3
Indoor and Outdoor Bioaerosol Backgrounds and Sampling Strategies

Understanding the normal state of the bioaerosol environment is an essential precursor for detecting unusual concentrations or populations that would be associated with a release of biological agents. As discussed below, not only can natural bioaerosol backgrounds be large compared to the biothreat agents that must be distinguished, but concentrations can also vary rapidly over a wide range. Unless these variations are understood and accounted for, they could overwhelm nonspecific detectors or result in an unacceptable rate of false alarms. As discussed below, a major problem with existing data on bioaerosol concentrations is that the data are generally reported as time averages. Few studies have focused on maximum concentrations, which are critical for predicting detector false alarm rates.

This chapter begins with a discussion of the various types of organisms and their by-products that make up bioaerosols. Both traditional and new methods of detection are discussed, as are associated sampling errors. Consistent with the two basic scenarios for biological agent releases examined in this report (see Chapter 2), both indoor and outdoor bioaerosol backgrounds are considered, including current data on such aerosols, areas where critical data are lacking, and what approaches might be taken to provide the needed data. Current methods used for control of natural aerosols in buildings are discussed. Finally, the committee offers its findings and recommendations.

ORGANISMS AND PARTICLES

Environmental aerosols contain numerous different kinds of particles of different sizes. As an example, biological particles that may be present in indoor air are listed in Table 3.1. Inadequate data have been published to estimate or predict concentrations of most particle types. Question marks indicate areas where no size data are available. For the purposes of detection, all of the organisms share some characteristics. They contain organic carbon, amino acids, DNA and/or RNA, and other materials indicative of their biological origin.

The actual particles that are present in the air may be single organisms, groups of organisms glued together with mucous secretions or other materials (e.g., droplet nuclei), single or grouped spores, fragments of organisms, or organic or inorganic rafts bearing one to many organisms.1

1  

T.M. Madelin and H.E. Johnson. 1992. Fungal and actinomycete spore aerosols measured at different humidities with an aerodynamic particle sizer. J. Appl. Bacteriol. 72:400-409.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases 3 Indoor and Outdoor Bioaerosol Backgrounds and Sampling Strategies Understanding the normal state of the bioaerosol environment is an essential precursor for detecting unusual concentrations or populations that would be associated with a release of biological agents. As discussed below, not only can natural bioaerosol backgrounds be large compared to the biothreat agents that must be distinguished, but concentrations can also vary rapidly over a wide range. Unless these variations are understood and accounted for, they could overwhelm nonspecific detectors or result in an unacceptable rate of false alarms. As discussed below, a major problem with existing data on bioaerosol concentrations is that the data are generally reported as time averages. Few studies have focused on maximum concentrations, which are critical for predicting detector false alarm rates. This chapter begins with a discussion of the various types of organisms and their by-products that make up bioaerosols. Both traditional and new methods of detection are discussed, as are associated sampling errors. Consistent with the two basic scenarios for biological agent releases examined in this report (see Chapter 2), both indoor and outdoor bioaerosol backgrounds are considered, including current data on such aerosols, areas where critical data are lacking, and what approaches might be taken to provide the needed data. Current methods used for control of natural aerosols in buildings are discussed. Finally, the committee offers its findings and recommendations. ORGANISMS AND PARTICLES Environmental aerosols contain numerous different kinds of particles of different sizes. As an example, biological particles that may be present in indoor air are listed in Table 3.1. Inadequate data have been published to estimate or predict concentrations of most particle types. Question marks indicate areas where no size data are available. For the purposes of detection, all of the organisms share some characteristics. They contain organic carbon, amino acids, DNA and/or RNA, and other materials indicative of their biological origin. The actual particles that are present in the air may be single organisms, groups of organisms glued together with mucous secretions or other materials (e.g., droplet nuclei), single or grouped spores, fragments of organisms, or organic or inorganic rafts bearing one to many organisms.1 1   T.M. Madelin and H.E. Johnson. 1992. Fungal and actinomycete spore aerosols measured at different humidities with an aerodynamic particle sizer. J. Appl. Bacteriol. 72:400-409.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases TABLE 3.1 Biological Particles That May Be Present in Indoor Air and Their Sizes Source Organism Particle Size Range (μm) Common Examples Unique Characteristics Virus One or more virions in droplet nucleus <0.1-3 Influenza Particle may be much larger than organism; RNA or DNA, not both Mycoplasma One or more organisms in droplet nucleus 1-5 M. pneumoniae No cell wall Chlamydia One or more organisms in droplet nucleus 1-5 Chlamydia psittasi   Rickettsia One or more organisms in droplet nucleus 1-5 Coxiella burnetii Obligate intracellular pathogen Bacteria One or more bacteria in droplet nucleus or on a raft 1-5 Micrococcus luteus Variable in size, shape, cell wall composition   Single or grouped dry spores 0.5-5 3 Bacillus cereus Thermoactinomyces Highly resistant endospores   Cell wall fragments <0.1     Algae One or more cells 5-10 Chlorococcus Chlorophyll, cellulose Nonvascular plants One or more spores 15-30 Mosses Chlorophyll, cellulose Vascular plants Spore 15-30 Lycopodium ferns     Pollen 10-50 Trees, grasses, weeds Sporopollenin   Pollen allergens ?       Hairs 10-100   Cellulose   Fragments ? Soy beans Cellulose Arthropods Fragments ? Cockroach, dust mite Chitin   Fecal material 20-30     Animals Fragments ? Cats, dogs, mice Keratin   Skin scales 10-50     Fungi One or more spores 1.5-100 Mushrooms, aspergillus Chitin   One or more hyphae 1.5-100   Ergosterol   Fragments ?   1-3 β-d-glucan Particle sizes range from less than 0.1 micrometers to more than 100 micrometers and can vary with relative humidity. Spores are generally smaller in air than when mounted in liquid media for microscopy. In addition to particles directly derived from living organisms, other particles in air may also share properties with the bioaerosols. Some examples include latex particles and combustion products derived

