-
Figure 1 shows the 72-h backward air mass trajectories in the four seasons. The SDZ is highly affected by northwest air masses transported from Mongolia and Inner Mongolia during autumn, winter, and spring. This site is also affected by southwestern air masses traveling over Datong (a significant coal production site) and urban Beijing in autumn. In spring, 21% of the air masses originate from the NCP. In summer, the site is influenced by the regions to the southeast (Shandong Province, Hebei Province, and Tianjin); 32% of the air masses originate from the nearby BTH metropolitan area, which is densely populated with high SO2 and NOx levels from agricultural and industrial activities (Xu et al., 2019; Khan et al., 2021; Su et al., 2021).
-
The statistics of the concentrations of carbonaceous components and WSIIs in PM2.5 at the SDZ are summarized in Table 1. TC was the sum of OC and EC, with an average of 6.3 ± 4.5 µg m–3. OC ranged from 0.66 µg m–3 to 23 µg m–3, with an average of 5.6 ± 4.0 µg m–3. EC ranged from 0.0 µg m–3 to 3.2 µg m–3, with an average of 0.70 ± 0.53 µg m–3. The average OC and EC concentrations at SDZ were approximately two to eight times lower than those in the adjacent cities within the NCP region (Beijing: 11 µg m–3 and 3.4 µg m–3; Tianjin: 12 µg m–3 and 3.1 µg m–3; Shijiazhuang: 23 µg m–3 and 5.4 µg m–3; Tangshan: 12 µg m–3 and 3.5 µg m–3, respectively) (Ji et al., 2019). The total WSIIs (
${\rm{NO}}_3^- $ ,${\rm{SO}}_4^{2-} $ , Cl−,${\rm{NO}}_2^- $ , F−,${\rm{PO}}_4^{3-} $ , MSA−,${\rm{NH}}_4^+ $ , K+, Na+, Ca2+, and Mg2+) varied from 1.1 µg m–3 to 118 µg m–3 with an average of 19 µg m–3. Secondary inorganic ions${\rm{NO}}_3^{-} $ ,${\rm{SO}}_4^{2-} $ , and${\rm{NH}}_4^+ $ were the major WSIIs, contributing 34%, 27%, and 19% to the total WSIIs mass, respectively.Components Autumn (n = 54) Winter (n = 89) Spring (n = 62) Summer (n = 23) Annual (n = 228) Range Mean ± SD Range Mean ± SD Range Mean ± SD Range Mean ± SD Range Mean ± SD Carbonaceous
Components
(µg m–3)OC 0.95–23 8.1±5.0 1.1–18 5.4±3.5 0.66–16 4.6±3.2 0.78–5.7 3.2±1.4 0.66–23 5.6±4.0 EC 0.09–3.3 1.0±0.73 0.11–1.8 0.66±0.41 0.03–2.0 0.57±0.40 0.0–0.68 0.39±0.19 0.0–3.2 0.70±0.53 TC 1.1–26 9.1±5.7 1.2–20 6.0±3.9 0.69–18 5.2±3.6 0.78–6.4 3.6±1.5 0.69–26 6.3±4.5 WSOC 0.28–12 3.4±2.6 0.66–8.4 2.1±1.6 0.51–10 2.7±2.0 0.36–4.7 2.3±1.1 0.28–12 2.6±2.0 WIOC 0.43–11 4.7±2.6 0.24–10 3.3±2.2 0.14–5.6 1.9±1.3 0.35–1.9 0.92±0.40 0.14–11 3.0±2.3 OM 1.3–32 11±7.0 1.5–25 7.5±5.0 0.92–22 6.4±4.5 1.1–8.0 4.5±1.9 0.92–32 7.8±5.6 Water-Soluble
Inorganic Ions
(µg m–3)${\rm{SO}}_4^{2-} $ 0.70–22 5.2±5.4 0.49–15 2.4±2.2 0.53–16 4.0±3.4 0.16–16 7.1±4.7 0.16–22 4.0±4.0 ${\rm{NO}}_3^- $ 0.28–65 14±18 0.18–25 4.3±6.0 0.25–60 11±13 0.1–10 3.4±3.4 0.10–65 8.4±12 ${\rm{NH}}_4^+ $ 0.14–25 5.7±6.8 0.19–14 2.3±2.9 0.0–25 4.4±5.4 0.18–10 4.0±2.5 0.0–25 3.8±5.0 K+ 0.03–1.5 0.44±0.34 0.03–3.2 0.31±0.39 0.02–1.1 0.29±0.23 0.0–0.23 0.08±0.06 0.0–3.2 0.31±0.33 Ca2+ BDL–4.7 0.78±1.0 BDL–2.3 0.05±0.25 BDL–0.61 0.06±0.13 BDL–0.81 0.02±0.17 BDL–4.7 