Characteristics and determinants of ambient fungal spores in Hualien, Taiwan

(1)

Atmospheric Environment 39 (2005) 5839–5850

Characteristics and determinants of ambient fungal

spores in Hualien, Taiwan

Hsiao-Man Ho

a

, Carol Y. Rao

b

, Hsiao-Hsien Hsu

c

, Yueh-Hsiu Chiu

c

,

Chi-Ming Liu

d,e

, H. Jasmine Chao

c,

a

Department of Mathematics and Science Education, National Taipei Teachers College, No. 134 Sec. 2 Hoping E. Rd., Taipei, 106 Taiwan

b

Department of Community Medicine, West Virginia University, Box 9190, Morgantown, WV 26506, USA c

Graduate Institute of Public Health, Taipei Medical University, No. 250 Wu-Hsing Street, Taipei 110, Taiwan d

Department of Medicine Research and Education, Cheng Hsin Rehabilitation Medical Center, No. 45 Cheng Hsin Street, Taipei 112, Taiwan

e_{Community Medicine Research Center and Institute of Public Health, National Yang-Ming University, No. 155 Sec. 2 Li-Nong St.,} Taipei 112, Taiwan

Received 24 November 2004; accepted 6 June 2005

Abstract

Characteristics and determinants of ambient aeroallergens are of much concern in recent years because of the apparent health impacts of allergens. Yet relatively little is known about the complex behaviors of ambient aeroallergens. To address this issue, we monitored ambient fungal spores in Hualien, Taiwan from 1993–1996 to examine the compositions and temporal variations of fungi, and to evaluate possible determinants. We used a Burkard seven-day volumetric spore trap to collect daily fungal spores. Air pollutants, meteorological factors, and Asian dust events were included in the statistical analyses to predict fungal levels. We found that the most dominant fungal categories were ascospores, followed by Cladosporium and Aspergillus/Penicillium. The majority of the fungal categories had signiﬁcant diurnal and seasonal variations. Total fungi, Cladosporium, Ganoderma, Arthrinium/ Papularia, Cercospora, Periconia, Alternaria, Botrytis, and PM10 had signiﬁcantly higher concentrations (po0.05)

during the period affected by Asian dust events. In multiple regression models, we found that temperature was consistently and positively associated with fungal concentrations. Other factors correlated with fungal concentrations included ozone, particulate matters with an aerodynamic diameter less than 10 mm (PM10), relative humidity, rainfall,

atmospheric pressure, total hydrocarbons, carbon monoxide, nitrogen dioxide, and sulfur dioxide. Most of the fungal categories had higher levels in 1994 than in 1995–96, probably due to urbanization of the study area. In this study, we demonstrated complicated interrelationships between fungi and air pollution/meteorological factors. In addition, long-range transport of air pollutants contributed signiﬁcantly to local aeroallergen levels. Future studies should examine the health impacts of aeroallergens, as well as the synergistic/antagonistic effects of weather, and local and global-scale air pollutions.

Keywords: Aerobiology; Aeroallergens; Asian dust storms; Fungi

www.elsevier.com/locate/atmosenv

Corresponding author. Tel.: +8862 2736 1661x6526; fax: +8862 2738 4831. E-mail address: hchao@tmu.edu.tw (H.J. Chao).

(2)

1. Introduction

With increasing evidence of adverse health effects of outdoor allergens (Dales et al., 2004;Delﬁno et al., 1997;

Lierl and Hornung, 2003;Neas et al., 1996;Ross et al., 2002), characteristics and determinants of ambient aeroallergens have drawn much attention in recent years. Among various outdoor allergens, fungal spores are of major concern because of their abundant sources and ubiquitous presence in environments (Burge and Rogers, 2000). Exposure to outdoor fungal spores has been associated with allergic respiratory symptoms, decreased lung functions among allergic asthmatics, asthma exacerbation, and asthma-related death (Cakmak et al., 2002; Dales et al., 2000, 2003, 2004;

Rosas et al., 1998; Rutherford et al., 2000; Targonski et al., 1995). Sensitivity to certain fungal spores also predicts the existence of asthma (Lehrer et al., 1994;

O’Hollaren et al., 1991). However, relatively little is known about the complex behaviors of fungal spores in ambient environments.

Various studies have been conducted worldwide to document the composition and variation of ambient fungal spores and to examine the association between fungal spores and meteorological factors (Corden and Millington, 2001;Di Giorgio et al., 1996;Henriquez et al., 2001;Mitakakis et al., 1997;Nikkels et al., 1996;

Oh et al., 1998;Satheesh and Rao, 1994). However, the relationships between ambient fungal spores and meteorological factors are explored thoroughly in rather few studies (Katial et al., 1997;Li and Kendrick, 1994; Troutt and Levetin, 2001). The associa-tions between fungi and air pollutants also lack com-prehensive evaluation. Because of the complicated interrelationships among climatic factors, air pol-lutants, and aeroallergens (Lin and Li, 2000), it will be difﬁcult to evaluate individual health risk with-out clear understanding of their interactions in atmosphere.

Transoceanic dust storms are another emerging issue concerning regional ambient aeroallergen compositions and concentrations. Although dust storm has been a long-existing weather phenomenon in Asia and Africa, not until recently abundant and viable microorganisms were discovered in dust clouds (Grifﬁn et al., 2003;

Shinn et al., 2003;Taylor, 2002). Asian dust storms are common meteorological events in the deserts of Mon-golia and China during late winter and early spring. Long-range transported coarse particles from dust storms (yellow sand) have been documented to deterio-rate air qualities and visibilities in other East Asian countries, including Taiwan (Chun et al., 2001; Fang et al., 2002;Ma et al., 2001). Two studies found higher concentrations of several fungal taxa during Asian dust events (Wu et al., 2004; Yeo and Kim, 2002). More investigations are needed to evaluate biological

compo-nents of Asian dust and the potential health impacts of these possibly foreign microorganisms.

