abundance and diurnal trends of fluorescent bioaerosols in ... yue et... · software: siyao yue,...

Abundance and Diurnal Trends of FluorescentBioaerosols in the Troposphere over Mt. Tai,China, in SpringSiyao Yue1,2,3 , Lujie Ren2, Tianli Song1 , Linjie Li1,3, Qiaorong Xie1,3, Weijun Li4 ,Mingjie Kang5,1, Wanyu Zhao1,3, Lianfang Wei1,3, Hong Ren1,3, Yele Sun1,3 , Zifa Wang1,3 ,Robert Mark Ellam2,6 , Cong‐Qiang Liu2, Kimitaka Kawamura7 , and Pingqing Fu2,1,3

1State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry, Institute of AtmosphericPhysics, Chinese Academy of Sciences, Beijing, China, 2Institute of Surface‐Earth System Science, Tianjin University,Tianjin, China, 3College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing, China,4Department of Atmospheric Sciences, School of Earth Sciences, Zhejiang University, Hangzhou, China, 5State KeyLaboratory of Water Environment Simulation, School of Environment, Beijing Normal University, Beijing, China,6Scottish Universities Environmental Research Centre, Glasgow, Scotland, UK, 7Chubu Institute for Advanced Studies,Chubu University, Kasugai, Japan

Abstract Primary biological aerosol particles are ubiquitous in the global atmosphere and can affectcloud formation, deteriorate air quality, and cause human infections. Mt. Tai (1,534 m a.s.l.) is an elevatedsite in the North China Plain where atmospheric aerosols reflect both regional advection and long‐rangetransport. In this study, we deployed a Wideband Integrated Bioaerosol Sensor (WIBS‐4A) and collectedtotal suspended particles and eight‐stage size‐segregated aerosol samples at the summit of Mt. Tai in springfrom 21 March to 8 April 2017 to quantify the abundance, size distributions, and diurnal variations offluorescent bioaerosols and to investigate the effect of different fluorescence thresholds of WIBS for ambientbioaerosol recognition. During the whole sampling period, the number concentration of fluorescentparticles (>0.8 μm) was 647 ± 533 L−1, accounting for 26.9% ± 10.0% by number of the total particles in thatsize range. Three‐dimensional excitation‐emission matrix fluorescence of water‐soluble organic matter insize‐segregated aerosols shows that humic‐like substances (HULIS) are mainly in the fine mode (<2.1 μm)while protein‐like substances are mainly in the coarse mode (>2.1 μm). From the diurnal variation, it isshown that bioaerosols can undergo transformation during long‐range transport and contribute to HULIS.For bioaerosol recognition, we find that 6σ‐threshold can lead to better classification of fluorescent aerosolparticles for fungal spores. Overall, our results constrain the abundance of primary bioaerosols in thetroposphere over East Asia and elucidate the processes for their evolution via mountain/valley breezes andlong‐range transport.

1. Introduction

Primary biological aerosol particles (PBAP) that are emitted directly from the biosphere contribute a largefraction of the atmospheric particulate mass (Després et al., 2012; Pöschl, 2005; Pöschl et al., 2010). PBAPcomprise animal and plant debris, pollen grains, fungal spores, bacteria, virus, proteins, and other materialsof biological origin (Després et al., 2012; Xie et al., 2017). Estimation of terrestrial PBAP emission is stillhighly uncertain, ranging from 56 to 1,000 Tg/year (Boucher et al., 2013; Jaenicke, 2005). The physicaldimensions of PBAP range from nanometers (e.g., viruses) to hundreds of micrometers (e.g., pollen grains;Fröhlich‐Nowoisky et al., 2016). The abundance and physical states of bioaerosols not only depend on theirsource emissions, but also are related to their atmospheric history, through processes such as condensationof organic coatings (Mikhailov et al., 2015; Pöhlker, Wiedemann, et al., 2012; Pöschl et al., 2010), photoox-idation by oxidants (Shiraiwa et al., 2012), clustering with dust particles (Barberán et al., 2015; Yamaguchiet al., 2012), and rupture due to exposure to high humidity (China et al., 2016).

Airborne measurements suggest that PBAP can be lofted to high altitudes and transported over regionaland continental distances (Fu et al., 2014; Hu et al., 2017; J. Lee et al., 2017; Pratt et al., 2009).Mountainous sites are ideal for the studies of aerosols in the free troposphere and its interaction withthe planetary boundary layer (PBL; Choularton et al., 2008; Fu et al., 2008, 2014; Matthias‐Maser

©2019. American Geophysical Union.All Rights Reserved.

RESEARCH ARTICLE10.1029/2018JD029486

Key Points:• Fluorescent particles (>0.8 μm)

account for 26.9% on average bynumber of the total particles in thatsize range over Mt. Tai in spring

• Bioaerosols are shown to undergotransformation during long‐rangetransport and contribute to HULIS

• For bioaerosol recognition,6σ‐threshold can lead to betterclassification of fungal spores

Supporting Information:• Supporting Information S1

Correspondence to:P. Fu,[email protected]

Citation:Yue, S., Ren, L., Song, T., Li, L., Xie, Q.,Li, W., et al. (2019). Abundance anddiurnal trends of fluorescentbioaerosols in the troposphere over Mt.Tai, China, in spring. Journal ofGeophysical Research: Atmospheres,124, 4158–4173. https://doi.org/10.1029/2018JD029486

Received 13 AUG 2018Accepted 2 MAR 2019Accepted article online 9 MAR 2019Published online 1 APR 2019

Author Contributions:Conceptualization: Siyao Yue,Pingqing FuFormal analysis: Siyao Yue, LujieRen, Tianli SongInvestigation: Siyao Yue, Lujie Ren,Tianli Song, Linjie LiMethodology: Siyao Yue, Lujie Ren,Tianli Song, Linjie Li, Qiaorong Xie,Mingjie Kang, Wanyu Zhao, LianfangWei, Hong Ren, Pingqing FuResources: Weijun Li, Pingqing FuSoftware: Siyao Yue, Lujie Ren, LinjieLi, Hong RenValidation: Weijun Li, Yele Sun, ZifaWang, Robert Mark Ellam, Cong‐QiangLiu, Kimitaka KawamuraWriting ‐ original draft: Siyao YueWriting – review & editing: Yele Sun,Zifa Wang, Robert Mark Ellam, Cong‐Qiang Liu, Kimitaka Kawamura,Pingqing Fu

YUE ET AL. 4158

et al., 2000; Wang et al., 2012). Mt. Tai is the most elevated location on the North China Plain, and hasbecome a focus for pollution research due to the complicated sources and evolution processes of pollu-tants (Fu et al., 2008; Gao et al., 2018; W. Li et al., 2011). The mountain is located at the east of theEast Asian Continent (Kawamura et al., 2013). When subtropical westerlies prevail, air masses transport-ing aerosols over the North China Plain can affect the downwind areas of the Korean Peninsula,Japanese Islands, and the North Pacific Ocean (Fu et al., 2012; Xu, Wei, Chen, Zhu, et al., 2017).According to a year‐round study at Mt. Tai (Zhang et al., 2014), the summit of Mt. Tai is within thePBL in the daytime of spring season. Freshly emitted PBAP and other nonbiological aerosols can betransported upward in daytime by valley breezes together with the large‐scale convective air mass.During this transport, aerosols will be involved in processes such as photooxidation and multiphase reac-tions in‐cloud droplets. While at night, especially after midnight, the PBL lowers to about half the alti-tude of the mountain (Zhang et al., 2014). The air mass over the summit station is thereforerepresentative of the lower free troposphere, which is most often derived from a regional scale orlong‐range transport (Fu et al., 2012; Zhang et al., 2014). Mountain breezes transport airborne aerosolsand gases downward, thereby influencing the local regions under the boundary layer.

Comparing with land surface environments, field characterization of bioaerosols at high altitudes is still verylimited (Després et al., 2012; Fu et al., 2014). The bioaerosols at this site have been discussed mainly based onbulk filter analysis for molecular tracers and genetic materials (Fu et al., 2008; Wang et al., 2009; M. Weiet al., 2017; Xu, Wei, Chen, Zhu, et al., 2017). Genetic sequencing of the fungal species and bacteria in aero-sols and cloud water has been reported for this site (M. Wei et al., 2017; Xu, Wei, Chen, Sui, et al., 2017; Xu,Wei, Chen, Zhu, et al., 2017). These studies presented fungal sequencing data showing variations in PM2.5

and PM1 for the summer, autumn, and winter seasons. Characteristics of bioaerosols in spring are still notwell known. Besides, direct online characterization of size‐resolved single PBAP and their variations havenot been reported at this site (Kanaya et al., 2013) and are also very limited at high altitudes around theworld (Crawford et al., 2016; Gabey et al., 2013; Perring et al., 2015; Ziemba et al., 2016).

The measurement of PBAP based on UV light/laser‐induced fluorescence (UV‐LIF) method is fast and non-destructive by using online instruments such as Ultraviolet Aerodynamic Particle Sizer (Hairston et al.,1997), Wideband Integrated Bioaerosol Sensors (WIBS; Kaye et al., 2005) and Single‐Particle FluorescenceSensor (Pan et al., 2007). This technique has been applied for online PBAP characterization in forest, rural,urban, and suburban environments (Gabey et al., 2011; Huffman et al., 2013; Pöschl et al., 2010; Yue et al.,2017). Autofluorescence of PBAP is caused by natural bio‐fluorophores. Typical biological fluorophoresinclude amino acids (e.g., tryptophan, tyrosine); structural chemicals such as chitin, cellulose, lignin, andsporopollenin; coenzymes such as Nicotinamide Adenine Dinucleotide (NADH) and NicotinamideAdenine Dinucleotide(P)hosphate (NAD(P)H), flavins and vitamins, pigments (e.g. chlorophyll‐a,chlorophyll‐b), nucleic acids (e.g. DNA and RNA), and secondary metabolites (Fennelly et al., 2017;Pöhlker, Huffman, et al., 2012). Nonbiological fluorescent compounds including HULIS, mineral dust con-taining transition metals and rare‐earth cations, PAHs, soot, and SOA are potential interferences for fluor-escent biological aerosol particles (FBAP) measured by UV‐LIF method (H. Lee et al., 2013; Pöhlker,Huffman, et al., 2012; Savage et al., 2017). While the excitation‐emission (Ex‐Em) spectral bands of WIBS‐4A are tuned to excite at the maximum fluorescence of tryptophan and NAD(P)H, it should be noted thatmany other biofluorophores, such as riboflavin, ergosterol, and chlorophylls, and the nonbiological fluoro-phores listed above can also fluoresce in the sensing spectral regions of WIBS‐4A albeit often with lowerquantum yields (Pöhlker, Huffman, et al., 2012; Savage et al., 2017).