0.23±0.60 Na+ 0.09–3.2 0.76±0.44 0.15–2.3 0.42±0.24 0.23–0.88 0.51±0.12 BDL–0.16 0.05±0.04 BDL–3.2 0.49±0.33 Cl− 0.19–4.2 1.2±0.84 0.0–3.8 0.95±0.81 0.09–2.3 0.65±0.51 BDL–0.33 0.09±0.11 BDL–4.2 0.85±0.78 MSA 0.0–0.39 0.06±0.07 0.0–0.12 0.01±0.02 0.0–0.07 0.01±0.01 0.0–0.04 0.01±0.01 0.0–0.39 0.02±0.04 ${\rm{NO}}_2^- $ 0.0–1.6 0.45±0.43 0.06–1.2 0.26±0.18 0.04–1.2 0.29±0.23 0.09–0.47 0.32±0.1 0.0–1.6 0.32±0.28 Mg2+ 0.0–0.48 0.12±0.14 0.0–0.64 0.01±0.07 0.0–0.19 0.02±0.03 0.0–0.13 0.04±0.03 0.0–0.64 0.05±0.09 ${\rm{PO}}_4^{3-} $ BDL–0.79 0.07±0.12 BDL–0.63 0.04±0.08 BDL–0.32 0.03±0.06 BDL–0.11 0.04±0.03 BDL–0.79 0.05±0.08 F− 0.0–0.14 0.02±0.03 0.0–0.06 0.01±0.01 0.0–0.03 0.0±0.0 0.0–0.0 0.0±0.0 0.0–0.14 0.01±0.02 Table 1. Seasonal variations in the mass concentrations of the chemical species in PM2.5 at SDZ. The detection limits for water-soluble inorganic ions are 0.001 µg m−3. BDL means below the detection limit.
The major chemical components were
${\rm{NO}}_3^- $ > OM >${\rm{SO}}_4^{2-} $ >${\rm{NH}}_4^+ $ > EC, according to their average concentrations (Fig. 2). The annual fraction of the three major ions (60%) was approximately twice that of the carbonaceous components (OM + EC: 32%). Consistent with other observations in the BTH region (Khan et al., 2021; Su et al., 2021),${\rm{NO}}_3^- $ was significantly higher than NH4+, likely because of some additional${\rm{NO}}_3^- $ sources. For example, the heterogeneous reaction HNO3 +NaCl→NaNO3 can provide an additional${\rm{NO}}_3^- $ in ambient aerosol particles (Wu et al., 2006).Figure 2. Contributions of the chemical species to total mass concentrations of PM2.5 (EC + OM + ∑anions + ∑cations) at the SDZ site: (a) autumn, (b) winter, (c) spring, (d) summer, and (e) the annual average. The minor cations include Ca2+ and Mg2+, and the minor anions include
${\rm{NO}}_2^- $ , F−,${\rm{PO}}_4^{3-} $ , and MSA−.Figure 3 and Table 1 show that the chemical components exhibited seasonal variations. All the carbonaceous components and most of the ions had the highest concentrations in autumn compared to other seasons. This result is inconsistent with the previous study at SDZ, in which the highest concentration was observed during spring or winter (Li et al., 2016). However, the
${\rm{SO}}_4^{2-} $ concentration peaked in summer, followed by autumn, and reached its lowest level in winter. The${\rm{NO}}_3^- $ concentration in autumn was about four times higher than that in summer. The seasonal contribution of each component to the total aerosol concentration (EC + OM + ∑cations +∑anions) is shown in Fig. 2.${\rm{NO}}_3^- $ was a major contributor, accounting for 39% in spring and 35% in autumn, followed by OM, accounting for > 20% in all seasons and as high as 39% in winter.${\rm{SO}}_4^{2-} $ was a major contributor in summer with a contribution of 35%, but it contributed only 12% in autumn and winter.Figure 3. Time series of the chemical composition concentrations in PM2.5 at the SDZ site. Additional data for the minor ions are shown in Fig. S5 in the ESM.
