Information on the density (rho) and refractive index m (=n-ik) of elemental carbon (ECa) and organic matter (OMa), the main carbon components of atmospheric aerosols, has frequently been obtained from closure calculations between physical and chemical aerosol properties. However, this approach has suffered from large uncertainties since there were more unknown (or poorly known) parameters than defining equations. In this study, we propose a method that avoids this ambiguity mainly by considering both optical and mass closure and by expressing the three ECa parameters (rho(ECa), n(ECa), k(ECa)) by a single (unknown) parameter. This allows mathematically rigorous determination of rho(Eca), m(ECa), rho(OMa) and m(OMa) from standard physico-chemical aerosol data and rigorous error analysis. The results are unambiguous and self-consistent, i.e., there is no difference between the chemically and physically derived p and m values of the atmospheric aerosol. Application of this method to our previously published data on biomass burning particles from Amazonia yields rho(ECa) = 1.8(+/-0.2) g/cm3, m(ECa) = 1.9(+/-0.1)-i0.20(-0.04/+0.02), rho(OMa) = 1.39(+/-0.13) g/cm3 and m(OMa) = 1.46(+/-0.02), where the launcertainty limits given in parenthesis are based on the principles of error propagation. The relatively low imaginary part of m(ECa) indicates the presence of only partially graphitized elemental carbon, which is consistentwith biomass burning aerosol dominated by smoldering combustion conditions.

Fine aerosol particles were collected separately during daytime and nighttime at a tropical pasture site in Rondonia, Brazil, during the burning and dry-to-wet transition period in 2002. Total carbon (TC) and water-soluble organic carbon (WSOC) were measured by evolved gas analysis (EGA). Based on the thermochemical properties of the fine aerosol, the relative amounts of the volatile and refractory compounds were estimated. It was found that the thermally refractory (possibly higher molecular weight) compounds dominated the TC composition. Their contribution to TC was higher in the daytime than in the nighttime samples. The relative share of WSOC also showed a statistically significant diel variation as did its refractory fraction. Anhydrosugars and phenolic acids were determined by GC-MS and their diel variation was studied. Based on the decrease of their relative concentrations between the biomass burning and transition periods and their distinctly different diel variations, we suggest that the phenolic acids may undergo chemical transformations in the aerosol phase, possibly towards more refractory compounds (humic-like substances, HULIS), as has been suggested previously. These conclusions are supported by the results of the thermally assisted hydrolysis and methylation gas chromatography-mass spectrometry of the same filter samples.

We present here the optical properties of humic-like substances (HULIS) isolated from the fine fraction of biomass-burning aerosol collected in the Amazon basin during the LBA-SMOCC (Large scale Biosphere atmosphere experiment in Amazonia – SMOke aerosols, Clouds, rainfall and Climate) experiment in September 2002. From the isolated HULIS, aerosol particles were generated and their scattering and absorption coefficients measured. The size distribution and mass of the particles were also recorded. The value of the index of refraction was derived from "closure" calculations based on particle size, scattering and absorption measurements. On average, the complex index of refraction at 532 nm of HULIS collected during day and nighttime was 1.65–0.0019i and 1.69–0.0016i, respectively. In addition, the imaginary part of the complex index of refraction was calculated using the measured absorption coefficient of the bulk HULIS. The mass absorption coefficient of the HULIS at 532 nm was found to be quite low (0.031 and 0.029 m2 g−1 for the day and night samples, respectively). However, due to the high absorption Ångström exponent (6–7) of HULIS, the specific absorption increases substantially towards shorter wavelengths (~2–3 m2 g−1 at 300 nm), causing a relatively high (up to 50%) contribution to the light absorption of our Amazonian aerosol at 300 nm. For the relative contribution of HULIS to light absorption in the entire solar spectrum, lower values (6.4–8.6%) are obtained, but those are still not negligible.

