylsulfide oxidation in the clear-sky remote marine boundary-layer, Atmos. Chem. Phys., 5, Markowski, G. R. (1987), Improving Twomey s algorithm for inversion of aerosol measurement data, Aerosol. Sci. Technol., 7, Mauldin, R. L., III, et al. (2003), Highlights of OH, H 2 SO 4, and methane sulfonic acid measurements made aboard the NASA P-3B during transport and chemical evolution over the Pacific, J. Geophys. Res., 108(D20), 8796, doi: /2003jd McMurry, P. H., and K. S. Woo (2002), Size distributions of nm urban Atlanta aerosols: Measurements and observations, J. Aerosol Med. Deposition Clearance Eff. Lung, 15, McMurry, P. H., et al. (2005), A criterion for new particle formation in the sulfur-rich Atlanta atmosphere, J. Geophys. Res., 110, D22S02, doi: /2005jd Meters, S., D. Galgon, K. Schwirn, A. Nowak, K. Lehmann, A. Massling, A. Wiedensohler, and W. Wieprecht (2005), Evolutions of particle concentration and size distribution observed upwind, inside and downwind hill cap clouds at connected flow conditions during FEBUKO, Atmos. Environ., 39, Mirme, A., et al. (2007), A wide-range multi-channel air ion spectrometer, Boreal Environ. Res., 12, O Dowd, C. D., P. Aalto, K. Hämeri, M. Kulmala, and T. Hoffmann (2002), Atmospheric particles from organic vapors, Nature, 416, Paulson, S. E., A. D. Sen, P. Liu, J. D. Fenske, and M. J. Fox (1997), Evidence for the formation of OH radicals from the reaction of O3 with alkenes in the gas phase, Geophys. Res. Lett., 24, Paulson, S. E., M. Y. Chung, and A. S. Hasson (1999), OH radical formation from the gas-phase reaction of ozone with terminal alkenes and the relationship between structure and mechanism, J. Phys. Chem. A, 103, Rissler, J. A., A. Vestin, E. Swietlicki, G. Fisch, J. Zhou, P. Artaxo, and M. O. Andreae (2006), Size distribution and hygroscopic properties of aerosol particles from dry-season biomass burning in Amazonia, Atmos. Chem. Phys., 6, Rolph, G. D. (2003), Real-time Environmental Applications and Display system (READY) Website, NOAA Air Resources Laboratory, Silver Spring, MD. Russell, L. M., A. A. Mensah, E. V. Fischer, B. C. Sive, R. K. Varner, W. C. Keene, J. Stutz, and A. A. P. Pszenny (2007), Nanoparticle growth following photochemical a- and b-pinene oxidation at Appledore Island during International Consortium for Research on Transport and Transformation/Chemistry of Halogens at the Isles of Shoals 2004, J. Geophys. Res., 112, D10S21, doi: /2006jd Seinfeld, J. H., and S. N. Pandis (1998), Atmospheric Chemistry and Physics: From Air Pollution to Climate Change, pp , John Wiley, New York, NY. Stanier, C. O., A. Y. Kylstov, and S. N. Pandis (2004), Nucleation events during the Pittsburg air quality study: Description and relation to key meteorological, gas phase, and aerosol parameters, Aerosol Sci. Technol., 38, suppl. 1, Suni, T., et al. (2008), Formation and characteristics of ions and charged aerosol particles in a native Australian Eucalypt forest, Atmos. Chem. Phys., 8, Tunved, P., et al. (2006), High natural aerosol loading over boreal forests, Science, 312, Wiedensohler, A., et al. (1997), Night-time formation and occurrence of new particles associated with orographic clouds, Atmos. Environ., 31, Young, L.-H., et al. (2007), Enhanced new particle formation observed in the northern midlatitude tropopause region, J. Geophys. Res., 112, D10218, doi: /2006jd Zhang, R., I. Suh, J. Zhao, D. Zhang, E. C. Fortner, X. Tie, L. T. Molina, and M. J. Molina (2004), Atmospheric new particle formation enhanced by organic acids, Science, 304, D. R. Benson, S.-H. Lee, and L.-H. Young, Department of Chemistry, Kent State University, 302 Williams Hall, Kent, OH 44240, USA. (slee19@kent.edu) T. L. Campos, J. Jensen, and D. C. Rogers, Earth Observing Laboratory, National Center for Atmospheric Research, Broomfield, CO, USA. H. Junninen, M. Kulmala, and T. Suni, Department of Physical Sciences, University of Helsinki, Helsinki, Finland. 7of7