Secondary ions (
${\rm{NO}}_3^- $ ,${\rm{SO}}_4^{2-} $ , and NH4+) strongly correlated with each other (r ≥ 0.71, p ≤ 0.0001) during the observation period, except for${\rm{NO}}_3^- $ and${\rm{SO}}_4^{2-} $ during summer (r = 0.31) (Table S3 in the ESM). This suggests that these three major secondary inorganics have the same source or come from a similar reaction generation process. In all seasons, WSOC,${\rm{SO}}_4^{2-} $ ,${\rm{NO}}_3^- $ , and NH4+ positively correlated with EC (r ≥ 0.52, p ≤ 0.0001), indicating that primary anthropogenic activities like burning fossil fuel and biomass are strong sources of the precursors to secondary organic and inorganic components (Pavuluri et al., 2011). In the following, the sources of organic matter are further discussed in the PMF analysis.From the PMF source apportionment, six factors were resolved including two secondary formation factors, biomass burning, fossil fuel combustion, industry, and dust (Fig. S1). Figure 4 shows the fractional contributions of different factors to the mass concentrations of WSOM and WIOM. Annually, secondary formation (52%) and fossil fuel combustion (63%) are the largest sources of WSOM and WIOM, respectively. Fossil fuel combustion also contributes 1/3 of WSOM. Biomass burning is also nonnegligible, accounting for ~10% of both WSOM and WIOM. Seasonally, the secondary formation can represent ~90% of WSOM in summer, and even in winter it is ~1/3. Fossil fuel combustion is the dominant source of WIOM in each season, with a contribution between ~ 45% to 75%, while in summer, the two secondary factors as a whole are counterpart contributors to WIOM. Except in summer, biomass burning contributes to around 10%–15% of both WSOM and WIOM.
Figure 4. Modeled concentrations (a) and fractional contributions (b) of different factors to the mass concentrations of WSOM and WIOM. These six source factors are resolved by the PMF analysis. The measurement results for the seasonal and annual mean concentration of WSOM and WIOM (as in Table 1) are also shown in subplot (a) as circles. Major contributions (> 10%) in the annual average are marked in the plots.
As shown by the source apportionment of WSOM and WIOM, the concentration of WSOM in winter and summer is similar. But the concentration of WIOM is about four times higher in winter than in summer. Thus, the WSOM/OM ratio decreased from summer to winter, from ~0.7 to ~0.4. Besides, the increase of WIOM from summer to winter is largely driven by fossil fuel combustion and biomass burning, totally contributing to ~90% of WIOM in winter.