Attribution of the fluorescent particles to specific types of bioaerosols is an active field of research. Themajority of the atmospherically relevant biological fluorophores has been summarized in Pöhlker,Huffman, et al. (2012). A simple way of categorizing fluorescent bioaerosols is by using the particle's spectralproperty. For WIBS, seven types of fluorescent particles can be classified (for definitions please refer to themethod section, Table S1 and Perring et al., 2015). According to the spectral features of a vast number of bio-logical compounds in literature (Savage et al., 2017; Pöhlker, Huffman, et al., 2012; and references therein),bio‐fluorophores of high or medium relevance in the atmosphere for the fluorescence‐resolved fluorescentaerosol particle (FAP) measurement are tryptophan and tyrosine (FL A, FL AB); NADH and NADPH (FLC, FL BC); cellulose and lignin (FL AB); riboflavin (FL B, FL BC); vitamin B6 compounds (FL A); lipofuscinand ceroid (FL BC); phenolics (FL BC, FL C); terpenoids and dipicolinic acid (FL BC); ergosterol (FL ABC);

10.1029/2018JD029486Journal of Geophysical Research: Atmospheres

YUE ET AL. 4159

sporopollenin, flavonoids, chitin, and alkaloids (FL C). It should be bear in mind, however, that the actualfluorescence of these bio‐fluorophores are rather complex presenting fluorescence characteristics beside themain features shown above, due to effects such as physical quenching (Savage et al., 2017). With respect tobiological species (Hernandez et al., 2016), more than 90% of 13 types of bacteria were classified as FL A par-ticles or FL AB particles for Bacillus subtilis; nearly 80% of the fungal spores (27 types) were categorized as FLA particles, following by FL AB and FL ABC type; pollens were partitioned into FL ABC and FL BC particles,collectively representing approximately 80%, with FL C and FL AB also representing around 10% for specieslike Eucalyptus. Another study by Savage et al. (2017) shows that as the size of pollen materials and fungalspores increase, they have the tendency to be classified as particles with multiple Ex‐Em pairs (e.g., FL ABand FL ABC).

Currently, it is still difficult to robustly determine the representativeness of single fluorescent particles forPBAP characterization by fluorescence technique (Kaye et al., 2005; Perring et al., 2015; Pöhlker,Huffman, et al., 2012). To eliminate nonbiological fluorescent compounds or assess their influence onFBAP measurements, one method is to determine an appropriate fluorescence intensity threshold abovewhich particles are defined as fluorescent particles (Savage et al., 2017). While increasing the thresholdscan eliminate weakly fluoresced nonbiological particles, small bioaerosols with weak fluorescence intensityare also prone to be excluded. As shown in the previous experiments on dust, PAHs, soot, HULIS and browncarbon, 6σ‐ and 9σ‐thresholds (refer to section 2 for definitions) are not capable of eliminating PAHs (e.g.,pyrene, phenanthrene, and naphthalene), fullerene, and diesel soot (Savage et al., 2017). In the fine mode,the 9σ‐threshold helps to decrease the interference from wood smoke soot and brown carbon (e.g., methyl-glyoxal + glycine, glycolaldehyde + methylamine, and glyoxal + ammonium sulfate). In the coarse mode,HULIS (Pony Lake fluvic acid) was largely eliminated. For dust particles, although increasing the fluores-cence thresholds were effective to classify them to be non‐FAP, the mixing state of dust particles with biolo-gical particles was still not clear, hindering to reach a conclusive method to eliminate these particles.Moreover, it has also been revealed that, for some fungi species, like Aspergillus brasiliensis, Rhizopus stolo-nifera and Aspergillus versicolor, more than 80% of their fine mode particles were misclassified into non‐FAPby applying both 3σ‐ and 9σ‐threshold baselines, which indicates that even using 3σ‐threshold, biologicalaerosol abundance can be underestimated for some species (Savage et al., 2017). Currently, discussions ofthe effects of varying fluorescence intensity thresholds on FBAP characterization have not been appliedfor ambient measurements.

The objective of this study is to investigate the abundances, size distributions, and diurnal variations ofbiological aerosol particles in the troposphere over Mt. Tai (1534 m a.s.l.) in East China. WIBS‐4A was firstutilized at this site to provide real‐time monitoring of fluorescent bioaerosols. Offline fluorescence measure-ments and chemical characterization of molecular markers for airborne fungal spores were combined withonline fluorescent bioaerosol characterization. Taking advantage of this mountain site, the effect of long‐range transport of bioaerosols is also discussed. In addition, we discuss the effects of different fluorescencethreshold criteria on ambient biological aerosol categorization, in order to optimize the analytical strategiesusing UV‐LIF techniques to characterize bioaerosols in atmospheric samples.

2. Materials and Methods2.1. Sampling

Sampling was conducted at the summit of Mt. Tai, China (1534 m a.s.l.; 36.26°N, 117.10°E), from21 March to 8 April 2017. The location is the same as for the previous study by W. Li et al. (2011). Totalsuspended particle (TSP) samples (n = 34) were collected onto quartz fiber filters (TISSUQUARTZ‐2500QAT‐UP, Pallflex) on a day/night basis using a high‐volume air sampler (Tisch 4010126, UnitedStates). Size‐segregated aerosol samples (n = 6) were collected onto Munktell microglass fiber papers usingan eight‐stage nonviable Andersen cascade impactor (Anderson, United States). The 50% cutoff sizes of theimpactor are: >9.0, 9.0–5.8, 5.8–4.7, 4.7–3.3, 3.3–2.1, 2.1–1.1, 1.1–0.7, 0.7–0.4, and <0.4 μm. The filters werefirst enveloped with aluminum foils and baked at 450 °C for 6 hours. Sampled filters were stored at −20 °Cbefore the analysis. Details of the sampling record of the date and time are provided in Tables S2 and S3.Hourly meteorological data were obtained from the meteorological station at the summit of Mt. Tai(Station ID: 54826).


YUE ET AL. 4160

2.2. WIBS Measurement and Analyses

The principles of the operation and measurement of WIBS have been described in previous studies (Gabeyet al., 2011; Kaye et al., 2005; Yue et al., 2017). Briefly, individual airborne particles were introduced into ameasuring chamber through stainless‐steel and silicone tube. The tube was set vertically into the wind. Theinlet was around 2 m above ground. The tubing length was approximately 60 cm with bending less than 30°.For particle measurement, a continuous‐wave 635 nm diode laser was used for the initial determination ofthe optical diameters of particles and to count the number of particles. For particles larger than 0.5 μm, fluor-escent emissions of particles were detected in two wavelength ranges, 310–400 and 420–650 nm, excited byxenon wavelengths centered at 280 and 370 nm successively. The size threshold for total particle analysiswas set to 0.8 μm (Gabey et al., 2011; Healy et al., 2012; Yue et al., 2016). Assuming a density of 1 g/cm3,the mass concentrations of FAPs were calculated (Heald & Spracklen, 2009).

2.3. WIBS Fluorescence Thresholds

Fluorescence thresholds were determined as the average fluorescence intensities plus multifold standarddeviations by using forced trigger mode data for the three fluorescence channels FL1 (Ex/Em:280 nm/310–400 nm), FL2 (280 nm/420–650 nm) and FL3 (370 nm/420–650 nm; Gabey et al., 2010;Gabey et al., 2011; Mason et al., 2015; Perring et al., 2015; Twohy et al., 2016; Ziemba et al., 2016). The elas-tically scattered UV light at 370 nm is much higher than the fluorescence signal and therefore saturates thedetector of the 310–400 nm channel (Kaye et al., 2005). From the histogram of fluorescence intensity in back-ground mode (Figure S1), 3σ‐threshold (average + threefold standard deviations) criteria located at the rightbottom side of the Gaussian‐like distributions of the fluorescence intensities. In this study, 3σ‐thresholdswere applied to be consistent with many previous studies (Toprak & Schnaiter, 2013; Twohy et al., 2016;Yu et al., 2016; Yue et al., 2016, 2017), while the effect of applying more rigorous standards will be discussedat the end of the discussion session. FAPs were classified using the fluorescence‐based scheme and they werecategorized as FL A, FL AB, FL ABC, FL AC, FL B, FL BC, and FL C fluorescent particles following literatureconvention (Perring et al., 2015; Yue et al., 2017; Table S1).

2.4. Offline Fluorescence and Factor Analyses of Excitation‐Emission Matrices

Offline fluorescence was measured for size‐segregated aerosol and TSP samples. Three eighths (3/8) of eachstage of the size‐segregated aerosol samples and an aliquot (30 mm, diameter) of TSP samples, respectively,were cut for the extraction of the water‐soluble compositions by adding 20 ml Milli‐Q (18.2 MΩ/cm) waterand ultrasonicating the mixture for 20 min. The extracts were filtered via syringe‐driven filters (Millex‐GP,0.22 μm). A fluorometer (Fluoromax‐4, Horiba) and a UV‐Vis spectrophotometer (U3900H, Hitachi) wereused to analyze the excitation‐emission matrix (EEM) and absorbance of water‐soluble contents of aerosols,respectively (Fu et al., 2015; Yue et al., 2016). Excitation and emission wavelength ranges were 240–455(5 nm step) and 290–550 nm (2 nm step), respectively. Fluorescence spectra were obtained in aSignal/Reference mode with instrumental bias correction. The inner filter effect was corrected by absor-bance spectrum measured with the spectrophotometer (McKnight et al., 2001). The EEM spectra of all sam-ples were corrected by subtracting the blank sample, especially for each size range of size‐segregated aerosolsamples. First‐ and second‐order Rayleigh scattering were removed by interpolation. Fluorescence was con-verted into Raman units (RU L m−3).

