The effects of transported Asian dust on the composition and concentration of ambient fungi in Taiwan
- First Online:
- Cite this article as:
- Chao, H.J., Chan, C., Rao, C.Y. et al. Int J Biometeorol (2012) 56: 211. doi:10.1007/s00484-011-0413-x
- 309 Views
This study was conducted to evaluate the effects of transported Asian dust and other environmental parameters on the levels and compositions of ambient fungi in the atmosphere of northern Taiwan. We monitored Asian dust events in Taipei County, Taiwan from January 2003 to June 2004. We used duplicate Burkard portable air samplers to collect ambient fungi before, during, and after dust events. Six transported Asian dust events were monitored during the study period. Elevated concentrations of Aspergillus (A. niger, specifically), Coelomycetes, Rhinocladiella, Sporothrix and Verticillium were noted (p < 0.05) during Asian dust periods. Botryosporium and Trichothecium were only recovered during dust event days. Multiple regression analysis showed that fungal levels were positively associated with temperature, wind speed, rainfall, non-methane hydrocarbons and particulates with aerodynamic diameters ≤10 μm (PM10), and negatively correlated with relative humidity and ozone. Our results demonstrated that Asian dust events affected ambient fungal concentrations and compositions in northern Taiwan. Ambient fungi also had complex dynamics with air pollutants and meteorological factors. Future studies should explore the health impacts of ambient fungi during Asian dust events, adjusting for the synergistic/antagonistic effects of weather and air pollutants.
KeywordsAerobiologyAsian dust eventsBioaerosolsCulturable fungi
Asian dust storms (yellow sand) occur frequently during late winter and spring in arid areas of East Asia, including the Gobi, Takla Makan and Ordos Deserts and the Loess Plateau (Lin 2001; Natsagdorj et al. 2003; Wang et al. 2004). With favorable meteorological conditions, suspended dusts can be transported eastward to Japan, Korea and the west coasts of North America, or southward to Taiwan and Hong Kong (Chun et al. 2001; Fang et al. 1999, 2002; Husar et al. 2001; Kanayama et al. 2002; Kwon et al. 2002; Lin 2001; Uematsu et al. 1983; Zhou et al. 1996). Even after traveling thousands of miles, dust storms can decrease visibility and adversely affect the air quality of downwind environments (Chun et al. 2001; Fang et al. 2002; Lee et al. 2006; Ma et al. 2001).
Many countries monitor the magnitude, trajectories, and constituents of Asian dust plumes in order to understand the characteristics and environmental impacts (Chun et al. 2001; Chung and Yoon 1996; Fang et al. 2002; Lin 2001; Ma et al. 2001; Qian et al. 2004; Terada et al. 2002). In Taiwan, various studies have examined the physical and chemical properties of Asian dust (Chen et al. 2004a; Cheng et al. 2005; Fang et al. 2002; Hsu et al. 2005; Lee et al. 2006; Lin et al. 2005; Liu et al. 2006), but few for biological compositions (Ho et al. 2005; Wu et al. 2004). Long-range transported African dust has been recently shown to carry a variety of live microorganisms, including fungi and bacteria, as well as virus-like particles, to the Caribbean (Griffin et al. 2001; Griffin et al. 2003; Shinn et al. 2003; Taylor 2002). Long-range transport of live macro-scale organisms, such as African desert locusts, from Africa to the Caribbean and the South America has also been reported (Rosenberg and Burt 1999). In East Asia, a few studies found that concentrations of several fungal taxa increased during Asian dust events (Ho et al. 2005; Wu et al. 2004; Yeo and Kim 2002). However, detailed information regarding the biological composition of Asian dust in Taiwan is still needed.