Table S4 in the ESM compares the concentrations of carbonaceous components and major WSIIs between the measurements performed at the SDZ (this study and previous measurements dating back to a decade ago) and other GAW atmosphere background stations. The summary included global background stations at Mt. Waliguanshan (WLG, 36°17′N, 100°54′E, 3816 m a.s.l.) and Mt. Cimone (CMN, 44°12′N, 10°42′E, 2165 m a.s.l.) and regional background stations in Akdala located in northwestern China (AKD, 47°06′N, 87°58′E, 562 m a.s.l.), Lin' an in eastern China (LA, 38°18′N, 119°44′E, 131 m a.s.l.), and Mt. Longfengshan in northeastern China (LFS, 44°44′N, 127°36′E, 331 m a.s.l.). First, the annual average concentrations of carbonaceous components (OC and EC) at the SDZ decreased from 2009 to 2018. A remarkable downward atmospheric deposition of OC was also observed in rural regions of the NCP during 2016–20 (Cao et al., 2022). In addition, the concentration of the major ions (
${\rm{SO}}_4^{2-} $ ,${\rm{NO}}_3^- $ ,${\rm{NH}}_4^+ $ , and Cl−) decreased from 2009 to 2015. From 2015 to 2018, the${\rm{SO}}_4^{2-} $ concentration decreased,${\rm{NH}}_4^+ $ concentration was stable, and${\rm{NO}}_3^- $ and Cl- concentrations increased (Yan et al., 2012a, b; Zhao et al., 2013a; Li et al., 2016). Second, comparing these background stations, for each season, the concentrations of the carbonaceous components and ions were considerably higher at the regional background stations (AKD, LA, LFS, and SDZ) than at the global stations (WLG and CMN) (Yang et al., 1996; Qu et al., 2008, 2009; Zhao et al., 2013a; Carbone et al., 2014; Li et al., 2015), suggesting that global background stations are less affected by anthropogenic emissions than the regional counterparts. -
The average WSOC concentration was 2.6 µg m–3 (0.28–12 µg m–3), accounting for 48% of the OC; WIOC represented the other half (52%; average: 3.0 µg m−3) (Table 1). The contribution of WSOM to OM was highest in summer (70%), followed by spring (60%) (Table 2). The measurement of the EEM spectra together with the post-PARAFAC analysis was used to characterize the chromophores in WSOM (Chen et al., 2016b; Chen et al., 2017; Yan and Kim, 2017). Fluorescence intensity as a semi-quantitative measure of the abundance of fluorescent chromophores in atmospheric samples (Yan and Kim, 2017; Wu et al., 2021a; Chen et al., 2022) is reported and used in correlation analyses.
Parameter Autumn (n = 54) Winter (n = 89) Spring (n = 62) Summer (n = 23) Annual (n = 228) Range Mean ± SD Range Mean ± SD Range Mean ± SD Range Mean ± SD Range Mean± SD OC/EC 5.5–22 8.7±2.6 5.9–17 8.2±1.6 6.2–20 8.4±2.1 4.2–14 8.8±2.5 4.2–22 8.4±2.1 EC/TC 0.04–0.15 0.11±0.02 0.06–0.14 0.11±0.02 0.05–0.14 0.11±0.02 0.0–0.19 0.10±0.04 0.0–0.19 0.11±0.02 WSOC/OC* 0.14–0.60 0.40±0.09 0.21–0.78 0.40±0.11 0.44–0.82 0.60±0.09 0.46–0.82 0.70±0.10 0.14–0.82 0.48±0.15 ∑+/∑− 0.92–1.3 1.1±0.12 0.76–1.7 0.99±0.14 0.06–1.4 1.0±0.25 0.51–1.3 1.1±0.14 0.06–1.7 1.0±0.18 ${\rm{NO}}_3^- $/${\rm{SO}}_4^{2-} $ 0.21–5.4 2.2±1.5 0.14–4.5 1.4±1.1 0.23–6.9 2.3±1.5 0.04–1.7 0.55±0.51 0.04–6.9 1.8±1.4 K+/EC 0.10–1.0 0.44±0.20 0.12–4.6 0.46±0.55 0.13–2.0 0.52±0.28 0.08–0.37 0.20±0.08 0.08–4.6 0.45±0.40 HIX 1.0–3.7 2.2 ± 0.63 0.96–2.8 1.6 ± 0.52 1.6–4.3 2.6 ± 0.61 1.8–7.7 3.4 ± 1.2 0.96–7.7 2.2 ± 0.89 FI 1.2–1.5 1.4 ± 0.08 1.2–1.6 1.4 ± 0.09 1.2–1.4 1.3 ± 0.05 1.2–1.4 1.3 ± 0.04 1.2–1.6 1.4 ± 0.09 BIX 0.75–1.4 1.0 ± 0.17 0.99–1.4 1.2 ± 0.12 0.82–1.3 1.0 ± 0.11 0.71–1.0 0.87 ± 0.10 0.71–1.4 1.1 ± 0.18 * WSOC/OC = WSOM/OM. Table 2. Seasonal variations in the ratios between the species and the fluorescent indices in PM2.5 at SDZ.