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases from burning biological materials (e.g., tobacco). Some of these particles may carry other biological material (e.g., allergens).2 No single sampling or analysis approach will accurately measure the concentrations of all biological particles. The types and concentrations of particles recovered depend on both the size selectivity of the sample collector and the sensitivity and specificity of the analytical method. Although samplers can be designed to collect particles over a relatively broad range of sizes (see Chapter 4), those currently in use are size-selective and misrepresent the true size distribution of the aerosol. Most bioaerosol samples are collected with impactors, impingers, or filters, depending on the type of target particle and the preference of the investigator. Each of the collection devices has a value of d50—that is, the aerodynamic diameter3 that is collected with 50 percent efficiency. Good samplers have a steep diameter-efficiency curve, so that 100 percent efficiency is only a small step above the d50. Because bioaerosols have a broad range of sizes—between 1 and 100 micrometers (μm)—the d50 and the steepness of the efficiency curve are extremely important factors. The Andersen cascade impactor is commonly used to collect both indoor and outdoor culturable aerosols. It is highly efficient down to a particle size of 0.1 μm and collects those above 10 μm with reasonable efficiency provided the unit is wind-oriented and the inlet air speed is close to the suction speed. However, some investigators use other devices (with much lower efficiencies) for sample collection. For example, a centrifugal sampler with poor (less than 10 percent) collection efficiency for particles smaller than 5 μm has been used in a number of studies. Thus, the data from these studies underestimate the actual concentrations of culturable organisms, and the underestimates are larger for small particles. The rotorod sampler (a rotating arm impactor with a d50 of about 20 μm) is commonly used in the United States for outdoor pollen and fungal spore collection. It is excellent for pollen but drastically underestimates concentrations of the majority of (smaller) fungal spores. The Burkard spore trap is another commonly used instrument for measuring pollen and spore concentrations. It is reasonably efficient for spores, with a d50 of about 2.5 μm or less, depending on the slit width. It is also efficient for pollen providing the sampler is properly wind-oriented and the wind speed is not too high. Indoors, several impaction spore traps and filtration methods are used. The slit samplers (Burkard, Air-O-Cell, Allergenco) have d50s in the range 2.5 to 5 μm. Thus, while spore concentrations are underestimated to some extent, the fact that nonculturable spores can be counted probably far outweighs the losses of smaller spores. Filtration devices collect all particles larger than the pore size by interception and, in addition, very small particles by diffusion. However, they are not amenable to most of the traditional analysis methods. Cells trapped on a filter tend to dry out and die in the flowing airstream, and microscopy is difficult due to the large area and nonrandom deposition of cells. These may well be the samplers of the future, when dioxyribonucleic acid (DNA) or immunological (structure-based) assays are the norm. ANALYTICAL METHODS All methods for sample analysis are selective. Table 3.2 outlines data derived from a range of sample analysis approaches and notes various limitations of each. Errors introduced by selective analytical methods far outweigh errors due to sample collection efficiencies. All analytical methods are selective in some way, and to suggest that any one approach gives information on total organisms is indefensible. One of the most commonly used analytical methods involves cell culture. However, this method only recovers organisms that are viable, that can grow on the culture medium and under the conditions 2   R.B. Knox, C. Suphioglu, P. Taylor, R. Desai, H.C. Watson, J.L. Peng, and L.A. Bursill. 1977. Major grass pollen allergen Lol p 1 aerodynamic particle sizer. J. Appl. Bacteriol. 72:400-409. 3   The aerodynamic diameter of an arbitrary particle is the diameter of a sphere with density of 1 g/cm3 that settles at the same terminal velocity as the particle in question.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases TABLE 3.2 Limitations of Some Common Analysis Methods Analysis Method Types of Organisms Counted Limitations Culture Living organisms capable of growing under conditions provided, e.g., Aspergillus fumigatus Underestimates concentration of all organisms; nonculturable organisms are invisible; nonculturable is not the same as noninfective Microscopy Morphologically identifiable particles, e.g., grass pollen Identification limited to groupings of organisms (genera, groups of genera, spore types) Immunoassay Particles with specific epitopes matching the assay antibodies, e.g., Alternaria allergen Limited to organisms for which assays have been designed; cross-reactivity common PCR DNA matching the assay DNA, e.g., Bacillus anthracis DNA Limited to organisms for which assays have been designed; very specific; very sensitive Chemical assays Biomass of specific chemical, e.g., total ergosterol or ATP Indicators for large groups of organisms provided, and that can compete with other organisms in the culture. This means that only the fastest growing organisms are recovered, and only those that are easiest to culture. Data documenting these relationships are abundant. For example, Reponen et al. recovered only 0.5 percent of Thermoactinomyces vulgaris by cultural methods compared with direct counting.4 In animal handling facilities, where an abundance of living organisms is to be expected, Eduard and Lacey recovered only 0.5 percent of the bacteria and 3 percent of the fungi detected by direct count methods.5 The culturability also varies with a number of factors, so that the errors introduced are not constant.6 This variability in errors over time is exemplified by a study in Oregon in which fluxes of bacteria from agricultural fields were assayed by three methods: total cell counts (epifluorescence microscopy), total culturable bacteria (impinger), and size-selective cultural (cascade impactor) samplers. Differences in cell concentrations were documented among the methods; these differences varied with season and weather conditions.7 Prime examples of the selectivity of culture are represented by Legionella and by Pneumocystis carinii. Legionella is a bacterium with very stringent cultural requirements; it is never recovered on ordinary laboratory media and does not compete well even when ideal conditions are provided. Thus, it is still not clear how many cells are released from even the most heavily contaminated reservoir. Pneumocystis carinii is a nonculturable fungus that is responsible for much of the pneumonia in AIDS patients. It was predicted to be an airborne disease based on epidemiological theories and was only recently identified in air samples following development of DNA probes for analysis. Because of these problems, all cultural data represent underestimates of culturable organisms and completely miss those that are not culturable. These errors are probably at least two or three orders of magnitude. 