Outdoor allergen exposure is of particular concern in Taiwan, where subtropical climate with high tempera-ture and relative humidity year-round provides appro-priate environmental conditions for microbial growth. However, long-term ambient fungal levels have been documented in very few studies in Taiwan and are mostly restricted to the Taipei area (Tseng and Chen, 1979; Wang and Chen, 1979). Han and Chuang (Han and Chuang, 1981) monitored fungal air spora in Hualien and several other cities in Taiwan in 1976 and found that Hualien had the highest fungal concentration among all study areas. To address the above issues, we conducted a longitudinal study to monitor ambient fungal spores for 3 years in Hualien, Taiwan. The major goals were to assess the composition and temporal variation of ambient fungi in Hualien City and to examine the effects of meteorological factors, air pollutants, and long-range transported events on aero-allergen levels.

2. Materials and methods

2.1. Fungal sample collection and analysis

Airborne fungal spores were monitored from April 1993 to March 1996 in Hualien, Taiwan. Hualien City, the capital of Hualien County, is in eastern Taiwan (1211 360₂₃00_{E, 23158}0₂₀00_{N), facing the Paciﬁc Ocean, with an}

area of 29.41 km2 and a population of approximately 110,000 people. Although Hualien City has been gradually industrialized in recent years, it is still one of the least polluted areas in Taiwan and famous for its tourism. The average temperature in Hualien City is about 27.8 1C in summer and 20.5 1C in winter.

Air sampling was conducted on the 3rd-floor roof of the Hualien County Bureau of Health, located in downtown Hualien. We used a Burkard seven-day volumetric spore trap (Burkard Manufacturing Co. Limited, Rickmansworth Hertfordshire, England) to collect daily fungal spores with a flow rate of 10 l min1. The flow rate of the sampler was calibrated once a week. Briefly, Melinexs tape, coated with a thin layer of Gelvotal solution, was mounted on the drum inside the sampler. The tape was exposed to airflow through a 2 mm 14 mm orifice, rotating with the drum at a speed of 2 mm hr1. The drum was changed weekly and the exposed tape was removed and cut in 48 mm segments (a 24 h interval) for identification. Samples were analyzed under microscope every fourth day using the 12 traverse method to obtain 2 hr average concentrations (spores m3) (Rogers and Muilenberg, 2001). The spore categories counted included Cladosporium, Arthri-nium/Papularia, Aspergillus/Penicillium, Ganoderma,

(3)

Curvularia, Fusarium, Ulocladium/Stemphylium, Cercos-pora, Alternaria, Drechslera, Torula, Botrytis, Nigros-pora, Pithomyces, Periconia, ascospores, and other fungi. We classiﬁed unidentiﬁable spores and known spores not otherwise categorized as ‘‘other fungi.’’

2.2. Data of environmental parameters

Air pollution and meteorological data were provided by Taiwan Environmental Protection Administration (Taiwan EPA). The nearest EPA monitoring station is about 4 km southwest of the sampling location. Com-puterized hourly data is available starting from 1st January 1994. Environmental parameters subjected to statistical analysis included temperature, relative humid-ity (RH), rainfall, atmospheric pressure, sulfur dioxide (SO2), carbon monoxide (CO), ozone (O3), particulates

with aerodynamic diameters less than 10 mm (PM10),

nitrogen monoxide (NO), nitrogen dioxide (NO2), total

hydrocarbons (THC), methane (CH4), and non-methane

hydrocarbons (NMHC). The sampling dates affected by China dust storms starting 1994 were also supplied by Taiwan EPA (Taiwan EPA Grant EPA-92-L105-02-207,

http://cisk.atmos.pccu.edu.tw/duststorm/).

2.3. Statistical analysis

We used SAS statistical package (v.8.02, SAS Institute Inc., Cary, NC) to perform data analyses. Kruskal– Wallis test was utilized to examine diurnal and seasonal effects of ambient fungi. Mann–Whitney U test was used

to evaluate the impact of Asian dust storms on environmental factors. We used multiple regression to examine the relationships between ambient fungal spores and environmental parameters. We developed regression models for total fungi and major fungal categories, including ascospores, Cladosporium, Asper-gillus/Penicillium, Ganoderma, Arthrinium/Papularia, and other fungi. To account for the serial correlations of fungal measurements, autoregressive error models were used, selected automatically by stepwise autore-gression in SAS (PROC AUTOREG procedure). To perform regression analysis, fungal concentrations and several environmental factors, including PM10, SO2,

NO, and NMHC, were transformed using base-10 logarithm to approximate normality. In order to perform log transformation, we added 1 to the concentrations in the major fungal categories to avoid zero values.

3. Results

3.1. Characteristics of ambient fungal spores

Table 1 summarizes the distributions of ambient fungal spores during the study period using 2 h ave-rages (n ¼ 3224). Ascospores, Cladosporium, Aspergil-lus/Penicillium, Ganoderma, Arthrinium/Papularia and other fungi were the most dominant fungal categories in Hualien City, present in more than 60% of the samples.