As part of the Large Scale Biosphere-Atmosphere Experiment in Amazonia – Smoke, Aerosols, Clouds, Rainfall and Climate (LBA-SMOCC) campaign, detailed surface and airborne aerosol measurements were performed over the Amazon Basin during the dry to wet season from 16 September to 14 November 2002. Optical and physical properties of aerosols at the surface, and in the boundary layer (BL) and free troposphere (FT) during the dry season are discussed in this article. Carbon monoxide (CO) is used as a tracer for biomass burning emissions. At the surface, good correlation among the light scattering coefficient (σs at 545 nm), PM2.5, and CO indicates that biomass burning is the main source of aerosols. Accumulation of haze during some of the large-scale biomass burning events led to high PM2.5 (225 μg m−3), σs (1435 Mm−1), aerosol optical depth at 500 nm (3.0), and CO (3000 ppb). A few rainy episodes reduced the PM2.5, number concentration (CN) and CO concentration by two orders of magnitude. The correlation analysis between σs and aerosol optical thickness shows that most of the optically active aerosols are confined to a layer with a scale height of 1617 m during the burning season. This is confirmed by aircraft profiles. The average mass scattering and absorption efficiencies (545 nm) for small particles (diameter Dp<1.5 μm) at surface level are found to be 5.0 and 0.33 m2 g−1, respectively, when relating the aerosol optical properties to PM2.5 aerosols. The observed mean single scattering albedo (ωo at 545 nm) for submicron aerosols at the surface is 0.92±0.02. The light scattering by particles (Δσs/Δ CN) increase 2–10 times from the surface to the FT, most probably due to the combined affects of coagulation and condensation.

Fine aerosol particles were collected separately during daytime and nighttime at a tropical pasture site in Rondonia, Brazil, during the burning and dry-to-wet transition period in 2002. Total carbon (TC) and water-soluble organic carbon (WSOC) were measured by evolved gas analysis (EGA). Based on the thermochemical properties of the fine aerosol, the relative amounts of the low and higher molecular weight compounds were estimated. It was found that the thermally refractory (possibly higher molecular weight) compounds dominated the TC composition. Their contribution to TC was higher in the daytime samples than in the nighttime ones. The relative share of WSOC also showed a statistically significant diel variation and so did its refractory fraction. Anhydrosugars and phenolic acids were determined by GC-MS and their diel variation was studied. Based on the decrease of their relative concentrations between the biomass burning and transition periods and their distinctly different diel variations, we suggest that the phenolic acids may undergo chemical transformations in the aerosol phase, possibly towards more refractory compounds (humic-like substances, HULIS), as has been suggested previously. These conclusions are supported by the results of the thermally assisted hydrolysis and methylation gas chromatography-mass spectrometry of the same filter samples.

The aerosol characterization experiment performed within the Large‐Scale Biosphere‐Atmosphere Experiment in Amazonia–Smoke, Aerosols, Clouds, Rainfall and Climate (LBA‐SMOCC) field experiment carried out in Rondônia, Brazil, in the period from September to November 2002 provides a unique data set of size‐resolved chemical composition of boundary layer aerosol over the Amazon Basin from the intense biomass‐burning period to the onset of the wet season. Three main periods were clearly distinguished on the basis of the PM 10 concentration trend during the experiment: (1) dry period, with average PM 10 well above 50 μ g m −3 ; (2) transition period, during which the 24‐hour‐averaged PM 10 never exceeded 40 μ g m −3 and never dropped below 10 μ g m −3 ; (3) and wet period, characterized by 48‐hour‐averaged concentrations of PM 10 below 12 μ g m −3 and sometimes as low as 2 μ g m −3 . The trend of PM 10 reflects that of CO concentration and can be directly linked to the decreasing intensity of the biomass‐burning activities from September through November, because of the progressive onset of the wet season. Two prominent aerosol modes, in the submicron and supermicron size ranges, were detected throughout the experiment. Dry period size distributions are dominated by the fine mode, while the fine and coarse modes show almost the same concentrations during the wet period. The supermicron fraction of the aerosol is composed mainly of primary particles of crustal or biological origin, whereas submicron particles are produced in high concentrations only during the biomass‐burning periods and are mainly composed of organic material, mostly water‐soluble, and ∼10% of soluble inorganic salts, with sulphate as the major anion. Size‐resolved average aerosol chemical compositions are reported for the dry, transition, and wet periods. However, significant variations in the aerosol composition and concentrations were observed within each period, which can be classified into two categories: (1) diurnal oscillations, caused by the diurnal cycle of the boundary layer and the different combustion phase active during day (flaming) or night (smouldering); and (2) day‐to‐day variations, due to alternating phases of relatively wet and dry conditions. In a second part of the study, three subperiods representative of the conditions occurring in the dry, transition, and wet periods were isolated to follow the evolution of the aerosol chemical composition as a function of changes in rainfall rate and in the strength of the sources of particulate matter. The chemical data set provided by the SMOCC field experiment will be useful to characterize the aerosol hygroscopic properties and the ability of the particles to act as cloud condensation nuclei.

