Carbonaceous aerosol composition in air masses influenced by large-scale biomass burning: a case study in northwestern Vietnam

Abstract

We investigated concentrations of organic carbon (OC), elemental carbon (EC), and a wide range of particle-bound organic compounds in daily sampled PM2.5 at the remote Pha Din (PDI) – Global Atmosphere Watch (GAW) monitoring station in northwestern Vietnam during an intense 3-week sampling campaign from 23 March to 12 April 2015. The site is known to receive trans-regional air masses during large-scale biomass burning (BB) episodes. BB is a globally widespread phenomenon and BB emission characterization is of high scientific and societal relevance. Emissions composition is influenced by multiple factors (e.g., fuel and thereby vegetation type, fuel moisture, fire temperature, available oxygen). Due to regional variations in these parameters, studies in different world regions are needed. OC composition provides valuable information regarding the health- and climate-relevant properties of PM2.5. Yet, OC composition studies from PDI are missing in the scientific literature to date. Therefore, we quantified 51 organic compounds simultaneously by in situ derivatization thermal desorption gas chromatography and time-of-flight mass spectrometry (IDTD-GC-TOFMS). Anhydrosugars, methoxyphenols, n-alkanes, fatty acids, polycyclic aromatic hydrocarbons, oxygenated polycyclic aromatic hydrocarbons, nitrophenols, and OC were used in a hierarchical cluster analysis highlighting distinctive patterns for periods under low, medium, and high BB influence. The highest particle phase concentration of the typical primary organic aerosol (POA) and possible secondary organic aerosol (SOA) constituents, especially nitrophenols, were found on 5 and 6 April. We linked the trace gas mixing ratios of methane (CH4), carbon dioxide (CO2), carbon monoxide (CO), and ozone (O3) to the statistical classification of BB events based on OA composition and found increased CO and O3 levels during medium and high BB influence. Likewise, a backward trajectory analysis indicates different source regions for the identified periods based on the OA clusters, with cleaner air masses arriving from the northeast, i.e., mainland China and the Yellow Sea. The more polluted periods are characterized by trajectories from the southwest, with more continental recirculation of the medium cluster and more westerly advection for the high cluster. These findings highlight that BB activities in northern Southeast Asia significantly enhance the regional organic aerosol loading and also affect the carbonaceous PM2.5 constituents and the trace gases in northwestern Vietnam. The presented analysis adds valuable data on the carbonaceous and chemical composition of PM2.5, in particular of OC, in a region of scarce data availability, and thus offers a reference dataset from Southeast Asian large-scale BB for future studies. Such a reference dataset may be useful for the evaluation of atmospheric transport simulation models, or for comparison with other world regions and BB types, such as Australian bush fires, African savannah fires, or tropical peatland fires.