Table 1

Descriptive statistics for airborne fungal concentrations (spore m3) from April 1993 to March 1996 in Hualien, Taiwan

Fungal categories Freq (%)a _Mean _Median _{Std Dev} _Min _Max

Total spores 100.00 6548.34 3166.67 9337.61 13.33 85533.33 Other fungi 90.91 4625.26 1230.00 8388.21 0.00 83940.00 Ascospores 91.50 870.64 180.00 1871.52 0.00 22300.00 Cladosporium 96.22 556.54 220.00 1316.01 0.00 31086.67 Aspergillus/Penicillium 84.12 184.13 73.33 412.02 0.00 15553.33 Ganoderma 85.51 130.70 40.00 221.48 0.00 2000.00 Arthrinium/Papularia 60.67 58.31 6.67 167.93 0.00 2086.67 Cercospora 50.51 33.60 6.67 107.68 0.00 2066.67 Periconia 57.20 22.89 6.67 78.33 0.00 2246.67 Fusarium 41.07 17.00 0.00 52.13 0.00 1166.67 Curvularia 49.50 16.63 0.00 42.46 0.00 680.00 Drechslera 31.85 12.20 0.00 50.12 0.00 1086.67 Nigrospora 42.31 7.49 0.00 18.18 0.00 380.00 Torula 30.25 4.76 0.00 11.51 0.00 186.67 Ulocladium/Stemphylium 21.65 4.14 0.00 16.63 0.00 280.00 Alternaria 20.35 2.40 0.00 6.66 0.00 106.67 Pithomyces 11.10 1.15 0.00 4.43 0.00 86.67 Botrytis 4.46 0.76 0.00 5.33 0.00 200.00 a

(4)

Total fungi, other fungal spores, and ascospores had similar diurnal patterns, with highest levels in the early morning (4–6 am) and lowest in the afternoon (2–4 pm) (Fig. 1). The concentration of Cladosporium was highest during 8 to 10 am and lowest during 2 to 6 pm. Similar to ascospores, Ganoderma, Arthrinium/Papularia, and Drechslera had highest concentrations in early morning (4–6 am) and kept decreasing until late afternoon. The diurnal variations of Aspergillus/Penicillium and Fusar-ium were somewhat different, with highest concentra-tions around midnight. Cercospora had lower levels during daytime (10 am–4 pm). The fungal categories not plotted had rather zigzag diurnal variations, except for Alternaria, which had similar diurnal rhythm as Clados-porium. All the fungal categories identiﬁed had statisti-cally signiﬁcant diurnal (hourly) variations (po0.05).

Fig. 2 shows the seasonal trends of ambient fungi in Hualien, Taiwan. All the plotted fungal categories had highest levels during summer, except Periconia. Categories not shown in Fig. 2 did not display apparent seasonal variation. To examine the signiﬁcance of seasonal (monthly) variation for ambient fungi, daily average data were used (n ¼ 271). All the fungal categories had statistically signiﬁcant seasonal effects, except for Aspergil-lus/Penicillium, Alternaria, Botrytis, and ascospores. Sea-sonal trend for each fungal category was similar every year during the study period, but fungal concentrations decreased over the years. This might result from the vegetation changes in vicinity during the investigation time.

3.2. Associations between ambient fungal spores and Asian dust storm events

A total of 33 episodes of Asian dust storms affected Taiwan from January 1994 to March 1996. Among our fungal measurements, 20 days of data (n ¼ 235) were affected by dust events. Fungal measurements of nearest days before and after the selected episode days were used as background (about 25 days of data, n ¼ 297). Several episodes would often occur in a short time frame so that only one background day fell between two dust episodes.

Table 2shows the distributions of fungi and selected environmental parameters during background and episode days. Total fungi, Cladosporium, Ganoderma, Arthrinium/Papularia, Cercospora, Periconia, Alternaria, Botrytis, and PM10had signiﬁcantly higher

concentra-tions (po0.05) during episode days. Relative humidity and pressure were slightly lower during the dust storm periods. No other environmental parameters were signiﬁcantly associated with Asian dust storms episodes.

3.3. Associations between ambient fungal spores and environmental parameters

We examined the relationships between environmen-tal factors and ambient fungal concentrations using

daily averages (every fourth day) from January 1994 to March 1996 (n ¼ 200). Table 3 lists the descriptive statistics for major fungal categories and environmental parameters examined in the multiple regression analyses. Multiple regression models for major fungal cate-gories are summarized in Table 4. Among the meteor-ological factors, temperature was the most consistent predictor of fungal concentrations. Except for Aspergil-lus/Penicillium and other fungi, higher temperature was positively associated with higher fungal levels. Rainfall had a negative association with Ganoderma and other fungi. Relative humidity was positively correlated with ascospores and Cladosporium, and negatively associated with Aspergillus/Penicillium. Pressure only had a nega-tive correlation with other fungi. Several air pollutants were also signiﬁcant predictors of ambient fungi. Ozone was negatively correlated with total fungi, other fungi, and Ganoderma. PM10 was positively associated with

total fungi and Cladosporium. Other environmental factors, such as THC, CO, NO2, and SO2, also had

statistically signiﬁcant relationships with fungal concen-trations. Year was another important predictor of fungal levels in Hualien City. Most of the fungal categories had higher levels in 1994 than in 1995–96, concordant with the temporal trends previously men-tioned.

4. Discussion

The dynamics of aeroallergens in the atmosphere are complicated, attributable to climate, weather, local biological sources, anthropogenic pollutants, and nat-ural events (e.g., dust storms). Long-term monitoring and appropriate statistical modeling are valuable methods to explore the interrelationships among envir-onmental factors on ambient aeroallergens. This 3-year study examined the relationships between ambient fungal spores and air pollution/meteorological factors, using multiple regression analysis and accounting for serial correlations of the repeated measurements.