In this study, we used the EEM-PARAFAC method to identify and quantify the chromophoric components of PM2.5. Four fluorescence components (C1, C2, C3, and C4) were resolved (Fig. 5a). The identities of the chromophores were determined by EEM spectra regarding those in previous reports (Fu et al., 2015; Yue et al., 2016; Chen et al., 2019; Wu et al., 2019; Zhang et al., 2021). C1 (ex/em = 240 nm/396 nm), C2 (ex/em = 255 nm/472 nm), and C3 (ex/em = 245 nm/390 nm), with typical fluorescence peaks, are identified as humic-like substances (HULIS) with previous reports (Fu et al., 2015; Yue et al., 2016; Chen et al., 2019) and are denoted as HULIS-2, HULIS-1, and HULIS-3, respectively. HULIS-1 chromophores with long emission wavelengths originate from large conjugated systems that may contain heteroatoms in their highly aromatic conjugate structures (Chen et al., 2016a). According to Chen et al. (2016b), a HULIS component with a longer emission wavelength is potentially more oxygenated, suggesting secondary sources. This is indicated by the strong correlations of HULIS-1 and -2 with secondary formation factors, which is absent for HULIS-3 (Table 3). Besides, the burning factors (biomass burning and/or fossil fuel combustion) and their tracers (EC, Cl-, and K+) also correlate well with these three fluorescent components (Table 3 and Table S3), suggesting these primary emissions are also important sources to them.
Figure 5. Fluorescent properties of the water-soluble organic matter (WSOM) in PM2.5 at the SDZ site. (a) Excitation–emission matrix (EEM) spectrum fingerprints of the identified fluorescent components are resolved by the PARAFAC model. The excitation and emission wavelengths of the peak intensity of each component are shown in the plots. The four components, C1, C2, C3, and C4, are named HULIS-2, HULIS-1, PLOM, and HULIS-3, respectively. (b) Seasonal variations in the fluorescence intensities of the four components and their relative fractions. (c) Seasonal variations in the humification index (HIX), fluorescence index (FI), and biological index (BIX).
Source factors HULIS-1 HULIS-2 HULIS-3 PLOM HIX FI BIX Autumn Secondary formation-1 0.62 0.29 −0.11 −0.08 0.57 −0.44 −0.57 Secondary formation-2 0.35 0.11 −0.18 −0.15 0.51 −0.35 −0.53 Biomass burning 0.63 0.63 0.32 0.28 0.05 −0.11 −0.01 Fossil fuel combustion 0.66 0.71 0.70 0.68 −0.11 0.14 0.20 Dust 0.51 0.13 −0.06 −0.08 0.36 −0.31 −0.35 Industry −0.02 −0.12 −0.05 −0.08 0.22 −0.18 −0.29 Winter Secondary formation-1 0.89 0.80 0.31 0.14 0.60 −0.54 −0.56 Secondary formation-2 0.58 0.49 0.05 −0.10 0.60 −0.47 −0.50 Biomass burning 0.79 0.68 0.28 0.14 0.51 −0.46 −0.51 Fossil fuel combustion 0.60 0.75 0.80 0.77 −0.16 0.09 0.07 Dust 0.02 0.03 0.04 0.07 −0.04 0.01 −0.02 Industry −0.19 −0.12 0.08 0.19 −0.31 0.30 0.25 Spring Secondary formation-1 0.87 0.74 0.21 0.35 0.66 −0.42 −0.44 Secondary formation-2 0.73 0.57 0.07 0.20 0.70 −0.45 −0.54 Biomass burning 0.71 0.79 0.74 0.81 0.00 −0.06 0.10 Fossil fuel combustion 0.42 0.33 0.18 0.25 0.24 −0.18 −0.21 Dust 0.36 0.28 0.24 0.20 0.12 −0.15 −0.11 Industry −0.55 −0.59 −0.38 −0.46 −0.22 0.22 0.09 Summer Secondary formation-1 0.87 0.82 −0.04 0.76 −0.01 −0.33 −0.42 Secondary formation-2 0.77 0.67 −0.19 0.37 0.14 −0.39 −0.87 Biomass burning 0.79 0.70 −0.08 0.58 0.12 −0.45 −0.44 Fossil fuel combustion −0.14 −0.01 0.03 0.01 −0.24 0.40 0.22 Dust −0.21 −0.17 0.21 −0.20 0.00 0.15 0.12 Industry −0.21 −0.18 0.16 −0.25 0.01 0.15 0.05 Notes: Autumn: t-test is p ≤ 0.0001 for the correction where r is ≥ 0.51. Winter: