Search

The apparent volatility of atmospheric organic aerosol (OA) particles is determined by their chemical composition and environmental conditions (e.g., ambient temperature). A quantitative, experimental assessment of volatility and the respective importance of these two factors remains challenging, especially in ambient measurements. We present molecular composition and volatility of oxygenated OA (OOA) particles in different rural, urban, and mountain environments (including Chacaltaya, Bolivia; Alabama, US; Hyytiälä, Finland; Stuttgart and Karlsruhe, Germany; and Delhi, India) based on deployments of a filter inlet for gases and aerosols coupled to a high-resolution time-of-flight chemical ionization mass spectrometer (FIGAERO-CIMS). We find on average larger carbon numbers (nC​​​​​​​) and lower oxygen-to-carbon (O : C) ratios at the urban sites (nC: 9.8 ± 0.7; O : C: 0.76 ± 0.03; average ±1 standard deviation) compared to the rural (nC: 8.8 ± 0.6; O : C: 0.80 ± 0.05) and mountain stations (nC: 8.1 ± 0.8; O : C: 0.91 ± 0.07), indicative of different emission sources and chemistry. Compounds containing only carbon, hydrogen, and oxygen atoms (CHO) contribute the most to the total OOA mass at the rural sites (79.9 ± 5.2 %), in accordance with their proximity to forested areas (66.2 ± 5.5 % at the mountain sites and 72.6 ± 4.3 % at the urban sites). The largest contribution of nitrogen-containing compounds (CHON) is found at the urban stations (27.1 ± 4.3 %), consistent with their higher NOx levels. Moreover, we parametrize OOA volatility (saturation mass concentrations, Csat) using molecular composition information and compare it with the bulk apparent volatility derived from thermal desorption of the OOA particles within the FIGAERO. We find differences in Csat values of up to ∼ 3 orders of magnitude and variation in thermal desorption profiles (thermograms) across different locations and systems. From our study, we draw the general conclusion that environmental conditions (e.g., ambient temperature) do not directly affect OOA apparent volatility but rather indirectly by influencing the sources and chemistry of the environment and thus the chemical composition. The comprehensive dataset provides results that show the complex thermodynamics and chemistry of OOA and their changes during its lifetime in the atmosphere. We conclude that generally the chemical description of OOA suffices to predict its apparent volatility, at least qualitatively. Our study thus provides new insights that will help guide choices of, e.g., descriptions of OOA volatility in different model frameworks such as air quality models and cloud parcel models.

Carbon monoxide (CO) concentrations in the troposphere are decreasing globally, with Africa as an exception. Yet, the region is understudied, with a deficit of ground-based observations and highly uncertain CO emission inventories. This paper reports multiyear observational CO data from the Mt. Kenya Global Atmosphere Watch (GAW) station, as well as summertime CO isotope observations from both Mt. Kenya and Nairobi, Kenya. The CO variability at Mt. Kenya is characterized by slightly increased concentrations during dry periods and a strong influence of short-term pollution events. While some data gaps and differences in instrumentation complicate decadal-scale trend analysis, a small long-term increase is resolved. High-pollution events are consistent with isotopic signal from downwind savanna fires. The isotope fingerprint of CO in Nairobi indicates an overwhelming dominance (near 100 %) of primary emissions from fossil fuel combustion with implications for air pollution policy. In contrast, the isotope signature of CO intercepted at the large-footprint Mt. Kenya region suggests that at least 70 % is primary sourced, with a predominance likely from savanna fires in Africa. Taken together, this study provides quantitative constraints of primary vs. secondary CO in the eastern Africa region and in urban Nairobi, with implications for satellite-based emission inventories as well as for chemical transport and climate modeling.

