Brenda T. Shaffera and Bruce Lighthartb
aManTech Environmental, Environmental Research Laboratory, Corvallis, OR 97333, (503)754-4562, Fax (503)754-4711, email@example.com; and
bUS Environmental Protection Agency, Environmental Research Laboratory, Corvallis, OR 97333
To determine the risks of microbial air pollution from microorganisms used for pesticides and bioremediation, or from composting, fermentation tanks, or other agricultural and urban sources, current microbial levels must be evaluated. This study surveyed the atmospheric load of bacteria at four locations in Oregon: a city street, a rye grass field, a douglas fir forest, and at the Pacific coast. Thirty-minute samples were taken using slit and six-stage cascade samplers. Samples were taken for two 10 or 24-hour periods depending on the site. Meteorological measurements were made at each location.
The quantity and type of bacteria found varied by site and location. The highest average total bacterial population was exhibited at the Urban site (715 CFU/m3) and the lowest was found at the Coast site (113 CFU/m3). Bacterial concentration in general tended to peak at sunrise and sunset, decrease during the solar noon hours, with the lowest numbers occurring between 2100-0400 hours. Bacillus was found to be the highest single genus represented at all locations (15-45%). Pigmented bacteria represented between 15-64% of the total bacteria sampled with the highest percentage of pigmented bacteria found at the Rural site and the lowest percentage at the Forest site. The majority of bacteria found were on particles greater than 3 m in diameter. The particle size distribution also varied by time of day, with an increasing percentage of smaller bacteria found in the pre-dawn and evening hours. Information gathered from this survey combined with data from the literature and future surveys will be invaluable in the detection of microbial air pollution.
Key words: airborne, bacteria, Oregon
The use of microorganisms for pesticides, bioremediation, composting, fermentation, and other agricultural and industrial uses is increasing in popularity as a safer and more economical alternative to chemical use. However, to evaluate future risks in using bacteria, current levels must be monitored. If microbial air pollution is a future risk, how can we protect our environment without knowing the current "natural" levels and types of bacteria?
Speculations of airborne microorganisms date back to Hippocrates who believed that epidemics were caused when people inhaled infected air (Gregory, 1973). However, the first long-term survey of the microbial content of the atmosphere by volumetric methods was not performed until 1883 by Pierre Miquel (Gregory, 1973), estimating airborne bacterial concentration in Paris to be 1000/m3. In more recent atmospheric surveys bacterial numbers have been estimated between 175 - 8500 colony forming units (CFU)/m3 in Johannesburg, South Africa (Yousefi and Rama, 1992), 2-4000 CFU/m3 in Sweden (Bovallius, et al., 1978), 10-150 CFU/m3 in the high desert, Washington, USA (Lighthart and Shaffer, 1994), 50-5730 CFU/m3 in Moscow (Vlodavets, 1958), and 0-18 CFU/m3 in the polar air and 1-32 CFU/m3 in the tropical air over the Atlantic Ocean (Pady and Kelly, 1953). The wide range of numbers can be attributed to location, time of day and year, type of sampler, type of media, altitude, and meteorological conditions, just to name a few variables.
This study was initiated to survey the current atmospheric load of bacteria at representative locations in Oregon, and to determine and better understand the number, type, and distribution of bacteria in the outdoor environment.
Site description. Airborne bacteria were sampled at four locations in Oregon: a city street, a rye grass field, a douglas fir forest, and at the Pacific coast. The urban location, Corvallis, OR (pop. 43,715, 22-23 September 1993) was on a two-lane one-way street, which is a major route from the south to north end of town, traffic was fairly steady throughout the day (13 cars per minute) with an average speed of 35 mph. Traffic consisted of primarily cars and pick-up trucks, with occasional utility and lumber trucks, foot traffic was light, and weather conditions were clear. The forest location was at the Oregon State University (OSU) Experimental Forest, Corvallis, OR (5 and 7 October 1993), located off-trail in a primarily douglas fir forest. Foot traffic is periodic with a single hiker or jogger. Conditions were cloudy to partly cloudy with a break in sampling on the second day of sampling due to a light rain from 1130 to 1230 hrs. The coast location was at Yaquina Head, Newport, OR, which was an isolated site located near a Coast Guard relay station on top of a bluff approximately 100 m above ocean level. Conditions were foggy both days, with winds from the northwest on Day 1 (24 September 1993) and from the southwest on Day 2 (27 September 1993), which did not affect the overall counts.
The rye grass field (165 acres; Willamette Valley, OR) was sampled at three stages of the crop, mature grass (Time-1, 14-18 June 1993), cut grass (Time-2, 2-9 August 1993), and new grass (Time-3, 28 October-11 November 1993). The nearest house was 27 m with the next house approximately 1 km from the sample site. During Time-1 and -2, conditions were clear to partly cloudy, with little to no extraneous disturbances. During Time-3, there was moderate activity on the second day by workers setting up an electric fence for sheep grazing, conditions were clear to light haze.
Sample procedure. Thirty to sixty minute samples were taken using two replicate slit (New Brunswick Scientific Co., Inc., Edison, NJ) and six-stage cascade samplers (Andersen, Atlanta, GA). Samplers contained Luria Bertani agar (Difco Laboratories, Detroit, MI) amended with 200 g/mL cycloheximide (Sigma Chemical Co., St. Louis, MO). Plates were incubated at 25C for 7 days. Samples were taken continuously over two 10 hour periods or the equivalent of two 24 hour periods depending on the site. At each location, temperature, relative humidity, solar radiation, and wind speed and direction were monitored (Campbell Scientific, Logan, UT).
Bacterial identification. Approximately 70 bacteria from each site were positively identified using the Biolog MicroStationTMSytem (v. 3.5; Biolog, Inc., Hayward, CA). Samples were taken randomly from the slit sampler plates in an effort to estimate the total population found at each site.