The environmental and health risks associated with Asian dust events are of high concern in downwind countries. Long-range transport of plant pathogens (mostly fungi) has been documented (Brown and Hovmoller 2002; Garrison et al. 2003). Asian dust events may also have adverse human health effects (Griffin 2007). In animal studies, rats experimentally exposed to Asian dust particles developed adverse cardiovascular effects and lung inflammation/injury (Chang et al. 2007; Lei et al. 2004). In epidemiological studies, Asian dust events have been implicated in increased reports of conjunctivitis, respiratory diseases, daily mortalities rates, hospital admissions for cardiovascular diseases, and emergency visits for cardiopulmonary diseases in Taipei, Taiwan (Chan et al. 2008; Chen et al. 2004b; Chen and Yang 2005; Yang et al. 2005; Yang 2006). A study in Seoul, Korea indicated that Asian dust events contribute to the increased mortality rates of advanced cardiovascular and respiratory patients (Kwon et al. 2002). Although it is still inconclusive about the mechanisms of these health impacts, allergic and toxic effects resulting from fungal spores, microorganisms, endotoxin, or chemicals (e.g., sulfates) have been suggested (Kwon et al. 2002; Lei et al. 2004; Yang 2006).
In order to fully understand the impacts of the biological compositions of Asian dust events on public health and ecosystems, the varieties and variations of microorganisms carried by Asian dust must first be determined. We conducted an environmental monitoring study to characterize ambient fungal compositions and distributions in northern Taiwan during the Asian dust periods in 2003 and 2004. Our major goal was to examine the impacts of long-range transported dust and other environmental parameters on the levels and compositions of ambient fungi in a downwind country.
Materials and methods
Fungal sample collection and analysis
We monitored Asian dust events from January 2003 to June 2004. Air samples were collected at two ambient air quality monitoring stations of the Taiwan Environmental Protection Administration (Taiwan EPA) in Shin-Juang City (SJCity) (121°27′E, 25°02′N) and Shihmen Township (SMTown) (121°06′E, 25°03′N) in Taipei County, Taiwan. SJCity monitoring station is located in a metropolitan area with heavy traffic flow. SMTown station is located in a rural area, on the seashore of the East China Sea with minimal local air pollution.
We used duplicate battery-powered multi-holed single-stage impactors (Burkard portable samplers for agar plates; Burkard Manufacturing, Rickmansworth, UK) and Malt Extract Agar (MEA), a nonselective, broad-spectrum culture medium, to collect ambient fungi. Duplicate 2-min samples were collected each time with a flow rate of 20 l/min, oriented to the direction of the wind. Sampling periods for each Asian dust event included before, after, and during episodes, as determined by Taiwan EPA forecast. We collected air samples 6 times a day (twice each in the morning, afternoon and evening) in SMTown, and 4 times a day (twice each in the morning and afternoon) in SJCity. All collected samples were immediately shipped to the laboratory, processed, and incubated at room temperature for 7–10 days before enumeration and identification. All fungal colonies were counted and identified to genus and/or species level. We used a positive-hole correction table to adjust colony counts and corresponding concentrations (Willeke and Macher 1999). Concentrations are reported in colony-forming units per cubic meter of air (CFU/m3). The averages of duplicates were used for all statistical analyses.
Environmental parameters data collection
We obtained air pollution and meteorological data of both monitoring stations from the Taiwan EPA. Meteorological data included temperature, relative humidity (RH), dew point, rainfall, and wind speed. Air pollutant data included sulfur dioxide (SO2), carbon monoxide (CO), ozone (O3), particulates with aerodynamic diameters ≤10 μm (PM10), nitrogen monoxide (NO), nitrogen dioxide (NO2), methane (CH4), and non-methane hydrocarbons (NMHC). We used the hourly averages of these environmental parameters measured concurrently with fungal sampling for further analysis.
We used the SAS statistical package (v.8.2, SAS Institute, Cary, NC, USA) to perform data analyses. Wilcoxon–Mann–Whitney U test and exact test (if the recovery frequency was less than 10 for either background or episode period) were used to evaluate the impact of Asian dust events on fungal concentrations and other environmental factors. We used multiple regression analyses to examine the relationships between ambient fungi and environmental parameters, using all collected data. We developed regression models for total fungi and major fungal categories. To account for the serial correlations of fungal measurements, we used PROC MIXED procedure in SAS with an exponential correlation covariance model. To perform regression analysis, fungal concentrations were transformed using base-10 logarithm to approximate normality. In order to perform log transformations, we added 1 to the concentrations in the major fungal categories to avoid zero values.