Atmospheric pollution in urban regions is highly influenced by oxidants due to their important role in the formation of secondary organic aerosol (SOA) and smog. These include the nitrate radical (NO3), which is typically considered a nighttime oxidant, and the chlorine radical (Cl), an extremely potent oxidant that can be released in the morning in chloride-rich environments as a result of nocturnal build-up of nitryl chloride (ClNO2). Chloride makes up a higher percentage of particulate matter in Delhi than has been observed anywhere else in the world, which results in Cl having an unusually strong influence in this city. Here, we present observations and model results revealing that atmospheric chemistry in Delhi exhibits an unusual diel cycle that is controlled by high concentrations of NO during the night. As a result of this, the formation of both NO3 and dinitrogen pentoxide (N2O5), a precursor of ClNO2 and thus Cl, are suppressed at night and increase to unusually high levels during the day. Our results indicate that a substantial reduction in nighttime NO has the potential to increase both nocturnal oxidation via NO3 and the production of Cl during the day.

Dimethyl sulfide (DMS), a gas produced by phytoplankton, is the largest source of atmospheric sulfur over marine areas. DMS undergoes oxidation in the atmosphere to form a range of oxidation products, out of which sulfuric acid (SA) is well known for participating in the formation and growth of atmospheric aerosol particles, and the same is also presumed for methanesulfonic acid (MSA). Recently, a new oxidation product of DMS, hydroperoxymethyl thioformate (HPMTF), was discovered and later also measured in the atmosphere. Little is still known about the fate of this compound and its potential to partition into the particle phase. In this study, we present a full year (2020) of concurrent gas- and particle-phase observations of HPMTF, MSA, SA and other DMS oxidation products at the Zeppelin Observatory (Ny-Ålesund, Svalbard) located in the Arctic. This is the first time HPMTF has been measured in Svalbard and attempted to be observed in atmospheric particles. The results show that gas-phase HPMTF concentrations largely follow the same pattern as MSA during the sunlit months (April–September), indicating production of HPMTF around Svalbard. However, HPMTF was not observed in significant amounts in the particle phase, despite high gas-phase levels. Particulate MSA and SA were observed during the sunlit months, although the highest median levels of particulate SA were measured in February, coinciding with the highest gaseous SA levels with assumed anthropogenic origin. We further show that gas- and particle-phase MSA and SA are coupled in May–July, whereas HPMTF lies outside of this correlation due to the low particulate concentrations. These results provide more information about the relationship between HPMTF and other DMS oxidation products, in a part of the world where these have not been explored yet, and about HPMTF's ability to contribute to particle growth and cloud formation.

Secondary organic aerosol (SOA) is formed through the oxidation of volatile organic compounds (VOCs), which can be of both natural and anthropogenic origin. While the hydroxyl radical (OH) and ozone (O3) are the main atmospheric oxidants during the day, the nitrate radical (NO3) becomes more important during the nighttime. Yet, atmospheric nitrate chemistry has received less attention compared to OH and O3. The Nitrate Aerosol and Volatility Experiment (NArVE) aimed to study the NO3-induced SOA formation and evolution from three biogenic VOCs (BVOCs), namely isoprene, α-pinene, and β-caryophyllene. The volatility of aerosol particles was studied using isothermal evaporation chambers, temperature-dependent evaporation in a volatility tandem differential mobility analyzer (VTDMA), and thermal desorption in a filter inlet for gases and aerosols coupled to a chemical ionization mass spectrometer (FIGAERO-CIMS). Data from these three setups present a cohesive picture of the volatility of the SOA formed in the dark from the three biogenic precursors. Under our experimental conditions, the SOA formed from NO3 + α-pinene was generally more volatile than SOA from α-pinene ozonolysis, while the NO3 oxidation of isoprene produced similar although slightly less volatile SOA than α-pinene under our experimental conditions. β-Caryophyllene reactions with NO3 resulted in the least volatile species. Four different parameterizations for estimating the saturation vapor pressure of the oxidation products were tested for reproducing the observed evaporation in a kinetic modeling framework. Our results show that the SOA from nitrate oxidation of α-pinene or isoprene is dominated by low-volatility organic compounds (LVOCs) and semi-volatile organic compounds (SVOCs), while the corresponding SOA from β-caryophyllene consists primarily of extremely low-volatility organic compounds (ELVOCs) and LVOCs. The parameterizations yielded variable results in terms of reproducing the observed evaporation, and generally the comparisons pointed to a need for re-evaluating the treatment of the nitrate group in such parameterizations. Strategies for improving the predictive power of the volatility parameterizations, particularly in relation to the contribution from the nitrate group, are discussed.