A parallel factor analysis (PARAFAC) model with nonnegativity constraint was applied to extract indepen-dent components of EEMs treating TSP and size‐segregated aerosol samples as a whole (drEEM 0.2.0MATLAB toolbox; Murphy et al., 2013). Analyses were performed according to the tutorial (Murphy et al.,2013) and previous studies (Chen, Ikemori, et al., 2016; Chen, Miyazaki, et al., 2016; Yue et al., 2017).EEM spectra data were normalized to unit variance in sample mode. The number of components was deter-mined from the 2‐ to 7‐component models by inspecting the residual errors (Figure S2), spectral loadings(Figure S3), and split‐half analysis (S4C6T3, Figure S4; Murphy et al., 2013). The core consistency was41.9%, which is acceptable for real‐world data sets (Chen, Ikemori, et al., 2016; Murphy et al., 2013). A10−8 convergence criterion was used to check the model convergence. The three‐component model explains>98.9% of the variations within the data set. The three components were designated as PLOM (protein‐likeorganic matter), HULIS‐1, and HULIS‐2 by visual inspection of the peak location (Figure 3; Chen, Miyazaki,


YUE ET AL. 4161

et al., 2016; Coble, 2007). The data set was reverse‐normalized at the end of the analysis. The maximumintensities of the components in the original measurement scale (RU L m−3) are reported.

Besides, fluorescence indices were calculated from the excitation/emission pairs for the humification index(HIX, ratio of the integrated fluorescence intensity in the range of 436–480 to the range of 300–344 nm bothexcited at 255 nm) and biological index (BIX, ratio of emission intensity of 380 nm to 430 nm both excited at310 nm; Fu et al., 2015).

2.5. Organic Marker Analyses

Detailed procedures for the analysis of organic markers are well documented in previous studies (Fu et al.,2008; Schauer et al., 1996; Simoneit et al., 2004). Brieftly, for each TSP sample, a filter aliquot was extractedthree times with dichloromethane/methanol (2:1, v/v) under ultrasonication for 10 min. Solvent extractswere filtered through quartz wool, concentrated by a rotary evaporator and nitrogen‐stripped to dryness.Derivatization was then carried out by reaction with 30 μl of N,O‐bis‐(trimethylsilyl)trifluoroacetamide con-taining 1% trimethylsilyl chloride and 10 μl of pyridine at 70 °C for 3 hr. Prior to gas chromatography/massspectroscopy (GC/MS) analysis, the derivatives were mixed with 40 μl of n‐hexane containing 1.43 ng/μl ofthe internal standard (C13 n‐alkane). GC/MSmeasurements were performed on an Agilent GC (model 7890)coupled to Agilent Mass Selective Detector (model 5975C). The mass spectrometer was operated on ElectronIonization mode at 70 eV and scanned between 50 and 650 Da. GC/MS spectra for one real sample and theblank sample are presented in Figure S5. The response factors of target compounds were determined withauthentic standards. Recoveries of the tracers were better than 80%. Field blanks were subtracted from realsamples. Target compounds were identified by comparison with authentic standards and literature data (Fuet al., 2008; Fu et al., 2009).

3. Results and Discussion

Figure 1 shows the time series of meteorological parameters, number concentrations, and fractions of FAPsand non‐FAP. During the sampling period, intermittent cloud events occurred with sporadic rain and snowprecipitation. The average temperature was 3.1 °C (−4.8 to 12.6 °C) and the mean relative humidity was67.5% (21.0% to 99.0%; Table S4). For most time of the period, the winds were mainly from north, west,and south directions, indicating that the air masses coming from the surrounding North ChinaPlain dominated.

3.1. Abundance of Bioaerosols

The abundance of total FAP (>0.8 μm) in the ambient atmosphere by volume was 647 ± 533 L−1

(26.9% ± 10.0%, Figures 1 and S6 and Tables S5 and S6), which is at the scale of the concentration for cleanperiod in winter in Beijing (Yue et al., 2017), but one order of magnitude lower than the concentration inautumn in Nanjing (Yu et al., 2016). The most abundant type of FAP was FL B particles, which is the sameas the observations in Beijing and Nanjing (Yu et al., 2016; Yue et al., 2017). Laboratory experiments of bac-teria, fungal spores, and pollen grains have very rarely measured these PBAP as FL B particles (Hernandezet al., 2016; Savage et al., 2017), although Perring et al. (2015) showed that some mold spores are detected asthis type. The dominance of FL B particles may suggest nonbiological interferences, although particles con-taining riboflavin can also contribute to this type of aerosols (Savage et al., 2017). For mass concentrations,FAP represented more than half of total particles by average (61.4%, Figure S6).

The FAP concentration at Mt. Tai was much higher than the measurements in other high‐altitude sites. Forexample, the FAP concentration was two orders of magnitude higher than that above the High‐AltitudeResearch Station Jungfraujoch in winter (6.3 ± 5.7 L−1, 3,580 m a.s.l.; Crawford et al., 2016). The lowerFAP level at the Jungfraujoch site is a combination of the winter biological inactivity, the high altitude of thissite which causes weak interactions of PBL, and the dominant marine air masses at this site (Crawford et al.,2016; Z. Li et al., 2017). The FL1 particle concentration in this study was 190 ± 226 L−1 which is much higherthan the concentration at Puy de Dôme mountain in France (12 ± 6 L−1, 1,465 m a.s.l) in summer usingWIBS‐3 (Gabey et al., 2013). The average abundance of FL3 (281 L−1) fluorescent aerosol at this site was alsohigher than the Puy de Dôme mountain (95 L−1). Regional pollution can also lead to higher FAP concentra-tion since nonbiological fluorophores may be included, which can be seen from the WIBS measurements inBeijing, another city frequently influenced by pollution of North China Plain (K. Wei et al., 2016; Yue et al.,


YUE ET AL. 4162

2017). Alternatively, varying climatic zones, WIBS prototypes and FAP definitions may also contribute tothese differences (Pöhlker, Huffman, et al., 2012; Savage et al., 2017).

Arabitol, mannitol, and trehalose are produced in large amounts by a variety of fungi. Arabitol andmannitolcan also be emitted by lichen (Lewis & Smith, 1967). Algae and higher plants can also contribute to mannitol(Lewis & Smith, 1967). Trehalose is also present in bacteria, yeast, and a few higher plants (Medeiros et al.,2006). In this study, the mass concentrations of arabitol, mannitol, and trehalose of TSP samples were3.3 ± 2.4, 4.6 ± 3.5, and 3.9 ± 3.6 ng/m3, respectively. Their abundance are lower than those in previousreports during nondust periods in spring (23, 15, 13 ng/m3; Wang et al., 2012). The concentrations of thesesugar polyols were an order of magnitude lower than those in early June (90, 90, and 23 ng/m3, respectively),which was influenced by biomass burning pollution (Xu, Wei, Chen, Zhu, et al., 2017). Sugar polyol concen-trations were also lower than those in late June (11, 24, 9.0 ng/m3; Fu et al., 2008, 2012), when the higherconcentrations were attributable to the increased biological activity in the warmer seasons. A strong positivelinear correlation (R2 = 0.85, p < 0.001) was found between arabitol and mannitol, which supports the ideathat they are both tracers for airborne fungal spores since significantly multiple sources may lower thecorrelation level (Fu et al., 2008; Gosselin et al., 2016). Mannitol and trehalose were positively correlated(R2 = 0.78), while arabitol and trehalose were weakly correlated (R2 = 0.56). These sugar polyols are alsoused to trace the resuspension of soil dust (Fu et al., 2008; Simoneit et al., 2004).

Biomass burning is one of the main sources of fluorescent interferences of bioaerosol detection by UV‐LIFmethod. Levoglucosan, mannosan, and galactosan are three main tracers for biomass burning. The averagemass concentration of levoglucosan was 38.4 ± 29.8 ng/m3, which is much lower than those (mean426 ng/m3) reported in Mt. Tai aerosols collected in early summer when wheat‐burning was active in theNorth China Plain (Fu et al., 2008). The three biomass burning tracers strongly correlate with each other(R2 > 0.88, Table S7). The Pearson correlation coefficients (R2) of biomass burning tracers with fungal tracerswere less than 0.08 and did not pass the two‐tailed t test, which indicates that biomass burning was not asignificant source of fungal materials during this sampling period (Table S7). This result contrasts with

Figure 1. Time series of meteorological parameters and WIBS results. (a) Temperature and RH; (b) wind speed and winddirection; for simplicity, wind directions are defined as north wind (N): 315°–359° and 0°–45°, east wind (E): 45°–135°,south wind (S): 135°–225°, west wind (W): 225°–315°; (c) number fractions of fluorescent particles and non‐FAP to totalparticles; (d) number concentrations of fluorescent particles and non‐FAP. According to their spectral properties, fluor-escent particles are classified as FL A, FL AB, FL ABC, FL AC, FL B, FL BC, FL C (Table S1 for full description of defi-nitions). In‐cloud, snow, and rain periods are indicated by color shading. RH = relative humidity; FAP = fluorescentaerosol particle.


YUE ET AL. 4163

the biomass‐burning active period in early June, during which biomass contributed to fungal particlessignificantly (Fu et al., 2012).

PAHs are major interferences of the detection of FBAP (Pöhlker, Huffman, et al., 2012; Savage et al., 2017).The total concentration of PAHs was 4.76 ng/m3 (0.11–14.4), which is at the same scale of the concentrationin summer (14.6; Fu et al., 2008). The correlation coefficients (R2) between PAHs and the three biologicaltracers were lower than 0.01 (Table S7). All the coefficients between FL‐size‐resolved FAPs and PAHs werealso found to be <0.29 (Tables S8 and S9), which indicates that they were not important interferences of themeasurement of FBAP in this study.