Distribution of airborne culturable fungi (CFU/m3) from March 2003 to April 2004 in Taipei County, Taiwan
Distributions of selected fungi and environmental factors during background and episode periods in Taipei, Taiwan
Background period (n =167)
Episode period (n = 128)
p value b
Freq (%) a
Freq (%) a
Wind speed (m/s)
Multiple regression models for major fungal taxa in Taipei, Taiwan
Few studies have examined the effect of Asian dust storms on airborne fungal ecology. In Seosan, Korea, more diverse culturable fungi were recovered during Asian dust events including Fusarium, Aspergillus, Penicillium and Basipetospora (Yeo and Kim 2002). The recovery rates, however, of many fungal taxa were low, possibly because the collection method (a cascade impactor with membrane filters) may underestimate culturable fungal concentrations. A study conducted in eastern Taiwan found that total fungi, Cladosporium, Ganoderma, Arthrinium/Papularia, Cercospora, Periconia, Alternaria and Botrytis had statistically significant higher concentrations during the periods affected by Asian dust events (Ho et al. 2005). Wu et al. (2004) found basidiospores, Aspergillus/Penicillium, Nigrospora, Arthrinium, Curvularia, rusts, Stemphylium, Cercospora, Pithomyces and unidentified fungi had higher levels during dust storm days than background days in southern Taiwan. Both these studies conducted in Taiwan collected fungal spores using Burkard volumetric spore traps and concordantly observed higher levels of Cercospora and Arthrinium during Asian dust periods than during non-dust periods. Our study of northern Taiwan found Aspergillus niger, Coelomycetes, Rhinocladiella, Sporothrix and Verticillium had significant higher levels (p < 0.05) during the episode periods. Botryosporium and Trichothecium were only recovered during the Asian dust periods. The discrepancies between our findings and other studies conducted in Taiwan are mainly due to different sampling/analyzing methods (i.e., culture-based vs. direct microscopy methods) and geographical regions.
The most prevalent fungi observed during dust events in our study were Cladosporium, non-sporulating fungi, Penicillium and Aspergillus which is generally consistent with previous studies of Asian and African dust events. Cladosporium and Penicillium were commonly observed in both Asian and African dust events (Abdel 2003; Griffin et al. 2003; Ho et al. 2005; Kellogg et al. 2004; Shinn et al. 2003; Weir-Brush et al. 2004; Wu et al. 2004; Yeo and Kim 2002). Several studies found more abundant Cladosporium during dust events (Abdel 2003; Ho et al. 2005). Non-sporulating fungi, such as basidiospores, do not produce spores under the culture conditions provided. Wu et al. (2004), using a non-culture based method, found that the level of basidiospores increased during Asian dust periods in southern Taiwan. Aspergillus is consistently the most frequently recovered fungal genus in both Asian and African dust events. In our study, A. niger was specifically identified in dust events. A. niger is a ubiquitous species, usually isolated from soils, plant litter, plant rhizospheres, seeds, dried fruits and nuts. It is one of the most commonly reported fungi from food, indoor environments, and outdoor niches, occasionally causing aspergillosis in immunocompromised patients (Klich 2002; St-Germain and Summerbell 1996). A. niger is more prevalent in warmer climates because its melanin-containing spores grant protection from sunlight, providing a competitive advantage in such habitats (Pitt and Hocking 1999).
Previous studies found that the microorganisms recovered in African dust storms were mostly pigmented, which might be a survival mechanism for microbes to protect against ultraviolet (UV) radiation while airborne (Griffin et al. 2001; Kellogg et al. 2004). A variety of fungi recovered in Asian dust events also have dark spores (Ho et al. 2005; Wu et al. 2004). UV resistance in microorganisms apparently plays an important role in their survival during atmospheric transport. Researchers also suggest that the particles in upper dust plumes provide UV light shielding for the microorganisms at lower altitudes (Griffin et al. 2001; Herman et al. 1999). Microbes may also avoid exposure to UV light within cracks and crevasses of inorganic dust particles (Garrison et al. 2003; Griffin et al. 2001; Herman et al. 1999). Microbial characteristics (e.g., pigmented propagules) and the protection provided by dust plumes help protect microorganisms from desiccation and UV exposure during long-range atmospheric transport.