The role aerosol chemical composition plays in Arctic low-level cloud formation is still poorly understood. In this study we address this issue by combining in situ observations of the chemical characteristics of cloud residuals (dried liquid cloud droplets or ice crystals) and aerosol particles from the Zeppelin Observatory in Ny-Ålesund, Svalbard (approx. 480 m a.s.l.). These measurements were part of the 1-year-long Ny-Ålesund Aerosol and Cloud Experiment 2019–2020 (NASCENT). To obtain the chemical composition of cloud residuals at molecular level, we deployed a Filter Inlet for Gases and AEROsols coupled to a Chemical Ionization Mass Spectrometer (FIGAERO-CIMS) with iodide as the reagent ion behind a ground-based counterflow virtual impactor (GCVI). The station was enshrouded in clouds roughly 15 % of the time during NASCENT, out of which we analyzed 14 cloud events between December 2019 and December 2020. During the entire year, the composition of the cloud residuals shows contributions from oxygenated organic compounds, including organonitrates, and traces of the biomass burning tracer levoglucosan. In summer, methanesulfonic acid (MSA), an oxidation product of dimethyl sulfide (DMS), shows large contributions to the sampled mass, indicating marine natural sources of cloud condensation nuclei (CCN) and ice nucleating particle (INP) mass during the sunlit part of the year. In addition, we also find contributions of the inorganic acids nitric acid and sulfuric acid, with outstanding high absolute signals of sulfuric acid in one cloud residual sample in spring and one in late summer (21 May and 12 September 2020), probably caused by high anthropogenic sulfur emissions near the Barents Sea and Kara Sea. During one particular cloud event, on 18 May 2020, the air mass origin did not change before, during, or after the cloud. We therefore chose it as a case study to investigate cloud impact on aerosol physicochemical properties. We show that the overall chemical composition of the organic aerosol particles was similar before, during, and after the cloud, indicating that the particles had already undergone one or several cycles of cloud processing before being measured as residuals at the Zeppelin Observatory and/or that, on the timescales of the observed cloud event, cloud processing of the organic fraction can be neglected. Meanwhile, there were on average fewer particles but relatively more in the accumulation mode after the cloud. Comparing the signals of sulfur-containing compounds of cloud residuals with aerosols during cloud-free conditions, we find that sulfuric acid had a higher relative contribution to the cloud residuals than to aerosols during cloud-free conditions, but we did not observe an increase in particulate MSA due to the cloud. Overall, the chemical composition, especially of the organic fraction of the Arctic cloud residuals, reflected the overall composition of the general aerosol population well. Our results thus suggest that most aerosols can serve as seeds for low-level clouds in the Arctic.