3.2. Size Distribution

As shown in Figure 2, all types of FAPs showed a predominant mode. Except for the coarse peak (~2.5 μm) ofFL ABC particles, other FAPs were dominated by the fine mode (~1 μm). Compared to single Ex‐Em pair

Figure 2. Size distributions of FL‐resolved FAPs by applying different fluorescence thresholds (3σ‐, 4σ‐, 5σ‐, 6σ‐, and9σ‐threshold). Dotted lines show the size distributions of the eliminated FAPs from the 3σ‐ to 6σ‐threshold (black) andfrom the 6σ‐ to 9σ‐threshold (white). Solid lines show the fraction of FAPs excluded from the 3σ‐ to 6σ‐threshold relativeto 3σ‐threshold. The black and gray parts indicate the decrease and increase of particle number concentrations, respec-tively. (a) FL A, (b) FL AB, (c) FL ABC, (d) FL B, (e) FL BC, and (f) FL C. FAP = fluorescent aerosol particle.


YUE ET AL. 4164

FAPs, fluorescent particles with multiple Ex‐Em pairs (FL AB, AC, BC, and FL ABC) had greatercontribution of larger aerosols. This agrees with the findings of laboratory experiments which reveal achange pattern of the fluorescent particle type (A‐AB‐ABC, B‐BC‐ABC, and C‐BC‐ABC) for individualfungi, intact and fragmented pollen, and bio‐fluorophores (Savage et al., 2017). This can be explained thatas particle size increases, the increasing quantity of constituent biomolecules leads to fluorescenceintensity exceeding the baseline of other spectral band. This change pattern has also been observed in theatmosphere in winter in Beijing (Yue et al., 2017) and in laboratory experiments with nonbiologicalinterferences (Savage et al., 2017). Collectively, as to number concentration, the fluorescent bioaerosolsresided mainly in the fine mode (<~3 μm), while for mass concentration, they were dominated by coarse(2.5–10 μm) and Lcoarse particles (>10 μm; Figure S6).

To identify and quantify independent chromophore components of water‐soluble organic matter,PARAFAC model was applied to the size‐segregated aerosol samples. PARAFAC model can decomposethe excitation and emission matrix of fluorescent mixtures to individual chemical components (Murphyet al., 2013). Three components were found and identified as PLOM, HULIS‐1, and HULIS‐2 according totheir spectral location on the EEM maps (Figure 3a; Chen, Miyazaki, et al., 2016; Yue et al., 2017).

PLOM showed a bimodal distribution of a fine mode peak (around 0.6 μm) and a coarse mode peak (around5.0 μm; Figures 3b and S7). The fine mode peak may include protein molecules, decomposed peptides,amino acids and smaller bacteria, and so forth, while fungal spores, pollen fragments may constitute thecoarse mode peaks (Després et al., 2012; Pöschl, 2005). By comparison, the size distributions of the numberand surface area concentrations of FL1‐FAP are also shown here (Figure 3b). They showed a fine (~1 μm)and coarse (~6 μm) predominant mode, respectively. Tryptophan and tyrosine are two widely contained

Figure 3. Fluorescence fingerprints of the excitation‐emission matrix components, absolute and relative abundance ofbiological matter and HULIS for the water‐soluble contents of the eight‐stage size‐segregated aerosol samples, and thesize distributions of FL1‐FAP matter. (a) Excitation‐emission matrix components identified by the three‐componentresults of the PARAFAC model for water‐soluble organic matter. The three components were designated to PLOM (pro-tein‐like organic matter), HULIS‐1, and HULIS‐2 by visual inspection of the peak location. (b) Size distributions of theaverage fluorescence intensity (n = 6 samples) of the fluorescence components PLOM, HULIS‐1, and HULIS‐2. Themaximum fluorescence intensities of the components in the original measurement scale are reported. The size distribu-tions of the number concentration (FL1‐FAPN) and surface area concentration (FL1‐FAPS) of FL1‐FAP are plotted asshades and dotted lines, respectively. (c) Scatter plot of humification index versus biological index (n = 54).FAP = fluorescent aerosol particle.


YUE ET AL. 4165

amino acids which have relatively high fluorescence quantum yields(Pöhlker, Huffman, et al., 2012). Materials containing these amino acidscontribute to both FL1‐FAP and PLOM (Catalá et al., 2016; Pöhlker,Huffman, et al., 2012). It has been shown that the fluorescence intensityof single particles scale as the particle size to the power between twoand three (Sivaprakasam et al., 2011; Taketani et al., 2013). This isbecause the inner fluorescent molecules may not contribute to the totalfluorescence intensity, especially for large particles (Sivaprakasam et al.,2011; Taketani et al., 2013). However, our results contrast with thisfeature. The resolved PLOM component are water‐soluble organicmatter. Additional nonnitrogen‐containing compounds (e.g., phenolcompounds and naphthalene) can also be included (Chen, Ikemori,et al., 2016). The discrepancy may be also caused by the differentquantum yields of biological fluorophores in single particle state and inwater solution (Pöhlker, Huffman, et al., 2012). Different sizing methods(aerodynamic vs optical) for nonspherical aerosol particles may also be apotential reason for the shift of the size distribution (Ramachandran &Cooper, 2011).

Figure 5. Correlations between fine (0.8–2.5 μm) and coarse (2.5–10 μm)mode FL BC and HULIS‐1. Eight samples collected in entire in‐cloud peri-ods are excluded from the calculation.

Figure 4. Diurnal variations of the fluorescent aerosol particles for nighttime (00:00–06:00) and daytime (08:00–16:00). (a)The number concentrations and fractions of FAP, relative number concentration of FL‐resolved FAPs, and FL‐size‐resolved FAPs in nighttime and daytime. The areas of the pie charts represent the relative abundance of total particles(>0.8 μm). (b) Diurnal variations of the number concentrations of FAP, non‐FAP, FL‐resolved FAPs, and FAP fraction.FAP = fluorescent aerosol particle.


YUE ET AL. 4166

HULIS‐2 and HULIS‐1 also exhibited two size modes, however, the coarsemode peak of HULIS‐1 wasmuch weaker than the fine peak. According tothe spectral positions (Chen, Ikemori, et al., 2016; Chen, Miyazaki, et al.,2016), HULIS‐2 has been designated as a less‐oxygenated species of terres-trial origin, while HULIS‐1 has been appointed to a highly oxygenatedspecies which may be due to the photodegradation of organic chromo-phores, such as PLOM and HULIS‐2. The dominant fine mode ofHULIS‐1 can be explained by the fragmentation of chromophores throughthe oxygenation reaction pathways (Chen, Miyazaki, et al., 2016).

Fluorescence indices can provide insights into the origins of fluorophores.The BIX is used to estimate the contribution of autochthonous biologicalactivity. High values (>1) have been shown to correspond to a predomi-nately biological origin of organic matter, whereas low values (<0.6) indi-cated few biological materials (Fu et al., 2015; Huguet et al., 2009). TheHIX was introduced to estimate the maturation level of dissolved organicmatter in soil (Zsolnay et al., 1999). Through the humification process, themicrobial availability of organic matter decreases and its aromaticityincreases. Low values (<4) correspond to autochthonous or microbial ori-gin, while high values (>10) are indicative of strongly humified organics,mainly of terrestrial origin (Birdwell & Engel, 2010; Huguet et al., 2009).In this study, BIX was smaller for the fine mode particles (<2.1 μm,0.8–1.1, average: 0.9) than that for the coarse mode particles (>2.1 μm,0.9–1.5, 1.2). Similarly, HIX decreased from fine (1.7–3.4, 2.4) to coarsemode (0.6–3.0, 1.1). BIX of aerosols between 3.3 and 5.8 μm were in thevery lower‐right side of Figure 3c. BIX values of TSP aerosol samples inBeijing in spring were between 0.8 and 1.4, similar to the range in thisstudy (Yue et al., 2016). Comparison with previous data for dissolvedorganic matter in river and oceanic waters, aquatic and soil humic sub-stances, as well as rain and fog‐water samples, the scatter plot of HIXand BIX values shows that coarse aerosols were contributed by more bio-logical materials, while fine mode aerosols contained more humifiedmatter of terrestrial source (Birdwell & Engel, 2010; Fu et al., 2015).This is apparent in Figure 3b, where the ratio of PLOM to HULIS is muchhigher in the coarse size range than the fine size range.

3.3. Diurnal Variations of Bioaerosols and the Transformation ofBioaerosols to HULIS

As illustrated in Figure 4, the concentrations of total FAP (747 vs 458 L−1,>0.8 μm), non‐FAP and all FL‐resolved FAPs were higher during the day(08:00–16:00) than at night (00:00–06:00) and showed a diurnal peak inthe early afternoon andminimum shortly before sunrise. The fungal sporetracers also showed higher daytime abundance of biological matter(Figure S8). This feature was similar to most of the primary and secondaryorganic tracers during the nonbiomass burning periods in late June at thesame site (Fu et al., 2012). However, the fraction of FAP was similar at

night (23.2%) and during the day (20.8%). The relative abundance of FL‐resolved FAPs was also similarbetween day and night with the exception of FL A and FL BC (Figure 4a).

The inverse day/night contrast of these two kinds of fluorescent particles suggests the transformation ofprotein‐like bioaerosols of FL A type to HULIS detected as FL B particles. During long range transport,protein‐like substances FL A particles can be photo‐oxidized, contributing to HULIS which was part ofthe FL BC particles. The fluorescence fingerprint of the highly oxygenated HULIS‐1 (Figure 3a) shows thatit can also contribute to FL BC particles, which is further supported by the moderate correlation betweenfine (R2 = 0.60)/coarse (R2 = 0.50) mode FL BC and HULIS‐1 (Figure 5). One formation pathway of

Figure 6. Pearson correlation coefficients (R2, n = 26) between fluorescentaerosol particles and the fungal spore tracer mannitol of TSP samples.Fluorescent aerosol particles are categorized by fluorescence property andsize ranges: fine (0.8–2.5 μm), coarse (2.5–10 μm), and Lcoarse (>10 μm).Eight samples collected in entire in‐cloud periods are excluded from thecalculation. Details of correlation coefficients with arabitol, trehalose, andbiomass burning tracers levoglucosan, mannosan, and galactosan areshown in Tables S8 and S9. (* indicates that the correlation coefficient doesnot pass the two‐tailed t test with significance level p < 0.001). TSP = totalsuspended particle.


