Andrew G. Allen

A dedicated study of new particle formation and fate in the coastal environment (PARFORCE): overview of objectives and achievements

A dedicated study into the formation of new particles, New Particle Formation and Fate in the Coastal Environment (PARFORCE), was conducted over a period from 1998 to 1999 at the Mace Head Atmospheric Research Station on the western coast of Ireland. Continuous measurements of new particle formation were taken over the 2-year period while two intensive field campaigns were also conducted, one in September 1998 and the other in June 1999. New particle events were observed on ∼90% of days and occurred throughout the year and in all air mass types. These events lasted for, typically, a few hours, with some events lasting more than 8 hours, and occurred during daylight hours coinciding with the occurrence of low tide and exposed shorelines. During these events, peak aerosol concentrations often exceeded 106 cm−3 under clean air conditions, while measured formation rates of detectable particle sizes (i.e., d > 3 nm) were of the order of 104–105 cm−3 s−1. Nucleation rates of new particles were estimated to be, at least, of the order of 105–106 cm−3 s−1 and occurred for sulphuric acid concentrations above 2 × 106 molecules cm−3; however, no correlation existed between peak sulphuric acid concentrations, low tide occurrence, or nucleation events. Ternary nucleation theory of the H2SO4-H2O-NH3 system predicts that nucleation rates far in excess of 106 cm−3 s−1 can readily occur for the given sulphuric acid concentrations; however, aerosol growth modeling studies predict that there is insufficient sulphuric acid to grow new particles (of ∼1 nm in size) into detectable sizes of 3 nm. Hygroscopic growth factor analysis of recently formed 8-nm particles illustrate that these particles must comprise some species significantly less soluble than sulphate aerosol. The nucleation-mode hygroscopic data, combined with the lack of detectable VOC emissions from coastal biota, the strong emission of biogenic halocarbon species, and the fingerprinting of iodine in recently formed (7 nm) particles suggest that the most likely species resulting in the growth of new particles to detectable sizes is an iodine oxide as suggested by previous laboratory experiments. It remains an open question whether nucleation is driven by self nucleation of iodine species, a halocarbon derivative, or whether first, stable clusters are formed through ternary nucleation of sulphuric acid, ammonia, and water vapor, followed by condensation growth into detectable sizes by condensation of iodine species. Airborne measurements confirm that nucleation occurs all along the coastline and that the coastal biogenic aerosol plume can extend many hundreds of kilometers away from the source. During the evolution of the coastal plume, particle growth is observed up to radiatively active sizes of 100 nm. Modeling studies of the yield of cloud-condensation nuclei suggest that the cloud condensation nuclei population can increase by ∼100%. Given that the production of new particles from coastal biogenic sources occurs at least all along the western coast of Europe, and possibly many other coastlines, it is suggested that coastal aerosols contribute significantly to the natural background aerosol population.

Coastal new particle formation: Environmental conditions and aerosol physicochemical characteristics during nucleation bursts

Nucleation mode aerosol was characterized during coastal nucleation events at Mace Head during intensive New Particle Formation and Fate in the Coastal Environment (PARFORCE) field campaigns in September 1998 and June 1999. Nucleation events were observed almost on a daily basis during the occurrence of low tide and solar irradiation. In September 1998, average nucleation mode particle concentrations were 8600 cm-3 during clean air events and 2200 cm-3 during polluted events. By comparison, during June 1999, mean nucleation mode concentrations were 27,000 cm-3 during clean events and 3350 cm-3 during polluted conditions. Peak concentrations often reached 500,000-1,000,000 cm-3 during the most intense events and the duration of the events ranged from 2 to 8 hours with a mean of 4.5 hours. Source rates for detectable particle sizes (d > 3 nm) were estimated to be between 104 and 106 cm-3 s-1 and initial growth rates of new particles were as high as 0.1-0.35 nm s-1 at the tidal source region. Recently formed 8 nm particles were subjected to hygroscopic growth and were found to have a growth factor of 1.0-1.1 for humidification at 90% relative humidity. The low growth factors implicate a condensable gas with very low solubility leading to detectable particle formation. It is not clear if this condensable gas also leads to homogeneous nucleation; however, measured sulphuric acid and ammonia concentration suggest that ternary nucleation of thermodynamically stable sulphate clusters is still likely to occur. In clear air, significant particle production (>105 cm-3) was observed with sulphuric acid gas-phase concentration as low as 2 × 10 6 molecules cm-3 and under polluted conditions as high as 1.2 × 108 molecules cm-3. Copyright 2002 by the American Geophysical Union.