t-test is p < 0.0001 for the correction where r is ≥ 0.51. Spring: t-test is p < 0.0001 for the correction where r is ≥ 0.57. Summer: t-test is p < 0.0001 for the correction where r is > 0.67; t-test is 0.0001 < p < 0.01 for the correction where r is from 0.58 to 0.67. Table 3. Correlation coefficients (r) between the six source factors of WSOM and the fluorescent components and the indices measured in the water extracts in PM2.5 at the Shangdianzi (SDZ) site in (a) autumn, (b) winter, (c) spring, and (d) summer. Correlation coefficients r ≥ 0.50 or r ≤ –0.50 are highlighted in bold.
C4 (ex/em = 270 nm/336 nm) is termed PLOM because its EEM spectra are similar to those of amino acids (Ohno and Bro, 2006; Wu et al., 2019; Zhang et al., 2021). But it should be noted that nitrogen-free nonbiological components (non-proteinaceous matter) can also contribute to C4 (Coble, 2007; Chen et al., 2016b, 2021). Here, we did not resolve a source factor for primary bioaerosols, which may be due to a relatively low contribution of bioaerosols in PM2.5. It also should be noted that the biomass burning factor and the dust factor can also include co-emitted bioaerosols (Yue et al., 2019, 2022a), although the impact might be minor here. A higher correlation of PLOM with fossil fuel combustion (autumn and winter) and the biomass burning factor (spring and summer) reflect the non-biological contribution (Table 3). The fluorescence intensity of PLOM in both summer and spring is lower than in winter (Fig. 5b) and also reflects a significant non-biological contribution.
The fluorescence intensities of these four fluorescent components in PM2.5 greatly varied between seasons (Table S2 and Fig. 5b). All these four fluorescent components exhibited similar seasonal variations, with the highest intensity in winter/autumn and the lowest in summer. On the other hand, the concentration of WSOM in winter and summer is similar, but with a dominant secondary source in summer while the main contributions of fossil fuel and biomass burning occurred in winter (Fig. 4). Collectively, these results indicate that the fluorescence components are more abundant (or of higher intensity) per mass of WSOM from the burning sources than from the secondary formation. In summer, the contributions of HULIS-2 (49%, Fig. 5b) and HULIS-1 (31%, Fig. 5b) to total fluorescence intensity are high, which is due to two factors: 1) secondary source as a major source of them (Table 3); and 2) the dominance of secondary formation in the contribution to WSOM (~90%, Fig. 4). Whereas in winter, the relative contributions of PLOM (34%) and HULIS-3 (28%) to total fluorescence intensity increase compared to the summer condition (Fig. 5b), which is because: 1) fossil fuel combustion is their main source; and 2) the contribution of fossil fuel combustion in WSOM is dominant (50%, Fig. 4). With the relative increase of PLOM and HULIS-3, the fractional contributions of HULIS-2 and HULIS-1 to total fluorescence intensity cannot be as much as in winter.
-
To further investigate the potential sources of chromophoric components in WSOM, fluorescence indices were calculated based on the ratios of fluorescence intensity in specific spectral regions (section 2.3). Three fluorescence indices are typically used to track the origins, transformation, and chemical processing of atmospheric aerosols: HIX, FI, and BIX (Fu et al., 2015; Xie et al., 2016; Yue et al., 2016, 2019; Tang et al., 2021). The HIX, FI, and BIX variations are plotted in Fig. 5c and presented in Table 2.