In recent years, the Indian capital city of Delhi has been impacted by very high levels of air pollution, especially during winter. Comprehensive knowledge of the composition and sources of the organic aerosol (OA), which constitutes a substantial fraction of total particulate mass (PM) in Delhi, is central to formulating effective public health policies. Previous source apportionment studies in Delhi identified key sources of primary OA (POA) and showed that secondary OA (SOA) played a major role but were unable to resolve specific SOA sources. We address the latter through the first field deployment of an extractive electrospray ionization time-of-flight mass spectrometer (EESI-TOF) in Delhi, together with a high-resolution aerosol mass spectrometer (AMS). Measurements were conducted during the winter of 2018/19, and positive matrix factorization (PMF) was used separately on AMS and EESI-TOF datasets to apportion the sources of OA. AMS PMF analysis yielded three primary and two secondary factors which were attributed to hydrocarbon-like OA (HOA), biomass burning OA (BBOA-1 and BBOA-2), more oxidized oxygenated OA (MO-OOA), and less oxidized oxygenated OA (LO-OOA). On average, 40 % of the total OA mass was apportioned to the secondary factors. The SOA contribution to total OA mass varied greatly between the daytime (76.8 %, 10:00–16:00 local time (LT)) and nighttime (31.0 %, 21:00–04:00 LT). The higher chemical resolution of EESI-TOF data allowed identification of individual SOA sources. The EESI-TOF PMF analysis in total yielded six factors, two of which were primary factors (primary biomass burning and cooking-related OA). The remaining four factors were predominantly of secondary origin: aromatic SOA, biogenic SOA, aged biomass burning SOA, and mixed urban SOA. Due to the uncertainties in the EESI-TOF ion sensitivities, mass concentrations of EESI-TOF SOA-dominated factors were related to the total AMS SOA (i.e. MO-OOA + LO-OOA) by multiple linear regression (MLR). Aromatic SOA was the major SOA component during the daytime, with a 55.2 % contribution to total SOA mass (42.4 % contribution to total OA). Its contribution to total SOA, however, decreased to 25.4 % (7.9 % of total OA) during the nighttime. This factor was attributed to the oxidation of light aromatic compounds emitted mostly from traffic. Biogenic SOA accounted for 18.4 % of total SOA mass (14.2 % of total OA) during the daytime and 36.1 % of total SOA mass (11.2 % of total OA) during the nighttime. Aged biomass burning and mixed urban SOA accounted for 15.2 % and 11.0 % of total SOA mass (11.7 % and 8.5 % of total OA mass), respectively, during the daytime and 15.4 % and 22.9 % of total SOA mass (4.8 % and 7.1 % of total OA mass), respectively, during the nighttime. A simple dilution–partitioning model was applied on all EESI-TOF factors to estimate the fraction of observed daytime concentrations resulting from local photochemical production (SOA) or emissions (POA). Aromatic SOA, aged biomass burning, and mixed urban SOA were all found to be dominated by local photochemical production, likely from the oxidation of locally emitted volatile organic compounds (VOCs). In contrast, biogenic SOA was related to the oxidation of diffuse regional emissions of isoprene and monoterpenes. The findings of this study show that in Delhi, the nighttime high concentrations are caused by POA emissions led by traffic and biomass burning and the daytime OA is dominated by SOA, with aromatic SOA accounting for the largest fraction. Because aromatic SOA is possibly more toxic than biogenic SOA and primary OA, its dominance during the daytime suggests an increased OA toxicity and health-related consequences for the general public.

Vast stretches of agricultural land in southern and central Africa are burnt between June and September each year, which releases large quantities of aerosol into the atmosphere. The resulting smoke plumes are carried west over the Atlantic Ocean at altitudes between 2 and 4 km. As only limited observational data in West Africa have existed until now, whether this pollution has an impact at lower altitudes has remained unclear. The Dynamics-aerosol-chemistry-cloud interactions in West Africa (DACCIWA) aircraft campaign took place in southern West Africa during June and July 2016, with the aim of observing gas and aerosol properties in the region in order to assess anthropogenic and other influences on the atmosphere. Results presented here show that a significant mass of aged accumulation mode aerosol was present in the southern West African monsoon layer, over both the ocean and the continent. A median dry aerosol concentration of 6.2 µg m−3 (standard temperature and pressure, STP) was observed over the Atlantic Ocean upwind of the major cities, with an interquartile range from 5.3 to 8.0 µg m−3. This concentration increased to a median of 11.1 µg m−3 (8.6 to 15.7 µg m−3) in the immediate outflow from cities. In the continental air mass away from the cities, the median aerosol loading was 7.5 µg m−3 (5.9 to 10.5 µg m−3). The accumulation mode aerosol population over land displayed similar chemical properties to the upstream population, which implies that upstream aerosol is a significant source of aerosol pollution over the continent. The upstream aerosol is found to have most likely originated from central and southern African biomass burning. This demonstrates that biomass burning plumes are being advected northwards, after being entrained into the monsoon layer over the eastern tropical Atlantic Ocean. It is shown observationally for the first time that they contribute up to 80 % to the regional aerosol loading in the monsoon layer over southern West Africa. Results from the COSMO-ART (Consortium for Small-scale Modeling – Aerosol and Reactive Trace gases) and GEOS-Chem models support this conclusion, showing that observed aerosol concentrations over the northern Atlantic Ocean can only be reproduced when the contribution of transported biomass burning aerosol is taken into account. As a result, the large and growing emissions from the coastal cities are overlaid on an already substantial aerosol background. Simulations using COSMO-ART show that cloud droplet number concentrations can increase by up to 27 % as a result of transported biomass burning aerosol. On a regional scale this renders cloud properties and precipitation less sensitive to future increases in anthropogenic emissions. In addition, such high background loadings will lead to greater pollution exposure for the large and growing population in southern West Africa. These results emphasise the importance of including aerosol from across country borders in the development of air pollution policies and interventions in regions such as West Africa.