YUE ET AL. 4167

HULIS‐1 is the photooxidation of PLOM substances, such as tryptophan,tyrosine, phenol, and 4‐phenoxyphenol (Bianco et al., 2014; Laurentiiset al., 2013). The phenolic compounds can be produced by microorgan-isms, plants, and pyrolysis of lignin (Hättenschwiler et al., 2000; Yeeet al., 2013).

3.4. Generic Effect of Fluorescence Threshold onBioaerosol Characterization

The purpose of adjusting fluorescence threshold is to reduce the interfer-ences of nonbiological matter and maintain the biological aerosols. Herewe use the method through comparing FAP with biological tracers andwith non‐FAP to assess the effects of different fluorescence thresholdsfor bioaerosol characterization.

Physically, higher thresholds will eliminate weakly fluorescent particles in each channel, either nonbiologi-cal particles or fluorescent bioaerosols (Savage et al., 2017). This can specifically cause a change for the cate-gorization of fluorescent particle and of their size distributions.

First, in this field measurement, increasing fluorescence threshold from 3σ to 6σ removed a large fraction ofFAP. In general, by applying the 6σ‐threshold, the average number fraction of FAP in total particles largerthan 0.8 μm reduced by 15% (from 26.9% to 11.9%, Table S6). In particular, more than half of the fine (0.8–2.5 μm) and coarse (2.5–10 μm)mode FAP were recategorized as nonfluorescent particles (Figures 2 and S6).As an example, the concentration of FL A decreased by one order of magnitude, from 107 to 8.40 L−1. As tomass concentration, FAP contribution to total particulate reduced by 18.8% (from 61.4% to 42.6%, Figure S6).The decrease of FAPmass with 6σ‐threshold criteria was significant in finemode aerosols while Lcoarse par-ticles reduced only by a slight degree (Figure S6).

Second, applying more strict fluorescence thresholds causes the shift of FAP categorization. Namely, biolo-gical species classified into multiple Ex/Em band FAPs were reduced to be single Ex/Em band FAPs. Forinstance, from 3σ‐ to 6σ‐threshold, the number concentration of FL B (>~7 μm), Lcoarse mode FL ABand FL BC particles increased (Figure 2). This feature has also been observed in the laboratory measure-ments of bioaerosols (Savage et al., 2017). Savage et al. (2017) found that the fraction of fungiSaccharomyces cerevisiae to be classified as FL ABC and FL AB particles decreased, while the FL A fractionincreased across the whole size range when increasing the baseline from 3σ‐ to 9σ‐threshold.

Third, another effect is the shift of the size distribution of FL1 FAPs and FL BC to the coarse mode. Thismeans that the fine mode particles will be more prone to being influenced than coarse mode particles.One reason should be due to the smaller quantity of fluorophores they contain. Another reason can be thatnonbiological interferences such as SOA, HULIS mainly reside in the fine mode.

Furthermore, for bioaerosol characterization, here we show the evidence of the removal of both nonbiologi-cal matter and bioaerosols by increasing fluorescence threshold. On the one hand, as illustrated in Figure S9,compared to the correlation coefficients when using 3σ‐threshold (0.59 < R2 < 0.63), the coefficients of finemode FL Awith non‐FAP decreased to insignificant values by applying the 6σ‐threshold (R2 < 0.18), indicat-ing a removal of nonbiological particles from this particle type. FL AB, AC showed a similar trend. On theother hand, the correlation of FL ABC with non‐FAP in fine mode became stronger, from 0.12 to 0.31(R2). This feature suggests that by applying the 6σ‐threshold, a proportion of bioaerosols was misclassifiedinto non‐FAPs. In addition, FL B and FL C particles in fine mode still strongly correlated with the fine modenon‐FAPs (For FL C, 3σ: R2 = 0.72–0.90; 6σ: R2 = 0.69–0.90; 9σ: R2 = 0.67–0.88), which indicates that theymight have similar source with non‐FAPs. This is similar to the findings in winter in Beijing (Yue et al., 2017)and is also indicated by the moderate correlations with biomass burning tracers (Tables S8 and S9).

3.5. Improvements of Fungal Spore Classification

Specially, we explore the effect of fluorescence threshold on characterization of fungal spores. Eight in‐cloudsamples (TSP) were excluded from the following discussions for the sake of characterizing bioaerosols undergeneral background condition since precipitation and high RH can induce the release of bioaerosols(Huffman et al., 2013; Yue et al., 2016) and rupture of bioaerosols (China et al., 2016) by a large amount,

Table 1Statistics of the Estimated Number Concentrations of Fungal Spores FromMannitol, Arabitol, and Wideband Integrated Bioaerosol Sensor

Concentration Avg. Sdev. Min. Max.

N3σ (L−1) 45 23 6.7 86N6σ (L−1) 18 7.4 6.2 34Nmannitol (L

−1) 8.7 6.6 0.40 29Narabitol (L

−1) 8.0 5.8 0.54 26

Note. Same to Figure 7, the Wideband Integrated Bioaerosol Sensorresults are calculated as the sum of FL A (2.5–10 μm) + FL AB(2.5–10 μm) + FL ABC (>2.5 μm) + FL BC (>10 μm) under 3σ‐ and6σ‐threshold. Eight samples collected in entire in‐cloud periods areexcluded from the calculation.


YUE ET AL. 4168

respectively. This can also be seen from the spikes of FAPs in Figure 1. Themass concentrations of FAP particles were calculated (Heald & Spracklen,2009) and averaged over the high‐volume TSP sampling periods.Correlation analysis was performed between the mass concentrations offluorescent aerosols and molecular tracers. Overall, positive correlationswere found for fungal polyols with FAPs (3σ‐threshold), R2 = 0.72between FL1‐FAP and mannitol, and R2 = 0.68 between FL1‐FAP andarabitol. Compared with the 3σ‐threshold, the coarse mode FL1‐FAPmass concentration resulting from the 6σ‐threshold showed similarcorrelation coefficients with fungal polyols mannitol (R2 = 0.70) andarabitol (R2 = 0.68).

Size‐resolved correlation coefficients between mass concentrations ofFAPs and biological tracers (TSP) showed that the 6σ‐threshold could leadto better classification of FAPs for fungal spores (Figure 6 and Tables S8and S9). As an example, the best correlation with mannitol was foundfor Lcoarse FL ABC (R2 = 0.75), FL BC (R2 = 0.59), and coarse mode FLAB (R2 = 0.57). Generally, the fine mode FAPs showed no clear linearcorrelation with mannitol, which is reasonable as fungal spores oftenreside in the coarse mode (Fröhlich‐Nowoisky et al., 2016). By applyinga 6σ‐threshold, all the correlation coefficients increased except for FLABC and fine FL AB particles. The coefficients increased significantlyfor coarse FL A (from 0.29 to 0.54) and Lcoarse FL B (from 0.35 to 0.55).

These enhancements could be partly due to the exclusion of weakly fluorescent bioaerosols such as bacteriaor bacteria clusters and nonbiological interferences (Savage et al., 2017), whereas the decreased correlationcoefficients can be explained by the change of fluorescence particle type. Namely, for instance, when apply-ing higher fluorescence intensity thresholds, previously categorized FL ABC particles would fluoresceweakly in any of the three excitation/emission bands and be demoted to FL AB, FL BC, or FL AC particles,simultaneously contributing to the increased correlation relationships of other FAPs with mannitol.

Furthermore, we found evidence of better quantification of fungal spores from the comparison of the esti-mated concentration based on molecular tracers and WIBS. The number concentration of fungal sporeswas calculated by assuming 0.49 and 0.38 pg per spore for mannitol and arabitol, respectively (Liang et al.,

Figure 7. Correlations among the estimated number concentrations of fun-gal spores frommannitol, arabitol, andWIBS.WIBS results are calculated asthe sum of FL A (2.5–10 μm) + FL AB (2.5–10 μm) + FL ABC(>2.5 μm) + FL BC (>10 μm) under 3σ‐ and 6σ‐threshold. Eight samplescollected in entire in‐cloud periods are excluded from the calculation.WIBS = Wideband Integrated Bioaerosol Sensor.

Figure 8. Schematic diagram showing the effect of mountain‐valley breezes and regional wind transport on the biologicalmatter of tropospheric aerosols over Mt. Tai (1534 m a.s.l.). PBL = planetary boundary layer.


YUE ET AL. 4169

2013). As to WIBS, the criteria for the selection of fluorescence category and size range is that at least one ofthe R2 under 3σ‐ and 6σ‐threshold is larger than 0.50 (Figure 6). Thus, the sum of FL A (2.5–10 μm) + FL AB(2.5–10 μm) + FL ABC (>2.5 μm) + FL BC (>10 μm) for 3σ‐ and 6σ‐threshold was used as an estimate. Theestimated concentrations of fungal spores from mannitol (Nmannitol, 8.7 ± 6.6 L−1) and arabitol (Narabitol,8.0 ± 5.8 L−1) were equivalent and well correlated (R2 = 0.94, p < 0.001; Table 1 and Figure 7). The estima-tion of fungal spores from WIBS were 45 ± 23 (N3σ) and 18 ± 7.4 L−1 (N6σ), both of which were higher thanthe values from molecular tracers (Table 1). The positive correlation was found more significant betweenN6σ and Nmannitol than that between N3σ and Nmannitol (R

2: 0.62 vs 0.46, p < 0.001; Figure 7). Besides, thefitting slope also revealed a better representation of fungal spores under 6σ‐threshold (0.78) than3σ‐threshold (2.2).