4   T.A. Reponen, S.V. Gazenko, S.A. Grinshpun, K. Willeke, and E.C. Cole. 1998. Characteristics of airborne actinomycete spores. Appl. Environ. Microbiol. 64:3807-3812. 5   W. Eduard, J. Lacey, K. Karlsson, U. Palmgren, G. Strom, and G. Blomquist. 1997. Evaluation of methods for enumerating microorganisms in filter samples from highly contaminated occupational environments. Am. Ind. Hyg. Assoc. J. 51:427-436. 6   C. Beggs. 2002. A quantitative method for evaluating the photoreactivation of ultraviolet damaged microorganisms. Photochem. and Photobio. Sci. 1:431-437. Y. Tong and B. Lighthart. 1998. Effect of simulated solar radiation on mixed outdoor atmospheric bacterial populations. FEMS Microbiol. Ecol. 26:311-316. J.F. Heidelberg, M. Shahamat, M. Levin, I. Rahman, G. Stelma, C. Grim, and R.R. Colwell. 1997. Effect of aerosolization on culturability and variety of gram-negative bacteria. Appl. Environ. Microbiol. 63:3585-3588. 7   Y. Tong and B. Lighthart. 2000. The annual bacterial particle concentration and size distribution in the ambient atmosphere in a rural area of the Willamette Valley, Oregon. Aerosol Sci. Tech. 32:393-403. R.A. Haugland, J.L. Heckman, and L.J. Wymer. 1999. Evaluation of different methods for the extraction of DNA from fungal conidia by quantitative competitive PCR analysis. J. Microbiol. Meth. 37:165-176.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases Microscopic spore counting also is selective and is restricted to those particles that are morphologically distinctive. Thus, some inorganic particles resemble spores, and some spores resemble inorganic particles. Also, microscopic examination offers relatively little opportunity for accurate identification of most particles. Pollen grains are usually classifiable only by genus, and in some cases only by groups (e.g., grass pollen). Basidiospores and ascospores, the most abundant spore types present in outdoor air, are usually identifiable only as a member of one or the other class, and errors are significant for the smallest and least distinguished of these. A few fungi have distinctive spores that are identifiable to species (e.g., Epicoccum nigrum). Most others can be placed only in generic categories. Chemical assays for cell wall components (e.g., glucans), endotoxins, or membrane components (e.g., ergosterol) have been used for analysis but do not provide particle type or size information and recognize all organisms with these components with no differentiation. Haugland et al.8 have developed a series of polymerase chain reaction (PCR) primers (see Chapter 6) for use in a quantitative method for detecting specific fungi. However, they have only been used for dust samples at this point. Haugland et al.9 and Schafer et al.10 have developed and used a PCR-based method for measurement of Mycobacterium concentrations, but this approach has not come into common use for other bacteria. Immunoassays also exist for many microbial components. However, none of these new approaches has been used to document background concentrations of bioaerosols. Rotorods collect integrated samples over (usually) 24 hours. Spore traps usually offer some time discrimination capability, but the samples are usually analyzed in 24-hour units and only 24-hour averages reported. This has the effect of underestimating the impact of pollen and spores as background aerosol particles, since the particles tend to be released during relatively short periods of time, between which concentrations can be essentially zero. Peaks may be several orders of magnitude higher than average levels so that, while average levels may range from 1 to 10 particles per liter, peaks may reach 100 or even 1,000 particles or more per liter. SOURCES OF BIOAEROSOLS Estimating or formally predicting the nature of bioaerosol populations depends on recognizing environmental sources and their characteristics. Table 3.3 summarizes some of these sources for outdoor aerosols. Specific particle types are discussed below. Outdoor Pollen Pollen concentrations are seasonal and depend on the distribution and life cycle of source plants. Generally, there are three major pollen seasons: trees (spring), grass (late spring/early summer), and weeds (summer/early fall). There is a period during midsummer when pollen levels can be quite low, and counts remain near zero during the winter in climates with freezing weather. Pollen is usually released during specific hours (often early in the morning), but pollen peaks may occur much later.11 Release times depend on cycles of pollen production, mechanisms of pollen release, and on secondary aerosolization. For example, ragweed pollen is released in the still morning hours as plants begin to dry. Pollen falls onto subtending plant surfaces from which most aerosols are formed as a result of afternoon wind action. On the other hand, mountain cedar pollen is shaken directly into the air from pollen sacs, often forming visible clouds with pollen concentrations well in excess of 10,000 particles per liter. 8   Haugland et al., 1999. See note 7 above. 9   Haugland et al., 1999. See note 7 above. 10   M.P. Schafer, J.E. Fernback, and S.A. Jensen. 1998. Sampling and analytical method development for qualitative assessment of airborne mycobacterial species of the Mycobacterium tuberculosis complex. Am. Ind. Hyg. Assoc. J. 59:540-546. 11   J. Norris-Hill. 1999. The diurnal variation of Poaceae pollen concentrations in a rural area. Grana 38:581-585.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases TABLE 3.3 Overview of Sources for Outdoor Bioaerosols Particle Natural Sources Source Characteristics Man-made Sources Source Characteristics Viruses Infected organisms Probably ubiquitous Sewage, other? Point sources; sporadic Bacteria and related particles Living leaf surfaces Ubiquitous Sewage Point or line sources, sporadic Dead leaf surfaces   Compost     Water   Cooling towers         Biopesticides   Fungal particles Mushrooms, puffballs Ubiquitous Compost Point sources, variable   Infected plants   Infected agricultural products     Dead plants         Fecal material (i.e., animal droppings)   Colonized dead field crops     Water   Stored dead organic material (grain, straw, etc.), biopesticides     Soil     Pollen Vascular plants Ubiquitous Agricultural plants Large point sources, variable Other plant particles Ferns, mosses, clubmosses, horsetails, mosses, liverwort, algae Cosmopolitan, variable Horticulture, aquaculture, seed/grain powders Small area sources, variable Other aerosols Arthropods Cosmopolitan, variable Sewage, stored food Point sources The condition of pollen sources strongly affects pollen production. Recent studies have reported the effects of El Niño on pollen concentrations (it increases them) and the similarly positive effects of increasing CO2 in ambient air.12,13 Some sources (e.g., trees) are driven by conditions pertaining during the previous season while others (grasses, weeds) are more driven by current conditions. Weather strongly affects airborne pollen concentrations as well, with levels near zero during precipitation events. Pollen clouds can be transported for long distances. 