The most dominant genus identiﬁed in Hualien City was Cladosporium, similar to many other geographical areas in the world (Henriquez et al., 2001;Mitakakis et al., 1997;Nikkels et al., 1996;Satheesh and Rao, 1994). Ascospores, Aspergillus/Penicillium, and Ganoderma were also prevalent in Hualien City, recovered from more than 80% of our samples and with daily mean concentrations over 130 spores m3. Our results are similar to a previous study that used a gravity slide method, which found that the most prevalent fungal genera in Hualien was also Cladosporium, followed by Fusarium, Nigrospora, Alternaria, Curvularia, Drech-slera, and Ganoderma. Ascospores and Aspergillus/ Penicillium were not identiﬁed separately (Han and Chuang, 1981). The different rank order of the fungal

(5)

0 2000 4000 6000 8000 10000 12000 14000 0-2 2-4 4-6 6-8 8-10 10-12 12-14 14-16 16-18 18-20 20-22 22-24 Hour of Day

Fungal Conc. (spores/m

-3) 0 50 100 150 200 250 300 0 5 10 15 20 25 30 35 40 Total Others Ascospores Cladosporium 0-2 2-4 4-6 6-8 8-10 10-12 12-14 14-16 16-18 18-20 20-22 22-24 Hour of Day

-3)

Asp/Pen Ganoderma

-3) Arth/Pap Cercospora Periconia Fusarium Drechslera Nigrospora Hour of Day 0-2 2-4 4-6 6-8 8-10 10-12 12-14 14-16 16-18 18-20 20-22 22-24

(6)

0 10 20 30 40 50 60 0

Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec

-3) 0 50 100 150 200 250 300 13000 12000 11000 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 Total Others Ascospores Cladosporium

-3)

Ganoderma Arth/Pap Cercospora Periconia

Fusarium Drechslera Curvularia Pithomyces

(7)

Table 2

Distributions of fungi and selected environmental factors during background and dust storm episode days in Hualien, Taiwan Episode days (n ¼ 235) Background days (n ¼ 297) p valuea

Freq (%)b _Mean _Med _SD _{Freq (%)} _Mean _Med. _SD

Fungi (spores m3₎ Total 100.0 6078.1 3566.7 7247.2 100.0 4839.1 2586.7 6026.5 0.0073 Other 92.3 4002.9 1573.3 6557.2 92.6 3418.6 1060.0 5614.4 0.3458 Ascospores 91.5 791.5 246.7 1514.7 95.6 624.0 166.7 1332.6 0.2412 Cladosporium 99.6 759.7 293.3 1319.5 97.3 474.0 240.0 686.4 0.0076 Aspergillus/Penicillium 79.2 205.5 80.0 311.7 92.6 133.3 66.7 189.6 0.3967 Ganoderma 91.0 104.9 46.7 133.3 84.2 67.6 26.7 103.0 o0.0001 Arthrinium/Papularia 73.2 91.5 20.0 164.3 57.9 34.1 6.7 108.5 o0.0001 Cercospora 58.7 40.6 6.7 78.0 53.5 18.7 6.7 32.4 0.0130 Curvularia 39.2 6.6 0.0 12.7 51.5 16.7 6.7 37.0 0.0005 Periconia 68.5 32.6 13.3 58.7 57.9 14.6 6.7 24.7 o0.0001 Fusarium 40.9 8.4 0.0 14.7 42.8 13.9 0.0 28.2 0.3499 Drechslera 33.6 10.5 0.0 28.3 37.4 7.5 0.0 16.8 0.6102 Nigrospora 40.4 8.0 0.0 21.8 34.3 5.1 0.0 11.8 0.0854 Torula 34.5 5.2 0.0 11.0 32.7 5.0 0.0 11.9 0.6569 Ulocladium/Stemphylium 22.1 3.8 0.0 19.5 21.6 3.0 0.0 14.0 0.7873 Alternaria 31.5 4.5 0.0 9.8 17.2 1.7 0.0 4.8 o0.0001 Pithomyces 8.1 0.7 0.0 2.5 12.1 1.2 0.0 4.1 0.1269 Botrytis 11.1 1.4 0.0 5.3 3.0 0.2 0.0 1.3 0.0002 RH(%) — 75.2 74.8 8.0 — 76.8 76.7 7.7 0.0022 Pressure (mmHg) — 764.1 764.0 3.5 — 764.8 765.6 3.9 0.0196 PM10(mg m 3 ) — 57.6 53.0 29.1 — 46.1 42.5 25.3 o0.0001 a

Mann–Whitney U Test was used to examine the differences of fungal levels between episode and background days, except for Botrytis [for which Exact Test was used because of low recovery frequency (o10 times)].

b_{Frequency was the percentage of samples with presence of that speciﬁc fungal category.}

Table 3

Descriptive statistics for major fungal categories and environmental factors from January 1994 to March 1996 in Hualien, Taiwan

Variables Mean Median SD Min. Max.

Fungi (spores m3) Total 4844.41 3618.06 4428.70 46.67 22476.67 Other 2896.58 880.28 4204.83 0.00 19734.44 Ascospores 1109.54 336.94 1599.74 0.00 8254.44 Cladosporium 446.25 283.89 507.98 5.56 3678.89 Aspergillus/Penicillium 155.20 88.52 206.24 0.00 1351.11 Ganoderma 105.20 72.50 109.81 0.00 624.44 Arthrinium/Papularia 48.16 10.00 92.33 0.00 569.44 Temperature (1C) 22.86 23.00 4.54 13.32 30.23 RH(%) 77.40 78.05 8.39 57.77 97.46 Pressure (mmHg) 764.65 764.85 5.06 748.44 775.79 Rainfall (mm) 0.22 0.00 0.78 0.00 6.55 PM10(mg m 3 ) 44.58 42.58 17.36 17.54 134.08 SO2(ppb) 1.11 0.98 0.72 0.13 6.67 CO (ppm) 0.79 0.76 0.22 0.17 1.55 O3(ppb) 16.79 16.31 7.48 2.03 37.95 NO (ppb) 7.62 6.60 4.54 1.12 32.61 NO2(ppb) 15.92 15.86 5.60 2.31 30.25 THC (ppm) 2.09 2.12 0.24 0.39 2.80 CH4(ppm) 1.85 1.86 0.21 0.16 3.44 NMHC (ppm) 0.26 0.23 0.17 0.04 1.82

(8)

spore types might be a result of different sampling methods (the gravity slide method underestimates small spores) and local vegetation shift.