Low-level clouds (LLCs) cover a wide area of southern West Africa (SWA) during the summer monsoon months and have an important cooling effect on the regional climate. Previous studies of these clouds have focused on modelling and remote sensing via satellite. We present the first comprehensive set of in situ measurements of cloud microphysics from the region, taken during June–July 2016, as part of the DACCIWA (Dynamics–aerosol–chemistry–cloud interactions in West Africa) campaign. This novel dataset allows us to assess spatial, diurnal, and day-to-day variation in the properties of these clouds over the region. LLCs developed overnight and mean cloud cover peaked a few hundred kilometres inland around 10:00 local solar time (LST), before clouds began to dissipate and convection intensified in the afternoon. Regional variation in LLC cover was largely orographic, and no lasting impacts in cloud cover related to pollution plumes were observed downwind of major population centres. The boundary layer cloud drop number concentration (CDNC) was locally variable inland, ranging from 200 to 840 cm−3 (10th and 90th percentiles at standard temperature and pressure), but showed no systematic regional variations. Enhancements were seen in pollution plumes from the coastal cities but were not statistically significant across the region. A significant fraction of accumulation mode aerosols, and therefore cloud condensation nuclei, were from ubiquitous biomass burning smoke transported from the Southern Hemisphere. To assess the relative importance of local and transported aerosol on the cloud field, we isolated the local contribution to the aerosol population by comparing inland and offshore size and composition measurements. A parcel model sensitivity analysis showed that doubling or halving local emissions only changed the calculated cloud drop number concentration by 13 %–22 %, as the high background meant local emissions were a small fraction of total aerosol. As the population of SWA grows, local emissions are expected to rise. Biomass burning smoke transported from the Southern Hemisphere is likely to dampen any effect of these increased local emissions on cloud–aerosol interactions. An integrative analysis between local pollution and Central African biomass burning emissions must be considered when predicting anthropogenic impacts on the regional cloud field during the West African summer monsoon.