4. Conclusions

In this article, we first deployed the online bioaerosol instrument (WIBS) on the top site of Mt. Tai, NorthChina Plain, together with filter sampling of TSP and size‐segregated particles. The number concentrations,mass abundance, size distributions, and diurnal variations of fluorescent bioaerosols were characterized.The abundance of total fluorescent aerosols was 647 ± 533 L−1, accounting for approximately one fourthof total particles (>0.8 μm). FAP represented 61.4% (mean) of total aerosol mass for the size range. All typesof FAPs showed a predominant mode, ~2.5 μm for FL ABC particles and ~1 μm for the other six types ofFAPs. The PARAFAC model resolved three factors, PLOM, HULIS‐1, and HULIS‐2 for the water‐solublecontents of aerosols. Their size distributions showed that the fluorescence of coarse aerosols was contributedmainly by biological materials, while humified matter contributed largely to the fluorescence offine particles.

The effect of applying different fluorescence thresholds for WIBS to characterize ambient bioaerosolsare also discussed. The shift of the size distributions and recategorization of FAPs have been revealedas the laboratory experiments (Savage et al., 2017). Besides, we find that while applying stricter criteria(6σ‐threshold) can improve the characterization of biological materials, despite the risk to eliminate smallFBAP with weak fluorescence. Furthermore, here we recommend constraining the sum of FL A (2.5–10 μm) + FL AB (2.5–10 μm) + FL ABC (>2.5 μm) + FL BC (>10 μm) under 6σ‐threshold as an estimateof fungal spores, although cautions are still needed for other studies to quantify this type of bioaerosol.

As illustrated in Figure 8, this mountain, in addition to large‐scale atmospheric convection, can serve as anenhancer for transporting regional aerosols to free troposphere by valley breezes. During the long‐rangetransport, PBAP can be degraded by photooxidation and contribute to HULIS, which can influence cloudformation and oxidation potentials in the troposphere, thus potentially affecting the ocean‐atmosphereinteractions and marine biogeochemical cycles when the westerlies prevail (Ariya et al., 2009; Estilloreet al., 2016; Sun & Ariya, 2006). Mountain breezes at night can also bring long‐range transported aerosolsdown to the local areas of upwind side of the mountain. Our results provide useful information to betterunderstand the abundance and evolution of tropospheric aerosols with respect to the interactions of thePBL and atmospheric chemistry over East Asia.

ReferencesAriya, P. A., Sun, J., Eltouny, N. A., Hudson, E. D., Hayes, C. T., & Kos, G. (2009). Physical and chemical characterization of bioaerosols—

Implications for nucleation processes. International Reviews in Physical Chemistry, 28(1), 1–32. https://doi.org/10.1080/01442350802597438

Barberán, A., Ladau, J., Leff, J. W., Pollard, K. S., Menninger, H. L., Dunn, R. R., & Fierer, N. (2015). Continental‐scale distributions of dust‐associated bacteria and fungi. Proceedings of the National Academy of Sciences of the United States of America, 112(18), 5756–5761.https://doi.org/10.1073/pnas.1420815112

Bianco, A., Minella, M., Laurentiis, E., Maurino, V., Minero, C., & Vione, D. (2014). Photochemical generation of photoactive compoundswith fulvic‐like and humic‐like fluorescence in aqueous solution. Chemosphere, 111, 529–536. http://www.sciencedirect.com/science/article/pii/S0045653514005177. https://doi.org/10.1016/j.chemosphere.2014.04.035

Birdwell, J. E., & Engel, A. (2010). Characterization of dissolved organic matter in cave and spring waters using UV–Vis absorbance andfluorescence spectroscopy. Organic Geochemistry, 41(3), 270–280. https://doi.org/10.1016/j.orggeochem.2009.11.002

Boucher, O., Randall, D., Artaxo, P., Bretherton, C., Feingold, G., Forster, P., et al. (2013). Clouds and aerosols (pp. 571–657). Cambridge,United Kingdom and New York, NY, USA: Cambridge University Press. https://doi.org/10.1017/CBO9781107415324.016

Catalá, T. S., Álvarez‐Salgado, X. A., Otero, J., Iuculano, F., Companys, B., Horstkotte, B., et al. (2016). Drivers of fluorescent dissolvedorganic matter in the global epipelagic ocean. Limnology and Oceanography, 61(3), 1101–1119. https://doi.org/10.1002/lno.10281


YUE ET AL. 4170

AcknowledgmentsThis work was supported the NationalNatural Science Foundation of China(grant 41625014, 41475117,41571130024, and 91543205). Theauthors declare no competing financialinterests. The authors received supportfrom Shanfeng Yuan of Institute ofAtmospheric Physics, ChineseAcademy of Sciences for meteorologicaldata. Supporting description of analyz-ing methods, figures, and tables areprovided in the supporting information.The data used in this manuscript arelisted in the tables, figures, and sup-porting information.

Chen, Q. C., Ikemori, F., & Mochida, M. (2016). Light absorption and excitation–emission fluorescence of urban organic aerosol compo-nents and their relationship to chemical structure. Environmental Science & Technology, 50(20), 10,859–10,868. https://doi.org/10.1021/acs.est.6b02541

Chen, Q. C., Miyazaki, Y., Kawamura, K., Matsumoto, K., Coburn, S., Volkamer, R., et al. (2016). Characterization of chromophoric water‐soluble organic matter in urban, forest, and marine aerosols by HR‐ToF‐AMS analysis and excitation–emission matrix spectroscopy.Environmental Science & Technology, 50(19), 10,351–10,360. https://doi.org/10.1021/acs.est.6b01643

China, S., Wang, B., Weis, J., Rizzo, L., Brito, J., Cirino, G. G., et al. (2016). Rupturing of biological spores as a source of secondary particlesin Amazonia. Environmental Science & Technology, 50(22), 12,179–12186. https://doi.org/10.1021/acs.est.6b02896

Choularton, T. W., Bower, K. N., Weingartner, E., Crawford, I., Coe, H., Gallagher, M. W., et al. (2008). The influence of small aerosolparticles on the properties of water and ice clouds. Faraday Discussions, 137, 205, https://www.ncbi.nlm.nih.gov/pubmed/18214105–222. https://doi.org/10.1039/B702722M

Coble, P. G. (2007). Marine optical biogeochemistry: The chemistry of ocean color. Chemical Reviews, 107(2), 402418. https://doi.org/10.1021/cr050350

Crawford, I., Lloyd, G., Herrmann, E., Hoyle, C. R., Bower, K. N., Connolly, P. J., et al. (2016). Observations of fluorescent aerosol–cloudinteractions in the free troposphere at the High‐Altitude Research Station Jungfraujoch. Atmospheric Chemistry and Physics, 16(4),2273–2284. https://doi.org/10.5194/acp‐16‐2273‐2016

Després, V. R., Huffman, J. A., Burrows, S. M., Hoose, C., Safatov, A. S., Buryak, G., et al. (2012). Primary biological aerosol particles in theatmosphere: A review. Tellus B, 64, 15598. https://doi.org/10.3402/tellusb.v64i0.15598

Estillore, A. D., Trueblood, J. V., & Grassian, V. H. (2016). Atmospheric chemistry of bioaerosols: Heterogeneous and multiphase reactionswith atmospheric oxidants and other trace gases. Chemical Science, 7(11), 6604–6616. https://doi.org/10.1039/C6SC02353C

Fennelly, M., Sewell, G., Prentice, M., O'Connor, D., & Sodeau, J. (2017). Review: The use of real‐time fluorescence instrumentation tomonitor ambient primary biological aerosol particles (PBAP). Atmosphere, 9(1), 1–39. https://doi.org/10.3390/atmos9010001

Fröhlich‐Nowoisky, J., Kampf, C. J., Weber, B., Huffman, A. J., Pöhlker, C., Andreae, M. O., et al. (2016). Bioaerosols in the Earth system:Climate, health, and ecosystem interactions. Atmospheric Research, 182, 346–376. https://doi.org/10.1016/j.atmosres.2016.07.018

Fu, P. Q., Kawamura, K., Chen, J., & Barrie, L. A. (2009). Isoprene, monoterpene, and sesquiterpene oxidation products in the highArctic aerosols during late winter to early summer. Environmental Science & Technology, 43, 4022–4028. https://doi.org/10.1021/es803669a

Fu, P. Q., Kawamura, K., Chen, J., Li, J., Sun, Y. L., Liu, Y., et al. (2012). Diurnal variations of organic molecular tracers and stable carbonisotopic composition in atmospheric aerosols over Mt. Tai in the North China Plain: An influence of biomass burning. AtmosphericChemistry and Physics, 12(18), 8359–8375. https://doi.org/10.5194/acp‐12‐8359‐2012

Fu, P. Q., Kawamura, K., Chen, J., & Miyazaki, Y. (2014). Secondary production of organic aerosols from biogenic VOCs over Mt. Fuji,Japan. Environmental Science & Technology, 48(15), 8491–8497. https://doi.org/10.1021/es500794d

Fu, P. Q., Kawamura, K., Chen, J., Qin, M., Ren, L., Sun, Y., et al. (2015). Fluorescent water‐soluble organic aerosols in the high Arcticatmosphere. Scientific Reports, 5(1). https://doi.org/10.1038/srep09845

Fu, P. Q., Kawamura, K., Okuzawa, K., Aggarwal, S., Wang, G., Kanaya, Y., & Wang, Z. (2008). Organic molecular compositions andtemporal variations of summertime mountain aerosols over Mt. Tai, North China Plain. Journal of Geophysical Research, 113, D19107.https://doi.org/10.1029/2008JD009900

Gabey, A. M., Gallagher, M. W., Whitehead, J., Dorsey, J. R., Kaye, P. H., & Stanley, W. R. (2010). Measurements and comparison of pri-mary biological aerosol above and below a tropical forest canopy using a dual channel fluorescence spectrometer.Atmospheric Chemistryand Physics, 10, 4453–4466. https://doi.org/10.5194/acp‐10‐4453‐2010

Gabey, A. M., Stanley, W. R., Gallagher, M.W., & Kaye, P. H. (2011). The fluorescence properties of aerosol larger than 0.8 μm in urban andtropical rainforest locations. Atmospheric Chemistry and Physics, 11, 5491–5504. https://doi.org/10.5194/acp‐11‐5491‐2011