Although the existence of live microorganisms in long-range transported events is documented, the mechanisms of ambient fungal concentration/composition shifts during dust periods are still not clear. It is possible that foreign microorganisms are picked up at the source areas of dust storms and then transported overseas. Asian dust may also commingle with local bioaerosols while traversing mainland China before reaching Taiwan (Shinn et al. 2003). Higher wind speeds during Asian dust events might also play a role. Increased wind velocity could promote aerosolization of phylloplane and soil fungi from their reservoirs. Our study found higher wind speed was significantly and positively correlated with Cladosporium and Curvularia. Therefore, local microorganisms may be aerosolized by prevailing northeasterly during Asian dust events and contribute to ambient fungal flora (Griffin et al. 2003). For future studies, sampling at both source and effected areas of Asian dust storms, as well as DNA sequencing and genetic comparisons, will be valuable measures to verify the origins of microorganisms in Asian dust (Kakikawa et al. 2008).
In addition to fungal concentration and composition, other environmental parameters also vary during Asian dust periods. Kwon et al. (2002) found that PM10 levels increased and temperature decreased during the Asian dust periods observed from 1995 to 1998 in Seoul, South Korea. In 2002, a study of 7 episodes of Asian dust events in northern Taiwan conducted at the same monitoring location in SJCity found that PM2.5–10 and wind speed were higher and temperature was lower during Asian dust periods, compared to before and after the dust events (Lee et al. 2006). However, several air pollutants, including PM2.5 and the major PM2.5 species (i.e., carbons, sulfate and nitrate) had highest concentrations after the dust periods because of the accumulation of particles influenced by the anticyclonic outflow. In our study, similar shifts of PM10 levels, wind speed and temperature during Asian dust events were observed. The concentrations of SO2, CO, NO2 and NMHC were lower during Asian dust periods (Table 2), possibly due to the accumulation of local pollution after dust events caused by atmospheric stagnancy brought about by the anticyclonic outflow (Lee et al. 2006). Therefore, it is essential to note that air quality is affected not only during Asian dust periods but also for a time period afterwards.
Ambient fungi also have complex interactions with meteorological parameters, local air pollutants and sampling location (Burge and Rogers 2000; Ho et al. 2005; Jones and Harrison 2004; Lin and Li 2000; Wu et al. 2007). Wu et al. (2007) found that total airborne fungal concentrations and diversity of fungal species (as measured by the number of fungal taxa identified) were generally higher in urban than in rural areas. The differences are likely due to different distributions of local vegetation, types of human activities (e.g., local turbulence in high traffic locales resulting in spore aerosolization), and characteristics of meteorological factors (e.g., dilution effects of higher wind speed). Both temperature and wind speed were positively correlated with fungal levels in our study (Table 3). The relationship between wind speed and fungal concentrations has been discussed above. Temperature is one of the most essential environmental factors affecting fungal growth and survival (Burge and Otten 1999). Dry and hot weather also facilitates the liberation of dry-air spora (e.g., Cladosporium) (Burch and Levetin 2002; Burge and Rogers 2000; Troutt and Levetin 2001). In our study, ambient temperature was positively associated with total fungi, Cladosporium, and non-sporulating fungi, consistent with findings observed worldwide (Burch and Levetin 2002; Hollins et al. 2004; Katial et al. 1997; Sabariego et al. 2000). Rainfall and relative humidity were also associated with ambient fungal levels in our study. Both these factors are correlated with water availability in ambient environments, which is a determinant for fungal growth (Burge and Otten 1999). Mixed effects of these two factors on ambient fungal levels have been documented (Ho et al. 2005; Katial et al. 1997; Li and Kendrick 1994; Mitakakis et al. 1997). Rainfall causes release of spores by splash and tap-and-puff mechanisms. Rain also removes ambient particles by rainout and washout effects. Wet-air spora, such as ascospores and basidiospores, require increased moisture or rainfall to release (Burge and Rogers 2000; Weber 2003). Dry-air spora, on the other hand, are usually more abundant when humidity is low (Troutt and Levetin 2001). In this study, rainfall had positive but relative humidity had negative relationships with Curvularia and Alternaria, which are both dry-air spora. The interaction of rainfall and relative humidity was not significant in these regression models.