Characterization and evolution of tropospheric plumes from Lascar and Villarrica volcanoes, Chile

Chile, reveal that both are significant and sustained emitters of SO2 (28 and 3.7 kg s � 1 , respectively), HCl (9.6 and 1.3 kg s � 1 , respectively), HF (4.5 and 0.3 kg s � 1 , respectively) and near-source sulfate aerosol (0.5 and 0.1 kg s � 1 , respectively). Aerosol plumes are characterized by particle number fluxes (0.08–4.0 mm radius) of � 10 17 s � 1 (Lascar) and � 10 16 s � 1 (Villarrica), the majority of which will act as cloud condensation nuclei at supersaturations >0.1%. Impactor studies suggest that the majority of these particles contain soluble SO4� . Most aerosol size distributions were bimodal with maxima at radii of 0.1–0.2 mm and 0.7–1.5 mm. The mean particle effective radius (Reff) ranged from 0.1 to 1.5 mm, and particle size evolution during transport appears to be controlled by particle water uptake (Villarrica) or loss (Lascar) rather than sulfate production. INDEX TERMS: 0305 Atmospheric Composition and Structure: Aerosols and particles (0345, 4801); 0322 Atmospheric Composition and Structure: Constituent sources and sinks; 0365 Atmospheric Composition and Structure: Troposphere—composition and chemistry; 8409 Volcanology: Atmospheric effects (0370); 8494 Volcanology: Instruments and techniques; KEYWORDS: volcanoes, degassing, aerosol sulphur dioxide, sulphate, Llaima

Size-Resolved Characterisation of Soluble Ions in the Particles in the Tropospheric Plume of Masaya Volcano, Nicaragua: Origins and Plume Processing

We present the first application of a multi-stage impactor to study volcanic particle emissions to the troposphere from Masaya volcano, Nicaragua. Concentrations of soluble SO42−,Cl−, F−, NO3−, K+, Na+,NH4+, Ca2+ and Mg2+ were determined in 11 size bins from ∼0.07 μm to >25.5 μm. The near-source size distributions showed major modes at 0.5μm (SO42−, H+,NH4+); 0.2 μm and 5.0 μm (Cl−) and 2.0–5.0 μm(F−). K+ and Na+ mirrored the SO42− size-resolvedconcentrations closely, suggesting that these were transported primarily asK2SO4 and Na2SO4 in acidic solution, while Mg2+ andCa2+ presented modes in both <1 μm and >1 μm particles. Changes in relative humidity were studied by comparing daytime (transparent plume) and night-time (condensed plume) results. Enhanced particle growth rates were observed in the night-time plume as well as preferential scavenging of soluble gases, such as HCl, by condensed water. Neutralisation of the acidic aerosol by background ammonia was observed at the crater rim and to a greater extent approximately 15 km downwind of the active crater. We report measurements of re-suspended near-source volcanic dust, which may form a component of the plume downwind. Elevated levels ofSO42−, Cl−, F−,H+, Na+, K+ and Mg2+ were observed around the 10 μm particle diameter in this dust. The volcanic SO42− flux leaving the craterwas ∼0.07 kg s−1.