Gabey, A. M., Vaitilingom, M., Freney, E., Boulon, J., Sellegri, K., Gallagher, M. W., et al. (2013). Observations of fluorescent and biologicalaerosol at a high‐altitude site in central France. Atmospheric Chemistry and Physics, 13(15), 7415–7428. https://doi.org/10.5194/acp‐13‐7415‐2013

Gao, Y., Yang, L., Chen, J., Li, Y., Jiang, P., Zhang, J., et al. (2018). Nitro and oxy‐PAHs bounded in PM2.5 and PM1.0 under differentweather conditions at Mount Tai in eastern China: Sources, long‐distance transport, and cancer risk assessment. Science of the TotalEnvironment, 622‐623, 1400–1407. https://doi.org/10.1016/j.scitotenv.2017.11.200

Gosselin, M. I., Rathnayake, C. M., Crawford, I., Pöhlker, C., Fröhlich‐Nowoisky, J., Schmer, B., et al. (2016). Fluorescent bioaerosol par-ticle, molecular tracer, and fungal spore concentrations during dry and rainy periods in a semi‐arid forest. Atmospheric Chemistry andPhysics, 16(23), 15,165–15,184. https://doi.org/10.5194/acp‐16‐15165‐2016

Hairston, P., Ho, J., & Quant, F. (1997). Design of an instrument for real‐time detection of bioaerosols using simultaneous measurement ofparticle aerodynamic size and intrinsic fluorescence. Journal of Aerosol Science, 28(3), 471–482. https://doi.org/10.1016/S0021‐8502(96)00448‐X

Hättenschwiler, S., Vitousek, P. M., Hättenschwiler, S., & Vitousek, P. M. (2000). The role of polyphenols in terrestrial ecosystem nutrientcycling. Trends in Ecology & Evolution, 15(6), 238–243. https://doi.org/10.1016/s0169‐5347(00)01861‐9

Heald, C. L., & Spracklen, D. V. (2009). Atmospheric budget of primary biological aerosol particles from fungal spores. GeophysicalResearch Letters, 36, L09806. https://doi.org/10.1029/2009GL037493

Healy, D. A., O'Connor, D. J., & Sodeau, J. R. (2012). Measurement of the particle counting efficiency of the “waveband integratedbioaerosol sensor” model number 4 (WIBS‐4). Journal of Aerosol Science, 47, 94–99. https://doi.org/10.1016/j.jaerosci.2012.01.003

Hernandez, M., Perring, A. E., McCabe, K., Kok, G., Granger, G., & Baumgardner, D. (2016). Chamber catalogues of optical and fluorescentsignatures distinguish bioaerosol classes. Atmospheric Measurement Techniques, 9, 3283–3292. https://doi.org/10.5194/amt‐9‐3283‐2016

Hu, W., Murata, K., Horikawa, Y., Naganuma, A., & Zhang, D. (2017). Bacterial community composition in rainwater associated withsynoptic weather in an area downwind of the Asian continent. Science of the Total Environment, 601‐602, 1775–1784. https://doi.org/10.1016/j.scitotenv.2017.06.052

Huffman, J. A., Prenni, A. J., DeMott, P. J., Pöhlker, C., Mason, R. H., Robinson, N. H., et al. (2013). High concentrations of biologicalaerosol particles and ice nuclei during and after rain. Atmospheric Chemistry and Physics, 13(13), 6151–6164. https://doi.org/10.5194/acp‐13‐6151‐2013

Huguet, A., Vacher, L., Relexans, S., Saubusse, S., Froidefond, J. M., & Parlanti, E. (2009). Properties of fluorescent dissolved organic matterin the Gironde Estuary. Organic Geochemistry, 40(6), 706–719. https://doi.org/10.1016/j.orggeochem.2009.03.002

Jaenicke, R. (2005). Abundance of cellular material and proteins in the atmosphere. Science (New York, N.Y.), 308(5718), 73. https://doi.org/10.1126/science.1106335


YUE ET AL. 4171

Kanaya, Y., Akimoto, H., Wang, Z. F., Pochanart, P., Kawamura, K., Liu, Y., et al. (2013). Overview of the Mount Tai experiment(MTX2006) in central East China in June 2006: Studies of significant regional air pollution. Atmospheric Chemistry and Physics, 13(16),8265–8283. https://doi.org/10.5194/acp‐13‐8265‐2013

Kawamura, K., Tachibana, E., Okuzawa, K., Aggarwal, S. G., Kanaya, Y., & Wang, Z. F. (2013). High abundances of water‐soluble dicar-boxylic acids, ketocarboxylic acids and α‐dicarbonyls in the mountaintop aerosols over the North China Plain during wheat burningseason. Atmospheric Chemistry and Physics, 13(16), 8285–8302. https://doi.org/10.5194/acp‐13‐8285‐2013

Kaye, P. H., Stanley, W. R., Hirst, E., Foot, E. V., Baxter, K. L., & Barrington, S. J. (2005). Single particle multichannel bio‐aerosol fluor-escence sensor. Optics Express, 13(10), 3583–3593. https://doi.org/10.1364/OPEX.13.003583

Laurentiis, E. D., Sur, B., Pazzi, M., Maurino, V., Minero, C., Mailhot, G., et al. (2013). Phenol transformation and dimerisation, photo-sensitised by the triplet state of 1‐nitronaphthalene: A possible pathway to humic‐like substances (HULIS) in atmospheric waters.Atmospheric Environment, 70, 318–327. https://doi.org/10.1016/j.atmosenv.2013.01.014

Lee, H., Laskin, A., Laskin, J., & Nizkorodov, S. A. (2013). Excitation–emission spectra and fluorescence quantum yields for freshand aged biogenic secondary organic aerosols. Environmental Science & Technology, 47(11), 5763–5770. https://doi.org/10.1021/es400644c

Lee, J., Park, E., Lee, S., Ko, G., Honda, Y., Hashizume, M., et al. (2017). Airborne bacterial communities in three East Asian cities of China,South Korea, and Japan. Scientific Reports, 7(1), 5545. https://doi.org/10.1038/s41598‐017‐05862‐4

Lewis, D., & Smith, D. (1967). Sugar alcohols (polyols) in fungi and green plants. New Phytologist, 66(2), 185–204. https://doi.org/10.1111/j.1469‐8137.1967.tb05998.x

Li, W. J., Zhang, D. Z., Shao, L. Y., Zhou, S. Z., & Wang, W. X. (2011). Individual particle analysis of aerosols collected under haze and non‐haze conditions at a high‐elevation mountain site in the North China plain. Atmospheric Chemistry and Physics, 11(22), 11,733–11,744.https://doi.org/10.5194/acp‐11‐11733‐2011

Li, Z., Guo, J., Ding, A., Liao, H., Liu, J., Sun, Y., et al. (2017). Aerosol and boundary‐layer interactions and impact on air quality. NationalScience Review, 4(6), 810–833. https://doi.org/10.1093/nsr/nwx117

Liang, L., Engling, G., Cheng, Y., Duan, F., Du, Z., & He, K. (2013). Rapid detection and quantification of fungal spores in the urbanatmosphere by flow cytometry. Journal of Aerosol Science, 66(2013), 179–186. https://doi.org/10.1016/j.jaerosci.2013.08.013

Mason, R. H., Si, M., Li, J., Chou, C., Dickie, R., Toom‐Sauntry, D., et al. (2015). Ice nucleating particles at a coastal marine boundary layersite: Correlations with aerosol type and meteorological conditions. Atmospheric Chemistry and Physics, 15(21), 12,547–12,566. https://doi.org/10.5194/acp‐15‐12547‐2015

Matthias‐Maser, S., Reichert, K., & Jaenicke, R. (2000). Primary biological aerosol particles at the high alpine site ofJungfraujoch/Switzerland. Journal of Aerosol Science, 31(1), 955–956. https://doi.org/10.1016/s0021‐8502(00)90965‐0

McKnight, D. M., Boyer, E. W., Westerhoff, P. K., Doran, P. T., Kulbe, T., & Andersen, D. T. (2001). Spectrofluorometric characterization ofdissolved organic matter for indication of precursor organic material and aromaticity. Limnology and Oceanography, 46(1), 38–48.https://doi.org/10.4319/lo.2001.46.1.0038

Medeiros, P. M., Conte, M. H., Weber, J. C., & Simoneit, B. (2006). Sugars as source indicators of biogenic organic carbon in aerosols col-lected above the Howland Experimental Forest, Maine. Atmospheric Environment, 40(9), 1694–1705. https://doi.org/10.1016/j.atmosenv.2005.11.001

Mikhailov, E. F., Mironov, G. N., Pöhlker, C., Chi, X., Krüger, M. L., Shiraiwa, M., et al. (2015). Chemical composition, microstructure, andhygroscopic properties of aerosol particles at the Zotino tall tower observatory (ZOTTO), Siberia, during a summer campaign.Atmospheric Chemistry and Physics, 15(15), 8847–8869. https://doi.org/10.5194/acp‐15‐8847‐2015

Murphy, K. R., Stedmon, C. A., Graeber, D., & Bro, R. (2013). Fluorescence spectroscopy and multi‐way techniques. PARAFAC. AnalyticalMethods, 5(23), 6557–6566. https://doi.org/10.1039/C3AY41160E

Pan, Y.‐L., Pinnick, R. G., Hill, S. C., Rosen, J. M., & Chang, R. K. (2007). Single‐particle laser‐induced‐fluorescencespectra of biological andother organic‐carbon aerosols in the atmosphere: Measurements at New Haven, Connecticut, and Las Cruces, New Mexico. Journal ofGeophysical Research, 112, D24S19. https://doi.org/10.1029/2007JD008741

Perring, A. E., Schwarz, J. P., Baumgardner, D., Hernandez, M. T., Spracklen, D. V., Heald, C. L., et al. (2015). Airborne observations ofregional variation in fluorescent aerosol across the United States. Journal of Geophysical Research: Atmospheres, 120, 1153–1170. https://doi.org/10.1002/2014JD022495