All the fungal categories had statistically signiﬁcant diurnal variations in our study. Several fungal categories

(i.e., total fungi, other fungal spores, ascospores, Ganoderma, Arthrinium/Papularia, and Drechslera) fol-lowed a Night Pattern as coined byGregory (1973)with highest concentrations occurring between sunset and sunrise. The causes of this diurnal pattern are probably

Table 4

Multiple regression models for major fungal categories in Hualien, Taiwan

b Coeff. SE p value R2 Total Fungi 0.44 Intercept 2.7224 0.4196 o0.0001 Temperature 0.0267 0.0094 0.0050 Log10(PM10) 0.4066 0.1980 0.0414 O3 0.0110 0.0042 0.0094 THC 0.3532 0.1483 0.0183 CO 0.4117 0.1613 0.0115 Yr 1994 0.2608 0.0981 0.0085 Yr 1995–6 0 — — Other Fungi 0.77 Intercept 27.5590 9.7799 0.0054 Pressure 0.0323 0.0128 0.0126 Rainfall 0.1129 0.0469 0.0171 O3 0.0143 0.0055 0.0096 NO2 0.0261 0.0102 0.0114 THC 0.4700 0.2095 0.0260 Yr 1994 1.2617 0.2838 o0.0001 Yr 1995–6 0 — — Ascospores 0.64 Intercept 1.5503 0.6313 0.0150 Temperature 0.0424 0.0154 0.0064 RH0.0401 0.0066 o0.0001 Log10(SO2) 0.3448 0.1510 0.0235 Cladosporium 0.27 Intercept 0.9169 0.7408 0.2174 Temperature 0.0215 0.0102 0.0363 RH0.0227 0.0055 o0.0001 Log10(PM10) 0.5665 0.2212 0.0112 Yr 1994 0.4463 0.1206 0.0003 Yr 1995–6 0 — — Aspergillus/Penicillium 0.02 Intercept 2.6914 0.3958 o0.0001 RH 0.0106 0.0051 0.0385 Ganoderma 0.65 Intercept 0.0128 0.2278 0.9552 Temperature 0.0821 0.0086 o0.0001 Rainfall 0.0886 0.0297 0.0032 O3 0.0120 0.0037 0.0015 Yr 1994 0.2049 0.0891 0.0225 Yr 1995–6 0 — — Arthrinium/Papularia 0.51 Intercept 0.1573 0.2715 0.5631 Temperature 0.0434 0.0115 0.0002 Yr 1994 0.6938 0.1346 o0.0001 Yr 1995–6 0 — —

(9)

due in part to vertical mixing depth of air and radiation inversion at night. In addition, ascospores and basidios-pores are released as humidity increases, and thus tend to be more concentrated at late night and before dawn (Burch and Levetin, 2002). Spore types such as Cladosporium and Alternaria, are usually found in higher concentrations during the warmest part of the day, dry weather conditions, with greatest wind speed and turbulence, usually referred as Middle-Day Pattern (Burch and Levetin, 2002;Gregory, 1973; Troutt and Levetin, 2001). In our study, the concentrations of Cladosporium and Alternaria were highest during 8 to 10 am.

Seasonal variations were similar every year over the study period. Most fungal categories peaked during warmer months, from May to July (Fig. 2). Although peak months varied slightly with differing local climates, similar ﬁndings were observed in studies conducted in other parts of the world (Bush and Portnoy, 2001;Di Giorgio et al., 1996; Oh et al., 1998). In spite of the consistent seasonal variations, the concentrations of total fungal spores and several major fungal categories decreased over the sampling years in Hualien City (Table 4). Sources of ambient fungi are usually considered to remain constant over time. However, major changes in the primary fungal source, plant materials, will alter the concentrations/compositions of fungal air spora (Burge, 2002). Urbanization of Hualien City and diminishing vegetation in vicinity might be some major reasons for decreasing ambient fungal levels during the study period.

To date, only very limited data are available for biological composition of Asian dust storms. In a study in Seosan, Korea, conducted during Asian dust and non-Asian dust periods in Spring 2000, researchers qualitatively found more diverse culturable fungi in samples taken during Asian dust events, including Fusarium, Aspergillus, Penicillium, and Basipetospora. Although the multi-stage impactor filter samples might result in low viability for many fungi, this study pointed out potential effects of Asian dust storms on fine biological particles (1.1–2.1 mm) (Yeo and Kim, 2002). Wu et al. (Wu et al., 2004) found basidiospores, Aspergillus/Penicillium, Nigrospora, Arthrinium, Curvu-laria, Rusts, Stemphylium, Cercospora, Pithomyces, and unidentified fungi had significantly higher concentra-tions during dust storm days than those of background days in Tainan City (southern Taiwan). Our study, which was conducted approximately 4 years earlier, found that total fungal spores, Cladosporium, Ganoder-ma, Arthrinium/Papularia, Cercospora, Periconia, Alter-naria, Botrytis, as well as PM10had significantly higher

concentrations during episode days. Moreover, in our study, Periconia and Botrytis had signiﬁcantly higher levels during episode days; Curvularia and Pithomyces levels increased during background days. These

out-comes were opposite to the findings in Tainan City. The discrepancies between the two studies are in part due to different geographical regions and weather conditions. In a study of ambient air pollutants monitored throughout Taiwan, the researcher found that Asian dust events contributed higher concentrations of parti-culate pollutants in northern and eastern Taiwan (Liu, 2004). Therefore, the long-range transported air pollu-tants might not influence Tainan City (southern Taiwan) as much as in Hualien City (eastern Taiwan). More studies are needed to clarify the associations between Asian dust events and ambient fungal composition/ variation, preferably in areas with minimal influences of local pollutants.