One of the major limitations in advancing the understanding of tropospheric ozone and aerosol generation and developing strategies for their control is the technical ability to accurately measure volatile organic compounds (VOCs). This paper describes the design of a constant flow VOC sampler. The versatile sampler can be used for fully automated concentration and flux measurements of VOCs. The sampler incorporates a microprocessor control unit and provides highly accurate mass flow control and significant ease of operation. Sampling sequences can be programmed directly or by remote control through a PC. All important operational parameters necessary for a complete sampling audit trail are logged. Compact weatherproof housings and low power consumption allow operation at remote sites and locations which are sensitive to disturbances or have restricted access. Inner wetted surfaces of the sampler are constructed from non-contaminating materials that do not sorb or emit VOC, and thus permit the collection of representative samples even in environments with very low VOC concentrations. The cartridge magazine provides a maximum of 20 sequential cartridge samples, which allows for long-term air quality assessments. In the dual channel mode, two samples can be collected simultaneously through two independent sample loops, providing ten sequential sample pairs. This design allows the parallel collection of (a) quality assurance backup samples, (b) samples on two different types of cartridges/sorbents to allow a variety of analyses, or (c) differential samples for flux measurements using enclosure, aerodynamic profile, or relaxed eddy accumulation (REA) methods. Field applications including airplane profile measurements above a tropical rainforest area, as well as gradient and REA measurements over a mid-latitude mixed forest stand are described, and demonstrate the validity and flexibility of the system. In particular, the application of the VOC sampler as an integrated part of a REA system is emphasized.

As part of the LBA-SMOCC (Large-Scale Biosphere-Atmosphere Experiment in Amazonia - Smoke, Aerosols, Clouds, Rainfall, and Climate) 2002 campaign, we studied the emission of carbon monoxide (CO), carbon dioxide (CO2), and aerosol particles from Amazonian deforestation fires using an instrumented aircraft. Emission ratios for aerosol number (CN) relative to CO (ERCN/CO) fell in the range 14-32 cm-3 ppb-1 in most of the investigated smoke plumes. Particle number emission ratios have to our knowledge not been previously measured in tropical deforestation fires, but our results are in agreement with values usually found from tropical savanna fires. The number of particles emitted per amount biomass burned was found to be dependent on the fire conditions (combustion efficiency). Variability in ERCN/CO between fires was similar to the variability caused by variations in combustion behavior within each individual fire. This was confirmed by observations of CO-to-CO2 emission ratios (ERCO/CO2), which stretched across the same wide range of values for individual fires as for all the fires observed during the sampling campaign, reflecting the fact that flaming and smoldering phases are present simultaneously in deforestation fires. Emission factors (EF) for CO and aerosol particles were computed and a correction was applied for the residual smoldering combustion (RSC) fraction of emissions that are not sampled by the aircraft, which increased the EF by a factor of 1.5-2.1. Vertical transport of smoke from the boundary layer (BL) to the cloud detrainment layer (CDL) and the free troposphere (FT) was found to be a very common phenomenon. We observed a 20% loss in particle number as a result of this vertical transport and subsequent cloud processing, attributable to in-cloud coagulation. This small loss fraction suggests that this mode of transport is very efficient in terms of particle numbers and occurs mostly via non-precipitating clouds. The detrained aerosol particles released in the CDL and FT were larger than in the unprocessed smoke, mostly due to coagulation and secondary growth, and therefore more efficient at scattering radiation and nucleating cloud droplets. This process may have significant atmospheric implications on a regional and larger scale.