14,15 Events that can trigger large pollen releases generally involve disturbance of pollen reservoirs. For example, mowing a field of grass during active pollination will lead to sharp increases in pollen counts and probably in small particle pollen aerosols. The size distribution of pollen aerosols may shift toward smaller particle sizes during sharp changes in humidity such as occur during thunderstorms; this effect is due to release of starch grains as the internal pressure in the pollen increases due to water absorption. As mentioned above, sudden wind events can also trigger massive short-term pollen releases. 12   H.B. Freye, J. King, and C.M. Litwin. 2001. Variations of pollen and mold concentrations in 1998 during the strong El Niño event of 1997-1998 and their impact on clinical exacerbations of allergic rhinitis, asthma, and sinusitis. Allergy Asthma Proc. 22:239-247. 13   J. Emberlin. 1994. The effects of patterns in climate and pollen abundance on allergy. Allergy 49(18 Supp):15-20. 14   P.V.d. Water and E. Levetin. 2001. Contribution of upwind pollen sources to the characterization of Juniperus ashei phenology. Grana 40:133-141. 15   H.A. Burge and C.A. Rogers. 2000. Outdoor allergens. Environ. Health. Persp. 108 Suppl 4:653-659.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases Pollen Grain Concentrations A reasonable amount of data has been published concerning whole pollen concentrations throughout the world. In the United States, pollen prevalence patterns have been monitored for many years by the American Academy of Allergy, Asthma, and Immunology, in Milwaukee, Wisconsin.16 Airborne pollen concentrations and types are seasonal and geographically variable.17 Reported pollen concentrations depend primarily on the quality of the counting, which requires training and patience. Two sites in a single city may report drastically different concentrations, especially for specific pollen types. On the other hand, on a weekly average basis, two geographically similar cities may have highly correlated pollen concentrations.18 Many pollen concentration studies have been published. Virtually all are based on 24-hour average pollen concentrations. Yearly peaks for pollen generally fall into the 300 to 500 particles per cubic meter range. However, as stated above, airborne pollen concentrations are diurnal, with peaks occurring over a relatively small portion of a 24-hour day. Thus, minute-by-minute peaks can be very high (more than 10,000 per cubic meter) while the remainder of the day may be near zero. Pollen Allergen Concentrations (Including Small Particles) Seasonal prevalence patterns for outdoors as determined by pollen counts are probably indicative of both outdoor small-particle allergen prevalence and indoor levels of pollen and pollen allergens.19 A number of studies have related rainfall and release of allergen from pollen grains.20 Data on relationships to asthma indicate either a long-delayed response to pollen exposure or release of allergen during rainfall.21 Measured allergen concentrations in Melbourne were 6 to 15 nanograms per cubic meter during the grass pollen season. Allergen peaks more or less parallel pollen peaks but tend to follow them (on a 24-hour average basis).22 Microscopic visualization of pollen allergen has been accomplished using spore trapping and immunoassays. Quantitative estimates of allergen concentration were made, and allergen was associated with pollen grains, pauci-micronic particles, and fungal spores. Outdoor Fungi Fungi colonize most living leaves and all dead ones. Leaf populations are readily released into the air with wind and rain splash, with different organisms/particles being released under different environmental conditions. Generally, dry fungal spores are most likely to be released in dry windy weather, while wet spores become abundant during damp and wet weather. Soil is another important source for all kinds of biological particles, although aerosolization from plant sources is probably more important. Aerosols are released from soil during mechanical disturbance (rain splash, wind, human and 16   Available online at http://www.AAAAI.org. Accessed August 2003. 17   H.A. Burge and C.A. Rogers, 2000. See note 15 above. 18   D.J. Dvorin, J.J. Lee, G.J. Belecanech, M.F. Goldstein, and E.H. Dunsky. 2001. A comparative, volumetric survey of airborne pollen in Philadelphia, Pennsylvania (1991-1997). Ann. Allerg. Asthma Im. 87:394-404. 19   M.K. Agarwal, J.W. Yunginger, M.C. Swanson, and C.E. Reed. 1981. An immunochemical method to measure atmospheric allergens. J. Allergy Clin. Immun. 68:194-200. 20   A. Celenza, J. Fothergill, E. Kupek, and R.J. Shaw. 1996. Thunderstorm associated asthma: A detailed analysis of environmental factors. British Medical Journal 312(7031):604-607. G.F. Schappi, C. Suphioglu, P.E. Taylor, and R.B. Knox. 1997. Concentrations of the major birch tree allergen Bet v 1 in pollen and respirable fine particles in the atmosphere. J. Allergy Clin. Immun. 100:656-661. 21   Celenza et al., 1996. See note 20 above. 22   G.F. Schappi, P.E. Taylor, M.C. Pain, P.A. Cameron, A.W. Dent, I.A. Staff, and C. Suphioglu. 1999. Concentrations of major grass group 5 allergens in pollen grains and atmospheric particles: Implications for hay fever and allergic asthma sufferers sensitized to grass pollen allergens. Clin. Exp. Allergy 29:633-641.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases animal activities).23 Highly infectious fungal diseases (e.g., coccidiomycosis, histoplasmosis) are spread directly from soil. The fungal life cycle is controlled in part by climate. Many fungi spend the winter in resting stages and produce sexual spores in the spring. These sexual spores germinate to form new fungal mycelia, which then produce asexual spores later in the season. For many fungi that produce large fruiting bodies (e.g., mushrooms) mycelia expand throughout the growing season and the fruiting bodies are produced and release spores in the fall. Thus, concentrated basidiospore aerosols often occur in the fall. Most fungi rely on the atmosphere for transport and spread, and many have developed remarkable mechanisms to insert spores into the air. Most of the mechanisms depend on water in some way, and spores show marked periodicity based on the water content of the air. Many spores are released early in the morning, when the dew point is reached (e.g., basidiospores).24 Others become abundant later in the day, when drying causes the conidiophores to twist, shaking spores loose (e.g., Cladosporium). Rain events cause splash-dispersed spores to become abundant and induce forcible spore release in the ascomycetes. Ascospore concentrations can reach levels well in excess of 1,000 per liter during light rain. Human activity, such as farming, can also produce major spore plumes. When field crops are harvested after the grain matures, huge numbers of spores are released to the extent that spore clouds become visible. Even in air-conditioned combine cabs, spore concentrations can reach more than 10,000 per liter, and levels within