The HIX is an index used to evaluate the maturation of humification (Zsolnay et al., 1999; Fu et al., 2015). A high HIX (>10) indicates heavily humified or aromatic organics, primarily of terrestrial origin, whereas a low HIX (< 4.0) indicates OM of microbial origin (Zsolnay et al., 1999; McKnight et al., 2001; Ohno and Bro, 2006; Huguet et al., 2009; Lee et al., 2013; Fu et al., 2015; Qin et al., 2018; Yue et al., 2019). HIX ranged from 0.96 to 7.7 with an average of 2.2 ± 0.89, which was lower than that in urban aerosols (7.1) (Xie et al., 2020) but comparable to that at a suburb site (1.6 in winter) (Yi et al., 2020), in forest environments (2.4 both in autumn and spring) (Xie et al., 2016), and at a mountain site (2.4 and 1.1 for fine and coarse aerosols, respectively) (Yue et al., 2019) (Table 2). The low annual average HIX indicates that fluorophores in water-soluble PM2.5 are primarily contributed by microbial-derived OM. As presented in Table 2 and Fig. 5c, the HIX exhibits seasonal variations in the following order: summer (average: 3.4) > spring (2.6) > autumn (2.2) > winter (1.6). This indicates that compared to summer, atmospheric chromophores tend to be either less humified or less aromatic in winter, which may be associated with the higher fraction of secondary formation in winter (Fig. 4 and Table 3).
BIX is an index used to measure the contribution of biological sources (Huguet et al., 2009; Xie et al., 2020). The increase in BIX corresponds to the enhanced contribution of microbially derived organics (Fu et al., 2015). A high BIX (> 1.0) indicates a significant biological or WSOM of microbial origin, whereas a low BIX (< 0.60) suggests a scarce biological origin (Huguet et al., 2009; Fu et al., 2015; Yue et al., 2016, 2019). As presented in Table 2, BIX varies between 0.71 and 1.4 with an average of 1.1 ± 0.18, comparable to that in Xi’an suburbs (1.2–1.4 in winter) (Yi et al., 2020). BIX was slightly higher in SDZ than in Nanjing urban areas (0.88) (Xie et al., 2020). A high BIX implies a predominately biological PM2.5 source. The highest and lowest BIX occurred in winter (1.2 ± 0.12) and summer (0.87 ± 0.10), respectively (Table 2). This is consistent with the magnitude of HIX (low value corresponds to high microbial sources) and another FI index (Table 2 and Fig. 5c). A higher FI also indicates more microbial sources and lower aromaticity (McKnight et al., 2001; Fu et al., 2015; Yue et al., 2016; Chen et al., 2021). The relatively low biological contribution to the fluorophores in the WSOM in summer is likely because the fraction of SOA in the WSOM is more enhanced in this season, as suggested by the strong negative correlation of BIX with secondary factor (Table 3) and the dominance of this factor in the WSOM in summer (Fig. 4).