Nitric acid from volcanoes

Abstract Atmospheric cycling of nitric acid and other nitrogen-bearing compounds is an important biogeochemical process, with significant implications for ecosystems and human health. Volcanoes are rarely considered as part of the global nitrogen cycle, but here we show that they release a previously unconsidered flux of HNO3 vapour to the atmosphere. We report the first measurements of nitric acid vapour in the persistent plumes from four volcanoes: Masaya (Nicaragua); Etna (Italy); and Villarrica and Lascar (Chile). Mean near-source volcanic plume concentrations of HNO3 range from 1.8 to 5.6 μmol m−3, an enrichment of one to two orders of magnitude over background (0.1–1.5 μmol m−3). Using mean molar HNO3/SO2 ratios of 0.01, 0.02, 0.05, and 0.07 for Villarrica, Masaya, Etna, and Lascar respectively, combined with SO2 flux measurements, we calculate gaseous HNO3 fluxes from each of these volcanic systems, and extend this to estimate the global flux from high-temperature, non-explosive volcanism to be ∼0.02–0.06 Tg (N) yr−1. While comparatively small on the global scale, this flux could have important implications for regional fixed N budgets. The precise mechanism for the emission of this HNO3 remains unclear but we suggest that thermal nitrogen fixation followed by rapid oxidation of the product NO is most likely. In explosive, ash-rich plumes NO may result from, or at least be supplemented by, production from volcanic lightning rather than thermal N fixation. We have calculated NO production via this route to be of the order of 0.02 Tg (N) yr−1.

Biomass burning in the Amazon: characterization of the ionic component of aerosols generated from flaming and smoldering rainforest and savannah.

Samples (170) of biomass combustion smoke were obtained in Brazilian tropical rainforest and savannah during August-September 1992. Speciation of the ionic fraction of fine (d d > 2 µm) mode particles was achieved using ion chromatography, and the data were interpreted according to geographical region, vegetation type, and phase of combustion. The species measured were CH_3COO^-, HCOO^-, CI^-, NO_3^-, SO_4^(2-), C_2O_4^(2-), PO_4^(3-), Na^+, NH_4^+, K^+, Mg^(2+), and Ca^(2+). Mean total ionic contents for the fine mode were 5.6% (Brasilia) and 2.2% (Rondonia) and for the coarse mode were 8.3% (Brasilia) and 13.7% (Rondonia). The fine mode ionic fraction was dominated by potassium, chloride, and sulfate, while calcium and acetate were important in the coarse mode. Highest ionic contributions were always observed during flaming combustion. Sources of aerosols were condensation/coagulation, release of partially combusted plant material, and suspension of soil particles.

Volcanic source for fixed nitrogen in the early Earth's atmosphere

Hot volcanic vents promote the thermal fixation of atmospheric N2 into biologically available forms. The importance of this process for the global nitrogen cycle is poorly understood. At Masaya volcano, Nicaragua, NO and NO2 are intimately associated with volcanic aerosol, such that NOx levels reach as much as an order of magnitude above local background. In-plume HNO3 concentrations are elevated above background to an even greater extent (#50 mmol·m 23 ). We estimate the production efficiency of fixed nitrogen at hot vents to be ;3 3 10 28 mol·J 21 , implying present-day global production of ;10 9 mol of fixed N per year. Although conversion efficiency would have been lower in a preoxygenated atmosphere, we suggest that subaerial volcanoes potentially constituted an important source of fixed nitrogen in the early Earth, producing as much as ;10 11 mol·yr 21 of fixed N during major episodes of volcanism. These fluxes are comparable to estimated nitrogen-fixation rates in the prebiotic Earth from other major sources such as bolide impacts and thunderstorm and volcanic lightning.