Water in the atmosphere can exist in the solid, liquid or gas phase. At high humidities, if the aerosol population remains constant, more water vapour will condense onto the particles and cause them to swell, sometimes up to several times their original size. This significant change in size and chemical composition is termed hygroscopic growth and alters a particle's optical properties. Even in unsaturated conditions, this can change the aerosol direct effect, for example by increasing the extinction of incoming sunlight. This can have an impact on a region's energy balance and affect visibility. Here, aerosol and relative humidity measurements collected from aircraft and radiosondes during the Dynamics–Aerosol–Chemistry–Cloud Interactions in West Africa (DACCIWA) campaign were used to estimate the effect of highly humid layers of air on aerosol optical properties during the monsoon season in southern West Africa. The effects of hygroscopic growth in this region are of particular interest due to the regular occurrence of high humidity and the high levels of pollution in the region. The Zdanovskii, Stokes and Robinson (ZSR) mixing rule is used to estimate the hygroscopic growth of particles under different conditions based on chemical composition. These results are used to estimate the aerosol optical depth (AOD) at λ=525 nm for 63 relative humidity profiles. The median AOD in the region from these calculations was 0.36, the same as that measured by sun photometers at the ground site. The spread in the calculated AODs was less than the spread from the sun photometer measurements. In both cases, values above 0.5 were seen predominantly in the mornings and corresponded with high humidities. Observations of modest variations in aerosol load and composition are unable to explain the high and variable AODs observed using sun photometers, which can only be recreated by accounting for the very elevated and variable relative humidities (RHs) in the boundary layer. Most importantly, the highest AODs present in the mornings are not possible without the presence of high RH in excess of 95 %. Humid layers are found to have the most significant impact on AOD when they reach RH greater than 98 %, which can result in a wet AOD more than 1.8 times the dry AOD. Unsaturated humid layers were found to reach these high levels of RH in 37 % of observed cases. It can therefore be concluded that the high AODs present across the region are driven by the high humidities and are then moderated by changes in aerosol abundance. Aerosol concentrations in southern West Africa are projected to increase substantially in the coming years; results presented here show that the presence of highly humid layers in the region is likely to enhance the consequent effect on AOD significantly.

A number of campaigns have been carried out to establish the emission factors of pollutants from fuel combustion in West Africa, as part of work package 2 (Air Pollution and Health) of the DACCIWA (Dynamics-Aerosol-Chemistry-Cloud Interactions in West Africa) FP7 program. Emission sources considered here include wood (hevea and iroko) and charcoal burning, charcoal making, open trash burning, and vehicle emissions, including trucks, cars, buses and two-wheeled vehicles. Emission factors of total particulate matter (TPM), elemental carbon (EC), primary organic carbon (OC) and volatile organic compounds (VOCs) have been established. In addition, emission factor measurements were performed in combustion chambers in order to reproduce field burning conditions for a tropical hardwood (hevea), and obtain particulate emission factors by size (PM0.25, PM1, PM2.5 and PM10). Particle samples were collected on quartz fiber filters and analyzed using gravimetric method for TPM and thermal methods for EC and OC. The emission factors of 58 VOC species were determined using offline sampling on a sorbent tube. Emission factor results for two species of tropical hardwood burning of EC, OC and TPM are 0.98 ± 0.46 g kg−1 of fuel burned (g kg−1), 11.05 ± 4.55 and 41.12 ± 24.62 g kg−1, respectively. For traffic sources, the highest emission factors among particulate species are found for the two-wheeled vehicles with two-stroke engines (2.74 g kg−1 fuel for EC, 65.11 g kg−1 fuel for OC and 496 g kg−1 fuel for TPM). The largest VOC emissions are observed for two-stroke two-wheeled vehicles, which are up to 3 times higher than emissions from light-duty and heavy-duty vehicles. Isoprene and monoterpenes, which are usually associated with biogenic emissions, are present in almost all anthropogenic sources investigated during this work and could be as significant as aromatic emissions in wood burning (1 g kg−1 fuel). EC is primarily emitted in the ultrafine fraction, with 77 % of the total mass being emitted as particles smaller than 0.25 µm. The particles and VOC emission factors obtained in this study are generally higher than those in the literature whose values are discussed in this paper. This study underlines the important role of in situ measurements in deriving realistic and representative emission factors.