the cloud are more likely to be greater than 107 per liter.25 Another example is composting. Clouds of Aspergillus fumigatus spores (more than 1,000 per liter) and other spores are released from yard waste compost when the compost is disturbed.26 Spore Concentrations Several studies that document the broad range of spore concentrations that can be present in outdoor air are listed in Table 3.4. Note that most of the values shown here are averages. Cultural counts are averages or ranges of multiple grab samples, each collected over 1 to 5 minutes. The spore counts are generally a 24-hour average. Most spore reports drastically underestimate concentrations, primarily because many investigators lack the training and dedication required to count small colorless spores accurately. When properly counted, even ambient levels are frequently in excess of 50 per liter, and hourly averages can be greater than 200 per liter.27 The problem of averages discussed above also applies here. Molina et al.28 report 24-hour average levels over a year of greater than one spore per liter of air. Considering that Cladosporium is strongly diurnal and seasonal, this means that daily averages could be 10 to 100 per liter, and hourly averages could easily exceed 1,000 per liter.29 As for pollen, fungal components have been detected in air in the absence of countable particles. Several investigators have measured the allergen content of particle-free air.30 23   A.A. Kwaasi, R.S. Parhar, F.A. al-Mohanna, H.A. Harfi, K.S. Collison, and S.T. al-Sedairy. 1998. Aeroallergens and viable microbes in sandstorm dust. Potential triggers of allergic and nonallergic respiratory ailments. Allergy 53:255-265. 24   W.G.D. Fernando, J. Miller, L. Seaman, K. Seifert, and T.C. Paulitz. 2000. Daily and seasonal dynamics of airborne spores of Fusarium graminearum and other Fusarium species sampled over wheat plots. Can. J. Bot. 78:497-505. J. Molina Mediavilla, J. Angulo, E. Domínguez, A. Castro, and F. Infante. 1997. Annual and diurnal incidence of Cladosporium conidia in the atmosphere of Córdoba, Spain. J. Investigational Allergology and Clinical Immunology 7(3):179-182. 25   J.H. Chapman, H. Burge, M. Muilenberg. 1996. Fungus allergen exposure during Midwest USA Fall crop harvests. J. Allergy Clin. Immun. 77(1 part 2):200. 26   C.S. Clark, R. Rylander, and L. Larsson. 1983. Levels of gram-negative bacteria, Aspergillus, fumigatus, dust, and endotoxin at compost plants. Appl. Envrion. Microb. 45:1501-1505. 27   M. Burch and E. Levetin. 2002. Effects of meteorological conditions on spore plumes. Int. J. Biometeorol. 46:107-117. O. Carisse and V. Philion. 2002. Meteorological factors affecting periodicity and concentration of airborne spores of Bremia lactucea. Can. J. Plant Pathol. 24:184-193. 28   Molina et al., 1997. See note 24 above. 29   Molina et al., 1997. See note 24 above. 30   M.K. Agarwal, M.C. Swanson, and C.E. Reed. 1983. Immunochemical quantitation of airborne short ragweed, Alternaria, antigen

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases TABLE 3.4 Variability of Spore Concentrations Reported in Various Locations Analysis Range Condition Locale Authors Culture 0-17 cfu/litera Ambient   Gorny, Reponen et al.b Culture 0-7 cfu/liter Ambient Washington, D.C. Jones and Cooksonc Spore count 70-200 spores/liter Ambient Oklahoma Burch and Levetind Culture 100-1,000/liter Harvesting Egypt Hameed and Khodre Spore counts 5-26/liter Yard waste compost Illinois Hryhorczuk et al.f Culture 11 (6-3)/liter, winter 5 (2-12)/liter, summer Ambient Taiwan Pei-Chih et al.g Culture 1 cfu/liter Ambient Taiwan Li, Hsu et al.h Counts >1,000/liter Crop harvest Missouri Chapman, Burge et al.i RNA probes >25 organisms/liter Ambient peaks Sweden Biggins et al.i a Colony-forming unit, cfu. b R.L. Gorny, T. Reponen, K. Willeke, D. Schmechel, E. Robine, M. Boissier, and S.A. Grinshpun. 2002. Fungal fragments as indoor air biocontaminants. Appl. Environ. Microb. 68:3522-3531. c B.L. Jones and J.T. Cookson. 1983. Natural atmospheric microbial conditions in a typical suburban area. Appl. Environ. Microb. 45:919-934. d M. Burch and E. Levetin. 2002. Effects of meteorological conditions on spore plumes. Int. J. Biometeorol. 46:107-117. e A.A. Hameed and M.I. Khodr. 2001. Suspended particulates and bioaerosols emitted from an agricultureal nonpoint source. J. Environ. Monit. 3:206-209. f D. Hryhorczuk, L. Curtis, P. Scheff, J. Chung, M. Rizzo, C. Lewis, N. Keys, and M. Moomey. 2001. Bioaerosol emissions from a suburban yard waste composting facility. Ann. Agr. Environ. Med. 8:177-185. g W. Pei-Chih, S. Huey-Jen, and L. Chia-Yin. 2000. Characteristics of indoor and outdoor airborne fungi at suburban and urban homes in two seasons. Sci. Total Envir. 253:111-118. h C.S. Li, C.W. Hsu, and M.L. Tai. 1997. Indoor pollution and sick building syndrome symptoms among workers in day-care centers. Arch. Envir. Heal. 52:200-207. i J.H. Chapman, H. Burge, M. Muilenberg. 1996. Fungus allergen exposure during Midwest USA Fall crop harvests. J. Allergy Clin. Immun. 77(1 part 2):200. j P. Biggins, N. Pomeroy, M. Pearce, C. Stone, N. Brown, R.M Harrison, J. Hobman, and A. Jones. 2002. Characterisation of the ambient respirable biological aerosol in Proceedings of the Sixth Annual UK Review Meeting on Outdoor and Indoor Air Pollution Research. Available at http://www.le.ac.uk/ieh/pdf/w12.pdf. Accessed November 2003. pp. 75-77. Outdoor Bacteria Natural outdoor bacterial aerosols are derived from plants, soil, and water. Bacteria are common on or in all of these sources. Every leaf is colonized with a population of bacteria that probably is essential to the healthy life of the plant. Most bacteria are probably released from leaf surfaces, and the primary mechanisms for release are probably droplet splash and wind. Although as yet unmeasured, bacterial clouds released during rainfall are likely to be equivalent to the clouds produced by the fungi. Droplets falling into water create bubbles that scavenge bacteria from the liquid and introduce the cells into the air when the bubbles burst.31     E, and Alt-I allergens: A two-year prospective study. J. Allergy Clin. Immun. 72:40-45. C. Barnes, K. Schreiber, F. Pacheco, J. Landuyt, F. Hu, and J. Portnoy. 2000. Comparison of outdoor allergenic particles and allergen levels. Ann. Allerg. Asthma Im. 84:47-54. 31   R. Marks, K. Jankowska, M. Michalska, and M. Krolska. 1996. The sea to air bacteria transfer from the coastal waters. Bull. Inst. Marit. Trop. Med. Gdnyia. 47(1-4):93-103. C. Gomez-Suarez, H.J. Busscher, and H.C. van der Mei. 2001. Analysis of bacterial detachment from substratum surfaces by the passage of air-liquid interfaces. Appl. Environ. Microb. 68:3522-3531.