Global_Environmental_Research_Vol.27No.1
46/80

40 3.1 Carbonaceous Components 3. Chemical Compositions of Peatland Fire Aerosols March 2023. We identified 1,769 suitable articles in the databases. After removing duplicates and screening the titles and abstracts, we had 232 articles remaining. We reviewed the full-text articles using the study selection criteria mentioned in the paragraph above. Ultimately, we included 112 in our review, with 13 articles added through snowballing and nine review papers excluded. A comparison of the nine review articles is shown in Table S1 as described in Section 1, and the articles are included only the Supplementary found one newly published article after completing our screening process, and included it in our dataset. Lastly we created a dataset of the 113 articles using a Microsoft Excel file with 13 sheets, consisting of: 1) a cover sheet, 2) carbonaceous components, 3) water-soluble ions, 4) elements, 5) organic compounds excluding PAHs, 6) PAHs, 7) aerosol mass spectra analyses, 8) isotopes, 9) lab. experiments, 10) size distributions, 11) source apportionments, 12) a list of references and 13) an appendix. PM sampling information (during haze periods, and partly including non-haze periods), total PM mass concentration (or EF) and mass concentration of each species (EF or emission ratio) are contained in sheets 2) – 6). The contents of the sheets 2) – 13) are listed and described in the cover sheet. Species concentrations (or EF) are available for five countries: Indonesia, Brunei, Malaysia, Singapore and Thailand, but are missing for other countries in SEA. in the literature When haze was induced by IPFs, a significant increase in the concentrations of total carbon (TC), organic carbon (OC), water-soluble OC (WSOC) and carbon content of humic-like substances (HULIS-C) was typically observed at receptor sites as shown in the Supplementary Information (“Carbonaceous Components” sheet in the dataset). For example, the OC concentrations in the total suspended particulates (TSP) during a haze period in 1991 and a non-haze period in 1992 at Petaling Jaya in Malaysia were 74 and 14 μg m-3, respectively2. In Singapore, significantly higher OC concentrations were observed107 in the PM2.5 of smoke dominant samples (25 ± 7 μg m-3) compared to those of non-smoke dominant samples (5 ± 1 μg m-3) in 2012–2013 and 2015. In Thailand, the OC concentrations in PM2.5 during haze and non-haze periods in 2019 – 2020 were 10.1 and 2.1 ± 0.2 μg m-3, respectively120. Although the WSOC data are limited, there are some reports of an increase in WSOC in Malaysia105, concentrations during haze events Singapore91 and Thailand120,122. For HULIS-C, Fujii et al.105 reported that the HULIS-C concentrations during a strong haze event (17.5 μg m-3) at Bangi, Malaysia were higher than those during a non-haze period (1.3 – the dataset list of Information. We Y. FUJII and S. TOHNO 3.2 Water-Soluble Ions 4.4 μg m-3). In contrast to some of the above species such as OC and WSOC, the elemental carbon (EC) and black carbon (BC) concentrations showed no significant increases during haze events in Malaysia2, 50, 55, 56, 105, Singapore25, 43, 89, 91, 107 or Thailand120, 122. IPFs are smoldering fires undergoing a slow, low-temperature, flameless form of burning49, which produces much smaller EC (BC) emissions than OC emissions, as seen in the laboratory experiments in the dataset. Therefore, IPF-derived EC contributes less to non-haze levels of EC at to dilution during transportation. The chemical speciation of brown carbon (BrC) constituents is described in Section S4.3 of the Supplementary Information. in receptor Based During haze events induced by IPFs, a significant 2-) and increase in the concentrations of sulfate (SO4+) was typically observed at receptor ammonium (NH4sites. For example, significant differences in the total concentrations of water-soluble ions were observed between haze (33 ± 6.5 μg m-3) and non-haze (5.5 ± 2.6 μg m-3) periods in Pekanbaru, Indonesia85. They then + showed the average concentrations of SO4during haze events to have been 7.3 and 7.8 times higher than those during non-haze periods. Highly elevated according sites on laboratory chamber combustion experiments using Indonesian or Malaysian peat, the EFs of OC, WSOC and EC were determined as 6.02–18.5, 3.1–3.84, and 0.04 – 0.57 g kg-fuel-1, respectively17, 26, 90, 98. Iinuma et al.26 also reported EFs of water-insoluble OC (WIOC) as 4.9 g kg-fuel-1. Stockwell et al.63 and Jayarathne et al.77 presented EFs of OC and EC based on peatland fire source sampling in Central Kalimantan, Indonesia. Their EFs of OC (12.4 –16.0 g kg-fuel-1) and EC (0.24 g kg-fuel-1) were in the range of those based on peat burning experiments17, 26, 90, 98. Chow et al.83 and Watson et al.90, respectively, reported the species abundances in PM2.5 mass and EFs for carbonaceous components (OC, EC and WSOC) based on aerosol data from laboratory chamber combustion of fresh and aged Malaysian peat, separated by two- and seven-day photochemical aging times simulated with an oxidation flow reactor (below, “OFR”). The EF of OC for fresh particles was 17.9 – 18.2 g kg-fuel-1 and that for aged particles decreased to 14.0 – 14.7 g kg-fuel-1 90. In particular, reduction of OC1 (the lowest temperature OC fraction obtained from thermal-optical method, see the “Appendix” sheet in the dataset.) was revealed, due to the loss of high-vapor-pressure semi-volatile organic compounds upon aging. In contrast, the EF of WSOC, as well as mass 83, increased after oxidation percentage of WSOC in PM2.5in the OFR90. Based on the field observations in Malaysia, Fujii et al.106 mentioned a high possibility of secondary WSOC formation during transport from the IPF source to receptor sites. 2- and NH4

元のページ  ../index.html#46

このブックを見る