Temperature and sampling year were the most consistent variables predicting fungal concentrations (Table 4). Temperature is one of the most important environmental factors that inﬂuence fungal survival and growth (Burge and Otten, 1999). Similar to the results of other studies (Di Giorgio et al., 1996;Mitakakis et al., 1997; Troutt and Levetin, 2001), we found positive correlations between temperature and most fungal categories, except for Aspergillus/Penicillium and other fungi. The negative association between sampling year and fungal concentrations was likely due to local vegetation change, as previously discussed.

Water is another essential environmental factor determining fungal survival and growth (Burge and Otten, 1999). Both relative humidity and rainfall are measures associated with water availability outdoors. The effect of rainfall on ambient fungal level is mixed. Rainfall may cause release of fungal spores by splash and ‘‘tap-and-puff’’ mechanisms. Ascospores and basi-diospores require sufﬁcient moisture for dispersal, either during rainfall or as humidity increases. On the other hand, rain removes ambient fungal spores by both rainout and washout effects, especially for dry-weather spores, such as Cladosporium and Alternaria (Burge and Rogers, 2000;Weber, 2003). Therefore, variable correla-tions have been observed in the literature. Studies in Melbourne, Australia and Denver, Colorado, USA, found that total fungal spores, Cladosporium and Alternaria were negatively correlated with rainfall and positively correlated with average temperature and relative humidity (Katial et al., 1997;Mitakakis et al., 1997). In Seoul, Korea, Leptospheria ascospores con-centrations can abruptly increase during the monsoon season (Oh et al., 1998). Li and Kendrick (Li and Kendrick, 1994) found that, in Southern Ontario, Canada, the concentrations of Leptosphaeria and uni-dentiﬁed ascospores were higher during light rain, and basidiospores were most numerous during medium RH and rain. In our study, relative humidity was positively associated with ascospores and Cladosporium and negatively associated with Aspergillus/Penicillium. Rain-fall was negatively associated with other fungi and

(10)

Ganoderma, a basidiospore. This negative relationship might implicate a long-term washout effect of rainfall in Hualien City.

Penicillium and Aspergillus are both common soil fungi decaying dead plant material outdoors and are produced in response to temperature and moisture conditions (Burge and Rogers, 2000). Nevertheless, relative humidity was the only measured environmental factor signiﬁcantly associated with this fungal category. In addition, our statistical model could only account for 2% of the variation of Aspergillus/Penicillium concen-trations, as opposed to 27 to 77% of the other fungal categories. Additional environmental factors, such as local agricultural activities, near-by biological sources, wind direction and speed, and sunlight, should be considered in future studies to fully understand the behavior of Aspergillus/Penicillium.

Fungal concentrations were associated with several air pollutants in our study, including O3, PM10, CO, NO2,

SO2, and THC (Table 4). Ozone was negatively

correlated with total fungi, other fungi, and Ganoderma. A negative association between total fungal concentra-tions and ambient O3was also observed in several other

studies (Delﬁno et al., 1996;Lin and Li, 2000). Ozone is toxic to ambient microorganisms, especially after react-ing with atmospheric oleﬁns and formreact-ing so-called ‘‘open air factors’’ (Cox, 1995; Cox et al., 1973). On the other hand, it has been suggested that ozone might potentiate the health effect of ambient aeroallergens (Higgins et al., 2000). Therefore, the confounding effects of ozone and ambient fungi on respiratory symptoms should be examined in epidemiological studies, although the effects were not consistent in the past (Lewis et al., 2000;Rutherford et al., 2000).

In our study, PM10 was positively associated with

total fungi and Cladosporium. According to a research conducted in Queensland, Australia, fungal spores dominated the bioaerosol counts in the 2–10 mm range and fungal spores contributed approximately 5% to the total particulate mass (Glikson et al., 1995). Delﬁno et al. (Delﬁno et al., 1996) also found a positive correlation between total fungal spores and PM2.5. Therefore, it is

not surprising to observe associations between fungi and particulate matters. However, the relationships were variable in different studies (Glikson et al., 1995; Lin and Li, 2000;Wu et al., 2004). We also found fungal concentrations were associated with several other air pollutants, including CO, NO2, SO2, and THC. More

studies are needed to understand the mechanisms of these associations.

5. Conclusion

This study provided essential information of long-term fungal compositions and variations in Hualien

City, where air spora were more abundant than most of the areas in Taiwan. Signiﬁcant diurnal rhythms were found for all fungal categories and coincided with the periodicities of Night Pattern and Middle-Day Pattern. Most fungal spores varied with season and usually peaked at warmer months. Associations between ambi-ent fungi and environmambi-ental parameters were complex. Temperature was the most consistent factor positively associated with fungal levels. Several meteorological factors, such as rainfall and relative humidity, had mixed effects on fungal concentrations. Moreover, both local and long-range transported pollutants were likely to interact with aeroallergens, although most of the mechanisms remain to be studied. Because of signiﬁcant temporal variation and the complicated interrelation-ships among aeroallergens and air pollution/meteorolo-gical factors, futures studies should examine the health impacts of aeroallergens using a longitudinal study design and examining the synergistic/antagonistic effects of these environmental parameters.