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases Disturbance of compost, moldy hay, or other damp organic material will release large numbers of thermophilic bacteria (primarily actinomycetes and Bacillus species). Soil is another important source for airborne bacteria and has been considered a source for Q fever in French Guiana.32 Bacterial Concentrations Bacterial aerosols vary vertically, geographically, and over time. Meteorological factors such as changing wind direction may play a major role in the characteristics of bacterial populations.33 Few data are available on bacterial populations in outdoor air, and virtually all that is published is derived from cultural sampling and likely to be a gross underestimate of actual concentrations. Some of the data that are available are presented in Table 3.5. Biggins et al.34 reported two fungal peaks and three bacterial peaks in excess of 25,000 per cubic meter using counts derived from ribonucleic acid (RNA) probe data. Of the bacteria, 28 percent were unidentified, providing an example of how little is actually known about the outdoor bacterial aerosol. In studies evaluating airborne culturable bacteria in various localities in Sweden, bacterial levels above city streets were highest.35 However, when event-associated aerosol releases occur, rural areas are likely to experience the greatest and fastest releases due to the potential for disturbance of agricultural materials. Because many of the biological agents of concern are bacteria, it is important to know specifically what kinds of bacteria may be present in air, and whether or not there are natural populations of any of the biological threat agents. Bacillus species are especially common in the natural environment. Natural populations of Bacillus thuringensis were recovered from soil in Spain.36 In addition, B. thuringensis is used as an insecticide and is sprayed into the air and allowed to settle on affected plants. At least transiently, concentrations of this organism may be very high.37 Dust generated by combines (harvesters) is also a rich source for Bacillus aerosols, including B. brevis, B. cereus, B. circulans, B. coagulans, B. licheniformis, B. stearothermophilus, and B. subtilis.38 Bacterial cell wall fragments are probably abundant in outdoor air. These have been measured only as either endotoxin (gram-negative) or peptidoglycan (all bacteria), and measurements have not been correlated with particle counts. Other Outdoor Bioaerosols Actual concentrations of viral particles in outdoor air are unknown. However, they have been recovered from plumes above sewage treatment facilities.39 Clearly, transmission of viral disease can occur via transport through outdoor air. Foot and mouth and Newcastle viruses are animal disease agents for which epidemiological data have confirmed outdoor airborne spread. Also, enteric viruses 32   J. Gardon, J.M. Heraud, S. Laventure, A. Ladam, P. Capot, E. Foquet, J. Favre, S. Webber, D. Hommel, A. Hulin, Y. Couratte, and A. Talermin. 2001. Suburban transmission of Q fever in French Guiana: Evidence of a wild reservoir. J. Infect. Dis. 184:278-284. 33   B. Lightheart and A. Kirilenko. 1998. Simulation of summer-time diurnal bacterial dynamics in the atmospheric surface layer. Atmos. Environ. 32(14-15):2491-2496. 34   P. Biggins, N. Pomeroy, M. Pearce, C. Stone, N. Brown, R.M Harrison, J. Hobman, and A. Jones. 2002. Characterisation of the ambient respirable biological aerosol in Proceedings of the Sixth Annual UK Review Meeting on Outdoor and Indoor Air Pollution Research. Available at http://www.le.ac.uk/ieh/pdf/w12.pdf. Accessed August 2003, pp. 75-77. 35   A. Bovallius, B. Bucht, R. Roffey, and P. Anas. 1978. Three year investigation of the natural airborne bacterial flora at four localities in Sweden. Appl. Environ. Microb. 35:847-852. 36   J. Iriarte, Y. Bel, M.D. Ferrandis, R. Andrew, J. Murillo, J. Ferre, and P. Caballero. 1998. Environmental distribution and diversity of Bacillus thuringiensis in Spain. Syst. Appl. Microbiol. 21:97-106. 37   K. Teschke, Y. Chow, K. Bartlett, A. Ross, and C. van Netten. 2001. Spatial and temporal distribution of airborne Bacillus thuringiensis var. kurstaki during an aerial spray program for gypsy moth eradication. Environ. Heal. Persp. 109:47-54. 38   A.A. Shoreit and M.A. Ismail. 1992. Bacillus species associated with wheat and sorghum dusts from combine harvester. Zentrabl. Mikrobiol. 147:541-550. 39   E.R. Baylor, M.B. Baylor, D.C. Blanchard, L.D. Syzdek, and C. Appel. 1977. Virus transfer from surf to wind. Science 198(4317):575-580. A. Carducci, C. Gemelli, L. Cantiani, B. Casini, and E. Rovini. 1999. Assessment of microbial parameters as indicators of viral contamination of aerosol from urban sewage treatment plants. Lett. Appl. Microbiol. 28:207-210.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases TABLE 3.5 Reported Concentrations of Total Bacteria in Outdoor Air Assay Method Concentration Sample Environment Location Authors Culture 0.1-0.6 cfu/liter (24-hr average)a Urban air Oregon Shaffer and Lighthartb Culture 0-4 cfu/liter (3-year averages) Rural and urban air Sweden Bovallius, Bucht et al. c Culture 1.4-2.2 cfu/liter Above waste treatment plant   Brandi, Sisti et al. d Culture 0.09-4.7/liter   Eastern Europe Gorny, Reponen et al. e Culture 10-1,000/liter Wheat harvest Egypt Hameed and Khodr f Culture 0.5-78.9 cfu/liter Yard waste compost site Illinois Hryhorczuk, Curtis et al. g Culture 0.004-1.5 cfu/liter Ambient Washington, D.C. Jones and Cookson h Culture 10 cfu/liter Sewage sludge application Texas Pillai, Widmer et al. i Culture 0.4 cfu/liter Ambient Taiwan Li, Hsu et al. j RNA probes >25 organisms/liter Ambient peaks England Biggins, Pomeroy et al.k a Colony-forming unit, cfu. b B.T. Shaffer and B. Lighthart. 1997. Survey of culturable airborne bacteria at four diverse locations in Oregon: urban, rural, forest, and coastal. Microb. Ecol. 34:167-177. c A. Bovallius, B. Bucht, R. Roffey, and P. Anas. 1978. Three year investigation of the natural airborne bacterial flora at four localities in Sweden. Appl. Environ. Microb. 35:847-852. d G. Brandi, M. Sisti, and G. Amagliani. 2001. Evaluation of the environmental impact of microbial aerosols generated by wastewater treatment plants utilizing different aeration systems. J. Appl. Microb. 88:845-852. e R.L. Gorny, T. Reponen, K. Willeke, D. Schmechel, E. Robine, M. Boissier, and S.A. Grinshpun. 2002. Fungal fragments as indoor air biocontaminants. Appl. Environ. Microb. 68:3522-3531. f A.A. Hameed and M.I. Khodr. 2001. Suspended particulates and bioaerosols emitted from an agricultureal nonpoint source. J. Environ. Monit. 3:206-209. g D. Hryhorczuk, L. Curtis, P. Scheff, J. Chung, M. Rizzo, C. Lewis, N. Keys, and M. Moomey. 2001. Bioaerosol emissions from a suburban yard waste composting facility. Ann. Agr. Environ. Med. 8:177-185. h B.L. Jones and J.T. Cookson. 1983. Natural atmospheric microbial conditions in a typical suburban area. Appl. Environ. Microb. 45:919-934. i S.D. Pillai, K.W. Widmer, S.E. Dowd, and S.C. Ricke. 1996. Occurrence of airborne bacteria and pathogen indicators during land application of sewage sludge. Appl. Environ. Microb. 61:296-299. j C.S. Li, C.W. Hsu, and M.L. Tai. 1997. Indoor pollution and sick building syndrome symptoms among workers in day-care centers. Arch. Envir. Heal. 52:200-207. k P. Biggins, N. Pomeroy, M. Pearce, C. Stone, N. Brown, R.M. Harrison, J. Hobman, and A. Jones. 