| Secondary organic aerosols (SOA) have great influences on visibility, climate, temperature and human health; they have captured more and more attention due to their harmfulness. Fine particle (PM2.5) samples, collected during summer field campaigns in Sichuan Panzhihua Cycad National Nature Reserve (PZH), Gonggashan National Nature Reserve (GGS) and Wolong National Nature Reserve (WL), were extracted with mixed solution, derivatized with N,O-bis-(trimethylsilyl) trifluoroacetamide (BSTFA) and analyzed by gas chromatography/mass spectrometry (GC/MS) to measure SOA tracers. Emphasis was placed on chemical composition, concentration levels and time series and diel variations of polar organic tracers. The results were as follows:(1) The 24h average concentrations of total isoprene oxidation products were 51.2,109.5 and 61.7 ng·m-3, of α-/β-pinene oxidation products were 16.1,1.5 and 6.5 ng·m-3, of short chain carboxylic acids (malic acid and 2-hydroxylglutaric acid) were 17.3,4.3 and 35.0 ng·m-3, of (3-caryophyllinic acid 1.7,0.09 and 0.11 ng-m"3, of sugar alcohols and sugars were107.5,36.7 and 65.6 ng·m-3, of anhydrosugars were 120.0,16.0 and 102.0 ng·m-3, of organic carbon (OC) were 20.3,2.9 and 9.3μg·m-3 at PZH, GGS and WL, respectively.(2) Isoprene and monoterpenes are major SOA precursors. In this study, excellent correlation were observed between 2-methylthreitol and 2-methylerythritol, similarly, the three isomeric C5- alkene triols, i.e. cis-2-methyl-1,3,4-trihydroxy-1-butene,3-methyl-2, 3,4-trihydroxy-l-butene, trans-2-methyl-1,3,4-trihydroxy-l-butene were highly correlated with the C5-alkene triols at PZH, GGS and WL. The photooxidation products of α-/β-pinene:3-hydroxyglutaric acid,3-hydroxy-4,4-dimethylglutaric acid and 3-methyl-1, 2,3-butane-tricarboxylic acid were remarkably correlated with each other for 24h samples at PZH and GGS; however, only 3-hydroxyglutaric acid and 3-hydroxy-4,4-dimethyl-glutaric acid showed weak correlation with each other for 24h samples at WL, this may be explained by low values of photosynthetic active radiation in sampling period at WL.(3) Diel variations with the highest concentrations during day-time were observed for 2-methyltetrols, inversely,2-methylglyceric acid and C5-alkene triols exhibited day-night differences with maxima during night at PZH; diurnal variations with the highest concentrations at day-time were observed for photooxidation products of isoprene at WL and GGS. But relative to WL, isoprene oxidation products did not present obvious diurnal differences at GGS; in an other word, the average day-time and night-time concentrations for isoprene oxudation products were relatively close. The possible explanation for this phenomenon could be that the altitude was relatively high and the influence by movement of the planetary boundary layer was significant at GGS. In addition, α-/β-pinene oxidation products exhibited unconsistent day/night variations, the ratio of fresh oxidation product (norpinic acid) to total α-/β-pinene oxidation products were 0.13,0.15 and 0.77 at PZH, GGS and WL, respectively. The differences of vegetation types and environmental factors had great influences on the species distribution and diel variation trend of the formation of α-/β-pinene oxidation products.(4) The isoprene,α-/β-pinene and β-caryophyllene oxidation products account for 1.6%, 0.3% and 0.4% of OC for 24h aerosol samples at PZH, for 24.4%,0.2% and 0.1% at GGS, for 4.3%,0.3% and 0.05% at WL. In addition, biomass burning and fungal spores account for 6.5% and 0.7% of OC at PZH, for 5.5% and 3.4% at GGS, for 11.1% and 1.8% at WL for 24h aerosol samples, respectively. In this study, based on the concentrations of molecular marker, we calculated the sources of organic matter in research aerosols. The sources included isoprene, α-/β-pinene and β-caryophyllene oxidation products, biomass burning and fungal spores. These sources accunted for 9.5%,33.6% and 17.5% of OC at PZH, GGS and WL, respectively.(5) The altitude was high at GGS and the sampling site was far away from residential areas, aerosol samples were less affected by human activities. The concentrations of OC and EC (0.4μg·m-3) were low, natural source of secondary organic aerosols had high proportion, the concentration of marker compound of biomass burning—levoglucosan was low (13.0 ng·m-3). GGS could be served as a good background sampling point. |