As part of the Dynamics-Aerosol-Chemistry-Cloud Interactions in West Africa (DACCIWA) project, an airborne campaign was designed to measure a large range of atmospheric constituents, focusing on the effect of anthropogenic emissions on regional climate. The presented study details results of the French ATR42 research aircraft, which aimed to characterize gas-phase, aerosol and cloud properties in the region during the field campaign carried out in June/July 2016 in combination with the German Falcon 20 and the British Twin Otter aircraft. The aircraft flight paths covered large areas of Benin, Togo, Ghana and Côte d'Ivoire, focusing on emissions from large urban conurbations such as Abidjan, Accra and Lomé, as well as remote continental areas and the Gulf of Guinea. This paper focuses on aerosol particle measurements within the boundary layer ( Background concentrations (i.e. outside urban plumes) observed from the ATR42 indicate a fairly polluted region during the time of the campaign, with average concentrations of carbon monoxide of 131 ppb, ozone of 32 ppb, and aerosol particle number concentration ( > 15 nm) of 735 cm−3 stp. Regarding submicron aerosol composition (considering non-refractory species and black carbon, BC), organic aerosol (OA) is the most abundant species contributing 53 %, followed by SO4 (27 %), NH4 (11 %), BC (6 %), NO3 (2 %) and minor contribution of Cl (1 in the region was 5.9 µg m−3 stp. During measurements of urban pollution plumes, mainly focusing on the outflow of Abidjan, Accra and Lomé, pollutants are significantly enhanced (e.g. average concentration of CO of 176 ppb, and aerosol particle number concentration of 6500 cm−3 stp), as well as PM1 concentration (11.9 µg m−3 stp). Two classes of organic aerosols were estimated based on C-ToF-AMS: particulate organic nitrates (pONs) and isoprene epoxydiols secondary organic aerosols (IEPOX–SOA). Both classes are usually associated with the formation of particulate matter through complex interactions of anthropogenic and biogenic sources. During DACCIWA, pONs have a fairly small contribution to OA (around 5 %) and are more associated with long-range transport from central Africa than local formation. Conversely, IEPOX–SOA provides a significant contribution to OA (around 24 and 28 % under background and in-plume conditions). Furthermore, the fractional contribution of IEPOX–SOA is largely unaffected by changes in the aerosol composition (particularly the SO4 concentration), which suggests that IEPOX–SOA concentration is mainly driven by pre-existing aerosol surface, instead of aerosol chemical properties. At times of large in-plume SO4 enhancements (above 5 µg m−3), the fractional contribution of IEPOX–SOA to OA increases above 50 %, suggesting only then a change in the IEPOX–SOA-controlling mechanism. It is important to note that IEPOX–SOA constitutes a lower limit to the contribution of biogenic OA, given that other processes (e.g. non-IEPOX isoprene, monoterpene SOA) are likely in the region. Given the significant contribution to aerosol concentration, it is crucial that such complex biogenic–anthropogenic interactions are taken into account in both present-day and future scenario models of this fast-changing, highly sensitive region.

Particulate emissions from biomass burning can both alter the atmosphere's radiative balance and cause significant harm to human health. However, due to the large effect on emissions caused by even small alterations to the way in which a fuel burns, it is difficult to study particulate production of biomass combustion mechanistically and in a repeatable manner. In order to address this gap, in this study, small wood samples sourced from Côte D'Ivoire in West Africa were burned in a highly controlled laboratory environment. The shape and mass of samples, available airflow and surrounding thermal environment were carefully regulated. Organic aerosol and refractory black carbon emissions were measured in real time using an Aerosol Mass Spectrometer and a Single Particle Soot Photometer, respectively. This methodology produced remarkably repeatable results, allowing aerosol emissions to be mapped directly onto different phases of combustion. Emissions from pyrolysis were visible as a distinct phase before flaming was established. After flaming combustion was initiated, a black-carbon-dominant flame was observed during which very little organic aerosol was produced, followed by a period that was dominated by organic-carbon-producing smouldering combustion, despite the presence of residual flaming. During pyrolysis and smouldering, the two phases producing organic aerosol, distinct mass spectral signatures that correspond to previously reported variations in biofuel emissions measured in the atmosphere are found. Organic aerosol emission factors averaged over an entire combustion event were found to be representative of the time spent in the pyrolysis and smouldering phases, rather than reflecting a coupling between emissions and the mass loss of the sample. Further exploration of aerosol yields from similarly carefully controlled fires and a careful comparison with data from macroscopic fires and real-world emissions will help to deliver greater constraints on the variability of particulate emissions in atmospheric systems.