2002. Characterisation of the ambient respirable biological aerosol. In Proceedings of the Sixth Annual UK Review Meeting on Outdoor and Indoor Air Pollution Research. Available at http://www.le.ac.uk/ieh/pdf/w12.pdf. Accessed November 2003. pp. 75-77. have been recovered from the surf zone near where the Hudson River discharges into the Atlantic, and nearby residents have a higher than expected rate of disease related to these organisms.40 Indoor Aerosols Common sources for indoor bioaerosols are listed in Table 3.6. In naturally ventilated interiors, the outdoor aerosols strongly affect indoor air, especially during seasons with open windows (see below). Understanding variations in the outdoor aerosol for all relevant particles is essential to predicting 40   Baylor et al., 1977. See note 39 above.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases bacteria and fungi per minute into the air.41 Ultrasonic units may kill these organisms, but fragments are still aerosolized. Vacuuming with ordinary brush and beater vacuums equipped with ordinary bags can raise clouds of dust with entrained organic material and microorganisms.42 However, some studies have reported no association between vacuuming and airborne dust levels.43 Indoor Pollen and Pollen-Derived Particles One study has evaluated particle penetration into moving vehicles using ragweed pollen as an indicator.44 Another evaluated the effects of room air conditioners on indoor pollen concentrations.45 In both of these, there was a direct correlation between indoor and outdoor counts, with the indoor/outdoor ratio being related to whether or not windows were open. Similarly, allergen measurements in floor dust reveal a close correlation between specific pollen allergen concentrations and outdoor pollen counts. Probably most pollen enters the air directly from outdoors. However, it is also likely that pollen grains are captured by people and by animals and brought inside. These would be re-released during activities of those carrying them. Because pollen grains are large and fall rapidly, these aerosols are transient. Pollen Concentrations Except for studies of penetration and accumulation in dust from outdoors, indoor pollen concentrations have not been systematically studied. In fact, indoor sources of pollen are uncommon. Most plants grown indoors (except in specialized environments such as greenhouses) do not produce airborne pollen. Measured pollen levels in occupied spaces have almost always been reported as less than those outside, with the difference depending on pathways for penetration.46 No study has reported indoor levels that exceed those outdoors, and indoor levels are generally low (less than 1 grain per liter). Floor dust analysis is the method most commonly used to measure indoor allergens, including pollen. In Sweden, outdoor birch pollen peaks in May range from 80 to 140 grains per cubic meter, and indoor allergen concentration closely parallels outdoor pollen counts.47 Fahlbusch et al.48 measured 120 to 150 nanograms grass pollen allergen per square meter of carpet. Maximum levels ranged from 4,000 to 6,000 nanograms per square meter. Allergen concentrations were highest during pollen season and correlated reasonably well with pollen counts. Neither concentrations of airborne allergen nor concentrations of allergen-bearing particles have been reported for pollen. Indoor Fungi Indoor/outdoor relationships for fungal spores are complex and depend on the type of spore under consideration.49 Some fungi are rarely found growing indoors (i.e., most basidiomycetes), while others 41   W.R. Solomon. 1976. A volumetric study of winter fungus prevalence in the air of Midwestern homes. J. Allergy Clin. Immun. 57:46-55. R.L. Haddock and F.A. Nocon. 1994. Infant salmonellosis and vacuum cleaners. J. Trop. Pediatrics 40:53-54. 42   Haddock and Nocon, 1994. See note 41 above. 43   L. Lehtonen and P. Huovinen. 1993. Susceptibility of respiratory tract pathogens in Finland to cefixime and nine other antimicrobal agents. Scand. J. Infect. Dis. 25:373-378. 44   M.L. Muilenberg, W.S. Skellenger, H.A. Burge, and W.R. Solomon. 1991. Particle penetration into the automotive interior I. Influence of vehicle speed and ventilatory mode. J. Allergy Clin. Immun. 87:581-585. 45   W.R. Solomon, H.A. Burge, and J.R. Boise. 1980. Exclusion of particulate allergens by window air conditioners. J. Allergy Clin. Immun. 64:305-308. 46   D.A. Sterling and R.D. Lewis. 1998. Pollen and fungal spores indoor and outdoor of mobile homes. Ann. Allergy Asthma Immun. 80:279-285. 47   L. Holmquist, J. Weiner, and O. Vesterberg. 2001. Airborne birch and grass pollen allergens in street-level shops. Indoor Air 11:241-245. 48   B. Fahlbusch, D. Horning, J. Heinrich, and L. Jager. 2001. Predictors of group 5 grass-pollen allergens in settled house dust: Comparison between pollination and nonpollination seasons. Allergy 56:1081-1086. 49   D. Li and B. Kendrick. 1996. Functional and causal relationships between indoor and outdoor airborne fungi. Can. J. Bot. 74:194-

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases may become abundant indoors (e.g., Aspergillus and Penicillium species). However, most indoor fungi are derived directly from outdoor air and hence are present in lower concentrations indoors than outdoors.50 Penetration depends, as for pollen, on existing pathways into the building. Central ventilation system filtration plays some role in determining the size of the pathways although even with low-efficiency filtration, in large buildings most fungal spores do not travel all the way through the ventilation system to reach the occupied space. Indoor/outdoor ratios are often used as an indicator for whether or not an indoor space is supporting fungal growth. However, this ratio is strongly dependent on the concentration of spores outdoors, with low outdoor spore counts often leading to ratios greater than 1 in the absence of active growth.51 Activities strongly influence concentrations of fungal spores in indoor air. Any activity that disturbs a source is likely to increase levels. Some activities considered important are cleaning, bed making, etc.52 Scheff et al.53 calculated the rate of shedding of fungi from people in a building based on number of occupants, activity, and the number of fungi recovered from air during activity. This group estimated the release of 167 cfu per hour per person for total fungi. Fungi also actively release spores, primarily during changes in water activity. Thus, large peaks of spores may be released if the temperature in a room drops, causing the relative humidity to rise (at least locally) and condensation to occur. Indoor spore concentrations represent the sum of outdoor penetration and release from indoor reservoirs such as dust or active growth in one or more reservoirs. The largest spore clouds result from disturbance of active growth or, in some cases, from forcible spore discharge in such growth. Table 3.7 lists a few studies that report indoor spore levels. The concentrations presented in Table 3.7 are biased by the analytical method used, and all underestimate true particle concentrations, as discussed above. Some evidence exists for small particle aerosols of fungal allergens in the 1 to 5 μm range.54 However, this is an area that needs additional work. Indoor Bacteria Source/Release Factors As they did for fungi, Scheff et al.55 have estimated the number of bacteria released by people. They estimate 227 cfu/hr/person-min. Unlike fungi, which are shed primarily from clothing, most bacteria are commensal organisms and are shed from skin surfaces and expired with respiratory secretions. Skin surface organisms are released with physical activity and are abundant in schools for this reason. Respiratory organisms are released with sneezes, coughs, and, to a lesser extent, with singing and speaking. Bacteria from humans may become airborne either as droplet nuclei or on skin or fiber rafts. Droplet nuclei can be nearly as small as an individual bacterium and may stay airborne for many minutes. Rafts are usually much larger and probably fall within only a few minutes. These factors have not been studied sufficiently to allow predictions of the extent to which these aerosols might interfere with agent detection.     