Pöhlker, C., Huffman, J. A., & Pöschl, U. (2012). Autofluorescence of atmospheric bioaerosols—Fluorescent biomolecules and potentialinterferences. Atmospheric Measurement Techniques, 5(1), 37–71. https://doi.org/10.5194/amt‐5‐37‐2012

Pöhlker, C., Wiedemann, K. T., Sinha, B., Shiraiwa, M., Gunthe, S. S., Smith, M., et al. (2012). Biogenic potassium salt particles as seeds forsecondary organic aerosol in the Amazon. Science, 337(6098), 1075–1078, http://science.sciencemag.org/content/337/6098/1075.https://doi.org/10.1126/science.1223264

Pöschl, U. (2005). Atmospheric aerosols: Composition, transformation, climate and health effects. Angewandte Chemie InternationalEdition, 44(46), 7520–7540. https://doi.org/10.1002/anie.200501122

Pöschl, U., Martin, S. T., Sinha, B., Chen, Q., Gunthe, S. S., Huffman, J. A., et al. (2010). Rainforest aerosols as biogenic nuclei of clouds andprecipitation in the Amazon. Science (New York, N.Y.), 329(5998), 1513–1516. https://doi.org/10.1126/science.1191056

Pratt, K. A., DeMott, P. J., French, J. R., Wang, Z., Westphal, D. L., Heymsfield, A. J., et al. (2009). In situ detection of biological particles incloud ice‐crystals. Nature Geoscience, 2, 398–401. https://doi.org/10.1038/NGEO521

Ramachandran, G., & Cooper, D. W. (2011). Size distribution data analysis and presentation. In Aerosol Measurement (pp. 479–506).Hoboken, NJ: John Wiley. https://doi.org/10.1002/9781118001684.ch22

Savage, N. J., Krentz, C. E., Könemann, T., Han, T. T., Mainelis, G., Pöhlker, C., & Huffman, A. J. (2017). Systematic characterization andfluorescence threshold strategies for the wideband integrated bioaerosol sensor (WIBS) using size‐resolved biological and interferingparticles. Atmospheric Measurement Techniques, 10(11), 4279–4302. https://doi.org/10.5194/amt‐10‐4279‐2017

Schauer, J. J., Rogge, W. F., Hildemann, L. M., Mazurek, M. A., Cass, G. R., & Simoneit, B. R. (1996). Source apportionment of airborneparticulate matter using organic compounds as tracers. Atmospheric Environment, 30(22), 3837–3855. https://doi.org/10.1016/1352‐2310(96)00085‐4

Shiraiwa, M., Selzle, K., Yang, H., Sosedova, Y., Ammann, M., & Pöschl, U. (2012). Multiphase chemical kinetics of the nitration of aero-solized protein by ozone and nitrogen dioxide. Environmental Science & Technology, 46(12), 6672–6680. https://doi.org/10.1021/es300871b

Simoneit, B. R. T., Elias, V. O., Kobayashi, M., Kawamura, K., Rushdi, A. I., Medeiros, P. M., et al. (2004). Sugars–dominant water‐solubleorganic compounds in soils and characterization as tracers in atmospheric particulate matter. Environmental Science & Technology,38(22), 5939–5949. https://www.ncbi.nlm.nih.gov/pubmed/15573592, https://doi.org/10.1021/es0403099


YUE ET AL. 4172

Sivaprakasam, V., Lin, H.‐B., Huston, A. L., & Eversole, J. D. (2011). Spectral characterization of biological aerosol particles using two‐wavelength excited laser‐induced fluorescence and elastic scattering measurements. Optics Express, 19(7), 6191. https://doi.org/10.1364/OE.19.006191

Sun, J., & Ariya, P. A. (2006). Atmospheric organic and bio‐aerosols as cloud condensation nuclei (CCN): A review. AtmosphericEnvironment, 40(5), 795–820. https://doi.org/10.1016/j.atmosenv.2005.05.052

Taketani, F., Kanaya, Y., Nakamura, T., Koizumi, K., Moteki, N., & Takegawa, N. (2013). Measurement of fluorescence spectra fromatmospheric single submicron particle using laser‐induced fluorescence technique. Journal of Aerosol Science, 58, 1–8. https://doi.org/10.1016/j.jaerosci.2012.12.002

Toprak, E., & Schnaiter, M. (2013). Fluorescent biological aerosol particles measured with the waveband integrated bioaerosol sensorWIBS‐4: Laboratory tests combined with a one year field study. Atmospheric Chemistry and Physics, 13, 225–243. https://doi.org/10.5194/acp‐13‐225‐2013

Twohy, C. H., McMeeking, G. R., DeMott, P. J., McCluskey, C. S., Hill, T. C. J., Burrows, S. M., et al. (2016). Abundance of fluorescentbiological aerosol particles at temperatures conducive to the formation of mixed‐phase and cirrus clouds. Atmospheric Chemistry andPhysics, 16(13), 8205–8225. https://doi.org/10.5194/acp‐16‐8205‐2016

Wang, G. H., Kawamura, K., Umemoto, N., Xie, M., Hu, S., & Wang, Z. (2009). Water‐soluble organic compounds in PM2.5 and size‐segregated aerosols over Mount Tai in North China Plain. Journal of Geophysical Research, 114, D19208. https://doi.org/10.1029/2008JD011390

Wang, G. H., Li, J. J., Cheng, C. L., Zhou, B. H., Xie, M. J., Hu, S. Y., et al. (2012). Observation of atmospheric aerosols at Mt. Hua and Mt.Tai in central and East China during spring 2009—Part 2: Impact of dust storm on organic aerosol composition and size distribution.Atmospheric Chemistry and Physics, 12(9), 4065–4080. https://doi.org/10.5194/acp‐12‐4065‐2012

Wei, K., Zou, Z., Zheng, Y., Li, J., Shen, F., Wu, C.‐y., et al. (2016). Ambient bioaerosol particle dynamics observed during haze and sunnydays in Beijing. Science of the Total Environment, 550, 751–759. https://doi.org/10.1016/j.scitotenv.2016.01.137

Wei, M., Xu, C., Chen, J., Zhu, C., Li, J., & Lv, G. (2017). Characteristics of bacterial community in cloud water at Mt Tai: Similarity anddisparity under polluted and non‐polluted cloud episodes. Atmospheric Chemistry and Physics, 17(8), 5253–5270. https://doi.org/10.5194/acp‐17‐5253‐2017

Xie, J., Jin, L., Luo, X., Zhao, Z., & Li, X. (2017). Seasonal disparities in airborne bacteria and associated antibiotic resistance genes in PM2.5

between urban and rural sites. Environmental Science & Technology Letters, 5(2), 74–79. https://doi.org/10.1021/acs.estlett.7b00561Xu, C., Wei, M., Chen, J., Sui, X., Zhu, C., Li, J., et al. (2017). Investigation of diverse bacteria in cloud water at Mt. Tai, China. Science of the

Total Environment, 580, 258–265. https://doi.org/10.1016/j.scitotenv.2016.12.081Xu, C., Wei, M., Chen, J., Zhu, C., Li, J., Lv, G., et al. (2017). Fungi diversity in PM2. 5 and PM1 at the summit of Mt. Tai: Abundance, size

distribution, and seasonal variation. Atmospheric Chemistry and Physics, 17(18), 11,247–11,260. https://doi.org/10.5194/acp‐17‐11247‐2017

Yamaguchi, N., Ichijo, T., Sakotani, A., Baba, T., & Nasu, M. (2012). Global dispersion of bacterial cells on Asian dust. Scientific Reports,2(1), 525. https://doi.org/10.1038/srep00525

Yee, L. D., Kautzman, K. E., Loza, C. L., Schilling, K. A., Coggon, M. M., Chhabra, P. S., et al. (2013). Secondary organic aerosol formationfrom biomass burning intermediates: Phenol and methoxyphenols. Atmospheric Chemistry and Physics, 13(16), 8019–8043. https://doi.org/10.5194/acp‐13‐8019‐2013

Yu, X., Wang, Z., Zhang, M., Kuhn, U., Xie, Z., Cheng, Y., et al. (2016). Ambient measurement of fluorescent aerosol particles with a WIBSin the Yangtze River Delta of China: Potential impacts of combustion‐related aerosol particles. Atmospheric Chemistry and Physics,16(17), 11337–11348. https://doi.org/10.5194/acp‐16‐11337‐2016

Yue, S., Ren, H., Fan, S., Sun, Y., Wang, Z., & Fu, P. (2016). Springtime precipitation effects on the abundance of fluorescent biologicalaerosol particles and HULIS in Beijing. Scientific Reports, 6(1). https://doi.org/10.1038/srep29618

Yue, S., Ren, H., Fan, S., Wei, L., Zhao, J., Bao, M., et al. (2017). High abundance of fluorescent biological aerosol particles in winter inBeijing, China. ACS Earth and Space Chemistry, 1(8), 493–502. https://doi.org/10.1021/acsearthspacechem.7b00062

Zhang, Y. M., Zhang, X. Y., Sun, J. Y., Hu, G. Y., Shen, X. J., Wang, Y. Q., et al. (2014). Chemical composition and mass size distribution ofPM1 at an elevated site in central East China. Atmospheric Chemistry and Physics, 14(22), 12,237–12,249. https://doi.org/10.5194/acp‐14‐12237‐2014

Ziemba, L. D., Beyersdorf, A. J., Chen, G., Corr, C. A., Crumeyrolle, S. N., Diskin, G., et al. (2016). Airborne observations of bioaerosol overthe Southeast United States using a wideband integrated bioaerosol sensor. Journal of Geophysical Research: Atmospheres, 121,8506–8524. https://doi.org/10.1002/2015JD024669/abstract

Zsolnay, A., Baigar, E., Jimenez, M., Steinweg, B., & Saccomandi, F. (1999). Differentiating with fluorescence spectroscopy the sources ofdissolved organic matter in soils subjected to drying. Chemosphere, 38(1), 45–50. https://doi.org/10.1016/S0045‐6535(98)00166‐0


YUE ET AL. 4173

abundance and diurnal trends of fluorescent bioaerosols in ... yue et... · software: siyao yue,...

Documents