Components | Autumn (n = 54) | Winter (n = 89) | Spring (n = 62) | Summer (n = 23) | Annual (n = 228) | ||||||||||
Range | Mean ± SD | Range | Mean ± SD | Range | Mean ± SD | Range | Mean ± SD | Range | Mean ± SD | ||||||
Carbonaceous Components (µg m–3) | OC | 0.95–23 | 8.1±5.0 | 1.1–18 | 5.4±3.5 | 0.66–16 | 4.6±3.2 | 0.78–5.7 | 3.2±1.4 | 0.66–23 | 5.6±4.0 | ||||
EC | 0.09–3.3 | 1.0±0.73 | 0.11–1.8 | 0.66±0.41 | 0.03–2.0 | 0.57±0.40 | 0.0–0.68 | 0.39±0.19 | 0.0–3.2 | 0.70±0.53 | |||||
TC | 1.1–26 | 9.1±5.7 | 1.2–20 | 6.0±3.9 | 0.69–18 | 5.2±3.6 | 0.78–6.4 | 3.6±1.5 | 0.69–26 | 6.3±4.5 | |||||
WSOC | 0.28–12 | 3.4±2.6 | 0.66–8.4 | 2.1±1.6 | 0.51–10 | 2.7±2.0 | 0.36–4.7 | 2.3±1.1 | 0.28–12 | 2.6±2.0 | |||||
WIOC | 0.43–11 | 4.7±2.6 | 0.24–10 | 3.3±2.2 | 0.14–5.6 | 1.9±1.3 | 0.35–1.9 | 0.92±0.40 | 0.14–11 | 3.0±2.3 | |||||
OM | 1.3–32 | 11±7.0 | 1.5–25 | 7.5±5.0 | 0.92–22 | 6.4±4.5 | 1.1–8.0 | 4.5±1.9 | 0.92–32 | 7.8±5.6 | |||||
Water-Soluble Inorganic Ions (µg m–3) | ${\rm{SO}}_4^{2-} $ | 0.70–22 | 5.2±5.4 | 0.49–15 | 2.4±2.2 | 0.53–16 | 4.0±3.4 | 0.16–16 | 7.1±4.7 | 0.16–22 | 4.0±4.0 | ||||
${\rm{NO}}_3^- $ | 0.28–65 | 14±18 | 0.18–25 | 4.3±6.0 | 0.25–60 | 11±13 | 0.1–10 | 3.4±3.4 | 0.10–65 | 8.4±12 | |||||
${\rm{NH}}_4^+ $ | 0.14–25 | 5.7±6.8 | 0.19–14 | 2.3±2.9 | 0.0–25 | 4.4±5.4 | 0.18–10 | 4.0±2.5 | 0.0–25 | 3.8±5.0 | |||||
K+ | 0.03–1.5 | 0.44±0.34 | 0.03–3.2 | 0.31±0.39 | 0.02–1.1 | 0.29±0.23 | 0.0–0.23 | 0.08±0.06 | 0.0–3.2 | 0.31±0.33 | |||||
Ca2+ | BDL–4.7 | 0.78±1.0 | BDL–2.3 | 0.05±0.25 | BDL–0.61 | 0.06±0.13 | BDL–0.81 | 0.02±0.17 | BDL–4.7 | 0.23±0.60 | |||||
Na+ | 0.09–3.2 | 0.76±0.44 | 0.15–2.3 | 0.42±0.24 | 0.23–0.88 | 0.51±0.12 | BDL–0.16 | 0.05±0.04 | BDL–3.2 | 0.49±0.33 | |||||
Cl− | 0.19–4.2 | 1.2±0.84 | 0.0–3.8 | 0.95±0.81 | 0.09–2.3 | 0.65±0.51 | BDL–0.33 | 0.09±0.11 | BDL–4.2 | 0.85±0.78 | |||||
MSA | 0.0–0.39 | 0.06±0.07 | 0.0–0.12 | 0.01±0.02 | 0.0–0.07 | 0.01±0.01 | 0.0–0.04 | 0.01±0.01 | 0.0–0.39 | 0.02±0.04 | |||||
${\rm{NO}}_2^- $ | 0.0–1.6 | 0.45±0.43 | 0.06–1.2 | 0.26±0.18 | 0.04–1.2 | 0.29±0.23 | 0.09–0.47 | 0.32±0.1 | 0.0–1.6 | 0.32±0.28 | |||||
Mg2+ | 0.0–0.48 | 0.12±0.14 | 0.0–0.64 | 0.01±0.07 | 0.0–0.19 | 0.02±0.03 | 0.0–0.13 | 0.04±0.03 | 0.0–0.64 | 0.05±0.09 | |||||
${\rm{PO}}_4^{3-} $ | BDL–0.79 | 0.07±0.12 | BDL–0.63 | 0.04±0.08 | BDL–0.32 | 0.03±0.06 | BDL–0.11 | 0.04±0.03 | BDL–0.79 | 0.05±0.08 | |||||
F− | 0.0–0.14 | 0.02±0.03 | 0.0–0.06 | 0.01±0.01 | 0.0–0.03 | 0.0±0.0 | 0.0–0.0 | 0.0±0.0 | 0.0–0.14 | 0.01±0.02 |