Acknowledgement

This study was supported in part by the National Science Council of Taiwan (NSC-82-0211-B-152-001-B08, NSC-83-0211-B-152-001-B08, NSC-84-2311-B-152-001-B08, NSC-85-2311-B-152-001-B14) and by the Taiwan Environmental Protection Administration (EPA-93-L105-02-207).

References

Burch, M., Levetin, E., 2002. Effects of meteorological conditions on spore plumes. International Journal of Biometeorology 46, 107–117.

Burge, H.A., 2002. An update on pollen and fungal spore aerobiology. Journal of Allergy and Clinical Immunology 110, 544–552.

Burge, H.A., Otten, J.A., 1999. Fungi. In: Macher, J. (Ed.), Bioaerosols: Assessment and Control. American Conference of Governmental Industrial Hygienists (ACGIH), Cincin-nati, OH, pp. 19.1–19.13.

Burge, H.A., Rogers, C.A., 2000. Outdoor allergens. Environ-mental Health Perspectives 108, 653–659.

Bush, R.K., Portnoy, J.M., 2001. The role and abatement of fungal allergens in allergic diseases. Journal of Allergy and Clinical Immunology 107 (Suppl. 3), 430–440.

Cakmak, S., Dales, R.E., Burnett, R.T., Judek, S., Coates, F., Brook, J.R., 2002. Effect of airborne allergens on emergency visits by children for conjunctivitis and rhinitis. Lancet 359, 947–948.

Chun, Y., Kim, J., Choi, J.C., Boo, K.O., Oh, S.N., Lee, M., 2001. Characteristic number size distribution of aerosol during Asian dust period in Korea. Atmospheric Environ-ment 35, 2715–2721.

(11)

Corden, J.M., Millington, W.M., 2001. The long-term trends and seasonal variation of the aeroallergen Alternaria in Derby, UK. Aerobiologia 17, 127–136.

Cox, C.S., 1995. Stability of airborne microbes and allergens. In: Cox, C.S., Wathes, C.M. (Eds.), Bioaerosols Handbook. Lewis Publishers, New York, pp. 77–99.

Cox, C.S., Hood, A.M., Baxter, J., 1973. Method for comparing concentrations of the open-air factor. Applied Microbiology 26, 640–642.

Dales, R.E., Cakmak, S., Burnett, R.T., Judek, S., Coates, F., Brook, J.R., 2000. Inﬂuence of ambient fungal spores on emergency visits for asthma to a regional childrens hospital. American Journal of Respiratory and Critical Care Medi-cine 162, 2087–2090.

Dales, R.E., Cakmak, S., Judek, S., Dann, T., Coates, F., Brook, J.R., Burnett, R.T., 2003. The role of fungal spores in thunderstorm asthma. Chest 123, 745–750.

Dales, R.E., Cakmak, S., Judek, S., Dann, T., Coates, F., Brook, J.R., Burnett, R.T., 2004. Inﬂuence of outdoor aeroallergens on hospitalization for asthma in Canada. Journal of Allergy and Clinical Immunology 113, 303–306. Delﬁno, R.J., Coate, B.D., Zeiger, R.S., Seltzer, J.M., Street, D.H., Koutrakis, P., 1996. Daily asthma severity in relation to personal ozone exposure and outdoor fungal spores. American Journal of Respiratory and Critical Care Medi-cine 154, 633–641.

Delﬁno, R.J., Zeiger, R.S., Seltzer, J.M., D.H., S., Matteucci, R.M., Anderson, P.R., Kourtrakis, P., 1997. The effect of outdoor fungal spore concentrations on daily asthma severity. Environmental Health Perspective 105, 622–635. Di Giorgio, C., Krempff, A., Guiraud, H., Binder, P., Tiret, C.,

Dumenil, G., 1996. Atmospheric pollution by airborne microorganisms in the city of Marseilles. Atmospheric Environment 30, 155–160.

Fang, G.C., Chang, C.N., Wu, Y.S., Lu, S.C., Fu, P.P., Chang, S.C., Cheng, C.D., Yuen, W.H., 2002. Concentration of atmospheric particulates during a dust storm period in central Taiwan, Taichung. Science of the Total Environ-ment 287, 141–145.

Glikson, M., Rutherford, S., Simpson, R.W., Mitchell, C.A., Yago, A., 1995. Microscopic and submicron components of atmospheric particulate matter during high asthma periods in Brisbane, Queensland, Australia. Atmospheric Environ-ment 29, 549–562.

Gregory, P.H., 1973. The air spora near the earths surface. In: The Microbiology of the Atmosphere. Leonard Hill, Plymouth, England, pp. 146–169.

Grifﬁn, D.W., Kellogg, C.A., Garrison, V.H., Lisle, J.T., Borden, T.C., Shinn, E.A., 2003. Atmospheric microbiology in the northern Caribbean during African dust events. Aerobiologia 19, 143–157.

Han, S.H., Chuang, Y.C., 1981. Air-borne fungal spore counts in Taiwan. Chinese Medical Journal 28, 193–196. Henriquez, V.I., Villegas, G.R., Nolla, J.M.R., 2001. Airborne

fungi monitoring in Santiago, Chile. Aerobiologia 17, 137–142.

Higgins, B.G., Francis, H.C., Yates, C., Warburton, C.J., Fletcher, A.M., Pickering, C.A., Woodcock, A.A., 2000. Environmental exposure to air pollution and allergens and peak ﬂow changes. European Respiratory Journal 16, 61–66.

Katial, R.K., Zhang, Y., Jones, R.H., Dyer, P.D., 1997. Atmospheric mold spore counts in relation to meteorologi-cal parameters. International Journal of Biometeorology 41, 17–22.