The highest average total bacterial population was exhibited at the Urban site (715 CFU/m3) and the lowest was found at the Coast site (113 CFU/m3). This was not surprising in that the ocean air has been shown to be relatively clean by the dilution of bacteria with no near terrestrial sources (Pady and Kelly, 1954), and the urban air was constantly being stirred up by the steady stream of automobile traffic.
Bacterial concentration in general tended to peak at sunrise and sunset, decrease during the solar noon hours, with the lowest numbers occurring between 2100-0400 hours. In the evening, a stable nocturnal boundary layer develops where there is little mixing or turbulent air. At sunrise, the ground begins to heat from solar radiation, and a mixed layer due to convective heating begins to form, increasing air turbulence (Kim, 1994). However, as solar radiation increases, the bacteria are reduced in numbers due to the damaging effects of UV radiation (Cox, 1989).
Bacillus was found to be the highest single genus represented at all locations (15-45%). Since the atmosphere is a very harsh environment for bacteria, primarily due to desiccation, it is not surprising that a spore-former represents the largest genus present. Upon grouping the bacteria, it was found that Gram (+) endospore-forming rods and cocci exhibited the highest percentages at the Forest, Urban, and Rural Time-1 sites. Gram (+) irregular, non-sporing, rods was the group with the highest percentages at the Coast, Rural Time-2, and Rural Time-3 sites. Gram (-) bacteria represented between 6-28% of the population, with the highest percentage found at the Coast site.
Pigmented bacteria represented between 15-64% of the total bacteria sampled, with the highest percentage of pigmented bacteria found at the Rural site and the lowest percentage at the Forest site. Pigmentation in bacteria has been shown to protect airborne bacteria from the effects of solar/UV radiation (Gregory, 1973 and Liu et al., 1993). The maximum solar radiation recorded at the forest site was 0.05 kW/m2 whereas 0.9 kW/m2 was the maximum solar radiation recorded at the Rural sites.
The particle size distribution varied by site with 50% of the population associated with particles greater than or equal to the following diameter: 5.0 m at Rural Time-3, 4.3 m at Urban and Rural Time-1, 3.5 m at Forest, 3.0 m at Rural Time-2, and 2.2 m at Coast site. Lighthart et al. (1993), reported that 50% of the bacteria entrained from oat plants was rafted on particles greater than 3 m in diameter, therefore, it would follow that in other terrestrial environments a similar pattern would be exhibited. The particle size distribution also varied by time of day, with an increasing percentage of smaller bacteria found in the pre-dawn and evening hours. A single bacterium can survive longer in the air under less harsh conditions (Lighthart et al., 1993).
In comparing the three time periods at the Rural site, Time-2 (cut grass, 704 CFU/m3) had significantly higher total bacterial counts than Time-1 (mature grass, 127 CFU/m3) or Time-3 (new grass, 291 CFU/m3), Time-2 also had the smallest median particle size. In harvesting the rye grass seed, chaft and cut straw were scattered back on the ground by the combine, even after the straw was bailed and removed from the area, a large percentage of small, loose, dry plant material was left behind, with no new vegetation to stabilize this material. Thus, this material was probably the source for the increased counts and smaller particle size during Time-2.
In the future, we hope to survey additional locations representing different ecoregions to determine variations in bacterial type, number, and particle size distribution. With this information a database for airborne bacteria could be established to detect microbial air pollution.
Special thanks to Lynn Bucao and Rosalind James for helping during sampling under some not so favorable conditions. Also, thanks to Willamette West Reality, OSU Experimental Forest, US Environmental Protection Agency-Newport, OSU Oceanography, US Coast Guard, Bureau of Land Management, Oregon Department of Transportation, and Gloria Lighthart for allowing us to sample on their property.
This research was funded by the US Environmental Protection Agency, Corvallis, OR and the US Army, Dugway Proving Ground, UT.
Bovallius A, Bucht B, Roffey R, Anas P (1978) Three-year investigation of the natural airborne bacterial flora at four localities in Sweden. Appl Environ Microbiol 35(5):847-852.
Cox CS (1989) Airborne bacteria and viruses. Sci Prog Oxford 73:469-500.
Gregory PH (1973) Microbiology of the atmosphere. John Wiley & Sons, New York, NY.
Kim J (1994) Atmospheric environments of bioaerosols. In: Lighthart B and Mohr AJ (ed) Atmospheric microbial aerosols: theory and application. Chapman & Hall, New York, NY, pp 28-67.
Lighthart B, Shaffer BT (1994) Bacterial flux from chaparral into the atmosphere in mid-summer at a high desert location. Atmospheric Environment 28(7):1267-1274.
Lighthart B, Shaffer BT, Marthi B, Ganio L (1993) Artificial wind-gust liberation of microbial bioaerosols previously deposited on plants. Aerobiologia 9:189-196.
Liu YT, Sui MJ, Ji DD, Wu IH, Chou CC, Chen CC (1993) Protection from ultraviolet irradiation by melanin of mosquitocidal activity of Bacillus turingiensis var. israelensis. J Invert Path 62:131-136.
Pady SM, Kelly CD (1954) Aerobiological studies of fungi and bacteria over the atlantic ocean. Can J Botany 32:202-212.
Pady SM, Kelly CD (1953) Numbers of fungi and bacteria in transatlantic air. Science 117:607-609.
Vlodavets VV, Mats LI (1958) The influence of meteorological factors on the microflora of the atmospheric air in Moscow. J Microbiology 59:539-544.
Yousefi V, Rama DBK (1992) Monitoring of air for microbial and metal contamination at selectedsites in the vicinity of Johannesburg, South Africa. The Science of the Total Environment 116:159-167.