Several air pollutants were significantly associated with ambient fungi in our multiple regression models. Ozone was negatively associated with total fungi, non-sporulating fungi and Penicillium in our study (Table 3). Similar results have been observed previously (Delfino et al. 1996; Ho et al. 2005; Lin and Li 2000). Ambient ozone interacts with atmospheric olefins, forming a so-called open air factor, which is toxic to ambient microorganisms (Cox 1995; Cox et al. 1973). Previous studies indicated that fungal spores comprise 5 to >20% of total ambient particulates (Battarbee et al. 1997; Glikson et al. 1995). We also observed positive relationships between PM10 and fungi, including Alternaria and Aspergillus. This finding further supports the fact that viable microorganisms are transported by Asian dust events. The interrelationships between fungal concentrations and other air pollutants have also been observed in our and other studies (Ho et al. 2005; Lin and Li 2000). However, the mechanisms of the associations are still not clear and should be further examined. Atmospheric mixing height could be one of the reasons that fungal concentrations fluctuate similarly to other air pollutants.
It is of note that in this study we collected grab samples during the study periods (i.e., duplicate 2-min samples collected 4 and 6 times a day at 2 sampling sites) and used a culture-based method for fungal analysis. Therefore, our results might not reflect the full spectrum of ambient fungi in comparison to using a continuous Hirst-type volumetric sampler. Culture-based methods, however, produce more accurate identification of fungal taxa (generally to genus or species) than direct spore microscopy. For example, only culture-based methods (as used in our study) or PCR methods could have differentiated spores of Aspergillus and Penicillium.
This study demonstrated that atmospheric concentrations and diversity of fungi in northern Taiwan were influenced by Asian dust events. Although detailed mechanisms of the increased fungal varieties and levels during dust events is not understood, the resulting health impacts on susceptible populations (e.g., patients with allergies or respiratory diseases) during and after Asian dust events warrant further research. Because ambient fungi, local air pollution/meteorological factors, and long-range transported events have complicated interrelationships, carefully designed longitudinal studies are needed to clarify the synergistic/antagonistic effects of these environmental parameters on health.
We conducted a longitudinal monitoring study to characterize the effects of Asian dust events and other environmental parameters on the levels and compositions of ambient fungi in northern Taiwan. Several fungi were more prevalent during Asian dust periods, including Aspergillus (A. niger, specifically), Coelomycetes, Rhinocladiella, Sporothrix and Verticillium. Botryosporium and Trichothecium were only recovered during dust event days. Several environmental factors also varied during Asian dust events, including temperature, wind speed, PM10, SO2, NO2 and NMHC. Temperature and O3 were the most consistent parameters associated with fungal levels. Ambient fungal concentrations increased with higher temperature and lower O3 levels. Wind speed, rainfall, relatively humidity, NMHC and PM10 were also important predictors of fungal levels. This study demonstrated that Asian dust events affected ambient fungal levels and compositions in northern Taiwan. Moreover, ambient fungi had complex interrelationships with local meteorological factors and air pollutants. Future studies should be conducted to explore the health and ecological impacts of ambient fungi, especially during Asian dust events.
This study was supported in part by Environmental Protection Administration, Executive Yuan, Republic of China (EPA-92-U1L1-02-101 and EPA-93-L105-02-207).
The experiments of this study comply with the current laws of the Republic of China in which they were performed.