Sources, size distribution, and downwind grounding of aerosols from Mount Etna

The number concentrations and size distributions of aerosol particles >0.3 mm diameter were measured at the summit of Mount Etna and up to 10 km downwind from the degassing vents during July and August 2004. Aerosol number concentrations reached in excess of 9 106 L1 at summit vents, compared to 4–8 104 L1 in background air. Number concentrations of intermediate size particles were higher in emissions from the Northeast crater compared to other summit crater vents, and chemical composition measurements showed that Northeast crater aerosols contained a higher mineral cation content compared to those from Voragine or Bocca Nuova, attributed to Strombolian or gas puffing activity within the vent. Downwind from the summit the airborne plume was located using zenith sky ultraviolet spectroscopy. Simultaneous measurements indicated a coincidence of elevated ground level aerosol concentrations with overhead SO2, demonstrating rapid downward mixing of the plume onto the lower flanks of the volcano under certain meteorological conditions. At downwind sites the ground level particle number concentrations were elevated in all size fractions, notably in the 2.0–7.5 mm size range. These findings are relevant for assessing human health hazard and suggest that aerosol size distribution measurements may aid volcanic risk management.

Sources of atmospheric methanesulphonate, non-sea-salt sulphate, nitrate and related species over the temperate South Pacific

Aerosol species and trace gases were collected during three intensive sampling periods representing winter (2 July to 9 August 1991), spring (29 September to 6 November 1991) and summer (15 January to 29 February 1992) at Baring Head on the southern end of New Zealands North Island. In remote marine air, mean winter, spring and summer aerosol phase concentrations (ng m−3) were, respectively, non-sea-salt sulphate (NSSS) = 115, 139 and 187, methanesulphonate (MSA) = 1.5, 23 and 48, nitrate (NOD = 293, 70 and 84, and ammonium (NH4+) = 44, 39 and 59. Mean gas phase concentrations were sulphur dioxide (SO2−) = 39,39 and 30, nitric acid (HNO3) = 18,30 and 27 and ammonia (NH3) = 74,42 and 31. Although natural sources for atmospheric reactive sulphur and nitrogen predominate in this region, evidence was also found for the existence of additional, non-oceanic sources of NOn3 and NSSS, including a long-range transport source of ammonium sulphate. Diurnal fluctuations were seen with lower SO2and NH3 concentrations at night, while nighttime sources of NSSS and HNO3 were indicated. Molar ratios of MSA/NSSS in remote air were 1.3, 16.3 and 25.7% for winter, spring and summer samples, respectively. These could be only partially explained by known DMS oxidation mechanisms, and a low, relatively constant background concentration of non-DMS NSSS appeared to be present at all times.

An analysis of rapid increases in condensation nuclei concentrations at a remote coastal site in western Ireland.

Massive “bursts” in condensation nuclei (CN) concentration were recorded at a remote site on the west Irish coast during campaigns in summer 1996 and spring/summer 1997. Number concentrations of 3–7 nm diameter CN were observed to rise daily from 102–103 up to ∼105 /cm3 for 1–3 hours. Data were collected as part of the Atmospheric Chemistry Studies in the Oceanic Environment program. In a previous paper the burst phenomenon was linked to the movement of the tide, and it was suggested that enhanced biogenic emissions occurred near low tide with concomitant rapid homogeneous gas phase CN formation. In this paper possible chemical mechanisms for the burst phenomenon are investigated. Two approaches are adopted. First, by assuming a 20:80 sulfate:water molar composition and calculating the number distribution using data from condensation particle counters, the total mass of CN formed during a burst is evaluated. This is compared with that mass of sulfate produced by OH-initiated dimethyl sulfide (DMS) oxidation. The procedure is termed “mass balance.” Second, a variety of chemical species are coplotted with tidal height. DMS oxidation is not believed to play a major role in CN formation at this site because (1) the mass balance calculations imply ambient DMS concentrations higher than those observed, and (2) gas phase HCl, HNO3, SO2, and NH3 did not exhibit any discernible correlation with tidal height. Further, none of the suite of observed nonmethane hydrocarbons or DMS showed a tidal relation. No mechanism has to date been convincingly identified for the burst phenomenon.