Lehrer, S.B., Hughes, J.M., Altman, L.C., Bousquet, J., Davies, R.J., Gell, L., Li, J., Lopez, M., Malling, H.J., Mathison, D.A., et al., 1994. Prevalence of basidiomycete allergy in the USA and Europe and its relationship to allergic respiratory symptoms. Allergy 49, 460–465.

Lewis, S.A., Corden, J.M., Forster, G.E., Newlands, M., 2000. Combined effects of aerobiological pollutants, chemical pollutants and meteorological conditions on asthma admis-sions and A & E attendances in Derbyshire UK, 1993–96. Clinical and Experimental Allergy 30, 1724–1732. Li, D.W., Kendrick, B., 1994. Functional relationships between

airborne fungal spores and environmental factors in Kitchener–Waterloo, Ontario, as detected by Canonical correspondence analysis. Grana 33, 166–176.

Lierl, M.B., Hornung, R.W., 2003. Relationship of outdoor air quality to pediatric asthma exacerbations. Annals of Allergy, Asthma, and Immunology 90, 28–33.

Lin, W.H., Li, C.S., 2000. Associations of fungal aerosols, air pollutants, and meteorological factors. Aerosol Science & Technology 32, 359–368.

Liu, S.C., 2004. Asian dust and long-range transport of air pollutants. Workshop on Asian dust-storm study, Taipei, Taiwan. Taiwan Environmental Protection Administration/ National Science Council.

Ma, C.-J., Kasahara, M., Holler, R., Kamiya, T., 2001. Characteristics of single particles sampled in Japan during the Asian dust-storm period. Atmospheric Environment 35, 2707–2714.

Mitakakis, T., Ong, E.K., Stevens, A., Guest, D., Knox, R.B., 1997. Incidence of Cladosporium, Alternaria and total fungal spores in the atmosphere of Melbourne (Australia) over three years. Aberobiologia 13, 83–90.

Neas, L.M., Dockery, D.W., Burge, H., Koutrakis, P., Speizer, F.E., 1996. Fungus spores, air pollutants, and other determinants of peak expiratory ﬂow rate in children. American Journal of Epidemiology 143, 797–807. Nikkels, A.H., Terstegge, P., Spieksma, F.T.M., 1996. Ten

types of microscopically identiﬁable airborne fungal spores at Leiden, The Netherlands. Aerobiologia 12, 107–112.

Oh, J.W., Lee, H.B., Lee, H.R., Pyun, B.Y., Ahn, Y.M., Kim, K.E., Lee, S.Y., Lee, S.I., 1998. Aerobiological study of pollen and mold in Seoul, Korea. Allergology International 47, 263–270.

O’Hollaren, M.T., Yunginger, J.W., Offord, K.P., Somers, M.J., O’Connell, E.J., Ballard, D.J., Sachs, M.I., 1991. Exposure to an aeroallergen as a possible precipitating factor in respiratory arrest in young patients with asthma. New England Journal of Medicine 324, 359–363.

Rogers, C.A., Muilenberg, M., 2001. Pan-American Aerobiol-ogy Association Standardized Protocols—Comprehensive guidelines for the operation of Hirst-type suction bioaerosol samplers, The Pan American Aerobiology Association. http://www.paaa.org.

Rosas, I., McCartney, H.A., Payne, R.W., Calderon, C., Lacey, J., Chapela, R., Ruiz-Velazco, S., 1998. Analysis of the relationships between environmental factors (aeroallergens,

(12)

air pollution, and weather) and asthma emergency admis-sions to a hospital in Mexico City. Allergy 53, 394–401. Ross, M.A., Persky, V.W., Scheff, P.A., Chung, J., Curtis, L.,

Ramakrishnan, V., Wadden, R.A., Hryhorczuk, D.O., 2002. Effect of ozone and aeroallergens on the respiratory health of asthmatics. Archives of Environmental Health 57, 568–578.

Rutherford, S., Simpson, R., Williams, G., Mitchell, C., McCall, B., 2000. Relationships between environmental factors and lung function of asthmatic subjects in south east Queensland, Australia. Journal of Occupational and En-vironmental Medicine 42, 882–891.

Satheesh, R.P., Rao, G.R., 1994. Fungal spore concentrations in the air at Tiruchirapalli (India) 1987–1988. Aerobiologia 10, 71–75.

Shinn, E.A., Grifﬁn, D.W., Seba, D.B., 2003. Atmospheric transport of mold spores in clouds of desert dust. Archives of Environmental Health 58, 498–504.

Targonski, P.V., Persky, V.W., Ramekrishnan, V., 1995. Effect of environmental molds on risk of death from asthma during the pollen season. J Allergy Clin Immunol 95, 955–961.

Taylor, D.A., 2002. Dust in the wind. Environmental Health Perspectives 110, A80–A87.

Troutt, C., Levetin, E., 2001. Correlation of spring spore concentrations and meteorological conditions in Tulsa, Oklahoma. International Journal of Biometeorology 45, 64–74.

Tseng, H.Y., Chen, Z.C., 1979. The fungal air spora of Taipei as determined by the agar plate method. Taiwania 24, 54–63.

Wang, W.Y., Chen, T., 1979. The relationship between aero-fungi and respiratory allergy in Taipei city. Chinese Medical Journal 26, 151–159.

Weber, R.W., 2003. Meteorologic variables in aerobiology. Immunology and Allergy Clinics of North America 23, 411–422.

Wu, P.C., Tsai, J.C., Li, F.C., Lung, S.C., Su, H.J., 2004. Increased levels of ambient fungal spores in Taiwan are associated with dust events from China. Atmospheric Environment 38, 4879–4886.

Yeo, H.G., Kim, J.H., 2002. SPM and fungal spores in the ambient air of west Korea during the Asian dust (Yellow sand) period. Atmospheric Environment 36, 5437–5442.