Changes in aerosol composition associated with a cold front passage were examined during a field experiment in Tel Aviv, Israel (2–15 Dec, 2000). In addition to monitoring aerosol scattering and optical thickness, aerosol samples were collected for detailed chemical analyses. Data were compared to simultaneous measurements made at Sde Boker, a semi-remote site in the Negev Desert, to help determine what changes were due to local pollution as opposed to regional phenomena. During the pre-frontal period (2–7 Dec) both sites were influenced by air masses containing a relatively high content of sulphate and dust, originating from neighbouring regions of the Middle East. A steady build-up of local pollution was then observed in Tel Aviv due to vehicular emissions/industrial activities, as indicated by increasing concentrations of black carbon, organic carbon, V, Cu, Ni, Zn, Br, Pb, NO3− and PAHs. Identification of a number of organic biomass burning tracers (e.g., levoglucosan) indicates that smoke also contributed to the pollution build-up in Tel Aviv, while a range of sugars/sugar alcohols point to a microbial/bioaerosol component. Locally emitted pollutants tended to exhibit higher nighttime concentrations due to trapping of pollution under a nocturnal inversion. Fine aerosol iodine was the only element exhibiting higher daytime concentrations, hinting at a photochemical source. Post-frontal measurements (12–15 Dec) revealed a significant decrease in all pollutants due to dispersal of the haze by the cold front (8–9 Dec), with the air initially being dominated by marine aerosol. Concentrations of pollutants then began to increase, with backward trajectories indicating a possible contribution from Eastern Europe. Overall, the study identified a range of useful tracers for monitoring the contribution of different sources to the aerosol over Israel.

Detailed organic analysis of natural aerosols from the Amazonian rain forest showed considerable quantities of previously unobserved polar organic compounds, which were identified as a mixture of two diastereoisomeric 2-methyltetrols: 2-methylthreitol and 2-methylerythritol. These polyols, which have the isoprene skeleton, can be explained by OH radical-initiated photooxidation of isoprene. They have low vapor pressure, allowing them to condense onto preexisting particles. It is estimated that photooxidation of isoprene results in an annual global production of about 2 teragrams of the polyols, a substantial fraction of the Intergovernmental Panel on Climate Change estimate of between 8 and 40 teragrams per year of secondary organic aerosol from biogenic sources.

As part of the Large‐Scale Biosphere‐Atmosphere Experiment in Amazonia‐European Studies on Trace Gases and Atmospheric Chemistry (LBA‐EUSTACH), size‐fractionated aerosol samples were collected at a primary rain forest in the Brazilian Amazon during two field campaigns in April–May and September–October 1999. These two periods encompassed parts of the wet and dry seasons, respectively. Daytime‐nighttime‐segregated sampling was carried out at three different heights (above, within, and below canopy level) on a 54‐m meteorological tower at the forest site in order to better characterize the aerosol sources. The samples were analyzed for up to 19 trace elements by particle‐induced X‐ray emission analysis and for carbonaceous components by thermal‐optical analysis. Equivalent black carbon (BCe) and gravimetric analyses were also performed. The average mass concentrations for particles <2 μm diameter were 2.2 and 33.5 μg m−3for the wet and the dry seasons, respectively. The elements related to biomass burning and soil dust generally exhibited highest concentrations above the canopy and during daytime, while forest‐derived aerosol was more concentrated underneath the canopy and during nighttime. These variations can be largely attributed to daytime convective mixing and the formation of a shallow nocturnal boundary layer, along with the possibility of enhanced nighttime release of biogenic aerosol particles. Mass scattering (αs) and mass absorption efficiency (αa) data indicate that scattering was dominated by fine aerosol, while fine and coarse aerosol both contributed significantly to absorption during both seasons. The data also suggest that components other than elemental carbon were responsible for a substantial fraction of the absorption.