209. 50   H.A. Burge, D.L. Pierson, T.O. Groves, K.F. Strawn, and S.K. Mishra. 2000. Dynamics of airborne fungal populations in a large office building. Curr. Microbiol. 40:10-16. 51   Burge et al., 2000. See note 50 above. 52   M. Lehtonen, T. Reponen, and A. Nevalainen. 2003. Everyday activities and variation of fungal spore concentrations in indoor air. Int. Biodeter. Biodegr. 31:25-39. 53   P.A. Scheff, V.K. Paulius, L. Curtis, and L.M. Conroy. 2000. Indoor air quality in a middle school, Part II: Development of emission factors for particulate matter and bioaerosols. Appl. Occup. Environ. Hyg. 15:835-842. 54   M.Y. Menetrez, K.K Foarde, and D.S. Ensor. 2001. An analytical method for the measurement of nonviable bioaerosols. J. Air Waste Manage. 51:1436-1442. 55   Scheff et al., 2000. See note 53 above.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases TABLE 3.7 Sample Concentrations of Fungal Spores in Indoor Environments Analytic Method Concentration Condition Locale Authors Culture 1-2 cfu/litera University auditorium Italy Sessa, Di Pietro et al.b Culture 0.1-0.3 cfu/liter Apartment Italy Sessa, Di Pietro et al.b Direct microscopy 5-11 org./literc Large buildings U.S. Midwest Reynolds, Black et al.d Culture 9 (4-18) cfu/liter winter Residences Taiwan Pei-Chih, Huey-Jen et al.e   4 (2-12) cfu/liter summer   Culture 0.9 (0.5-3) cfu/liter Apartments France Duchaine and Meriauxf Culture 0 to 41 cfu/liter Residences Scotland Strachan, Flannigan et al.g Culture 0.01 to >20 cfu/liter Residences U.S. Northeast Solomonh Culture 0.846-1.033 cfu/liter Residences U.S. Northeast Ren, Jankun et al.i Culture 0.2-0.8 cfu/liter Homes, offices Poland Lis, Pastuszka et al.j Culture 1 (2-12) cfu/liter Day care centers Taiwan Li and Hsuk Culture 1-200 cfu/liter During office remediation Finland Rautiala, Reponen et al.l Spore count 100-1,000 spores/liter During office remediation Finland Rautiala, Reponen et al.l Culture 0-17 cfu/liter Residences Poland Gorny, Reponen et al.m Culture 5 cfu/litern Residences Poland Gorny, Reponen et al.m a Colony-forming units, cfu. b R. Sessa, M. Di Pietro, G. Schiavoni, I. Santino, A. Altieri, S. Pinelli, and M. Del Piano. 2002. Microbiological indoor air quality in healthy buildings. New Microbiol. 25:51-56. c Combination of directly counted individual bacterial and fungal cells. d S.J. Reynolds, D.W. Black, S.S. Borin, G. Breuer, L.F. Burmeister, L.F. Guortes, T.F. Smith, M.A. Stein, P. Subramanian, P.S. Thorns, and P. Whitten. 2000. Indoor environmental quality in six commercial office buildings in the Midwest United States. Appl. Occup. Environ. Hyg. 16:1065-1077. e W. Pei-Chih, S. Huey-Jen, and L. Chia-Yin. 2000. Characteristics of indoor and outdoor airborne fungi at suburban and urban homes in two seasons. Sci. Total Envir. 253:111-118. f C. Duchaine and A. Merieaux. 2000. Airborne microfungi from eastern Canada sawmills. Can. J. Microb. 46:612-617. g D.P. Strachen, B. Flannigan, E.M. McCabe, and F. McGarry. 1990. Quantification of airborne moulds in the homes of children with and without wheeze. Thorax 45:382-387. h W.R. Solomon. 1976. A volumetric study of winter fungus prevalence in the air of Midwestern homes. J. Allergy Clin. Immun. 57:46-55. i P. Ren, T.M. Jankun, K. Belanger, M.B. Bracken, and B.P. Leaderer. 2001. The relation between fungal propagules in indoor air and home characteristics. Allergy 56:419-424. j D.O. Lis, J.S. Pastuszka, and R.L. Górny. 1997. [The prevalence of bacterial and fungal aerosol in homes, offices and ambient air of Upper Silesia. Preliminary results.] Rocz. Panstw. Zakl. Hig. 48:59-68. k C.S. Li, C.W. Hsu, and M.L. Tai. 1997. Indoor pollution and sick building syndrome symptoms among workers in day-care centers. Archive of Environmental Health 52:200-207. l S. Rautiala, T. Reponen, A. Hyvärinen, A. Nevalainen, T. Husman, A. Vehviläinen, and P. Kalliokoski. 1996. Exposure to airborne microbes during the repair of moldy buildings. Am. Ind. Hyg. Assoc. J. 57:279-284. m R.L. Gorny, T. Reponen, K. Willeke, D. Schmechel, E. Robine, M. Boissier, and S.A. Grinshpun. 2002. Fungal fragments as indoor air biocontaminants. Appl. Environ. Microb. 68:3522-3531. n Exposure limit.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases Indoor Bacterial Concentrations Reported concentrations of bacteria in indoor air are listed in Table 3.8. Again, results are biased due to the method of analysis, and total concentrations far exceed those represented by culture (see discussion above). Bacterial products may be present in aerosols not associated with intact bacterial cells. Endotoxin is a prime example. Although endotoxin clearly is consistently present in both outdoor and indoor air and (especially indoors) can reach concentrations that impact human health, studies documenting the actual particles on which the endotoxin is borne have not been conducted. Most is probably present on intact gram-negative bacteria. Office building levels are usually relatively low (between 0.05 and 3 nanograms per liter), with naturally ventilated building levels lower than levels in mechanically ventilated ones.56,57 On the other hand, in areas where agricultural and animal confinement activities are occurring, levels are often much higher (e.g., 490 nanograms per cubic meter for swine confinement).58 TABLE 3.8 Reported Concentrations of Bacteria in Indoor Air in Various Circumstances Analysis Method Concentration Type of Building Location Authors Culture 0.9-1.2 cfu/liter Auditorium with people Rome Sessa, Di Pietro, et al.a Culture 0.735 cfu/liter Day care centers Taiwan Li, Hsu et al.b Culture 8-11org/literc Office buildings U.S. Midwest Reynolds, Black et al.d Culture 425 (167-930) cfu/liter Swine confinement Canada Duchaine, Grimard et al.e Culture 57-260 cfu/m3 Homes Poland Lis, Pastuszka et al.f Culture 19-118 cfu/m3 Offices Poland Lis, Pastuszka et al.f Culture 0-0.196 cfu/liter Offices Mauritius Bholah and Subrattyg Culture 0.088-16 cfu/liter Residences Poland Gorny, Reponen, et al.h Culture 5 cfu/literi Residences Poland Gorny, Reponen et al.h a R. Sessa, M. Di Pietro, G. Schiavoni, I. Santino, A. Altieri, S. Pinelli, and M. Del Piano. 2002. Microbiological indoor air quality in healthy buildings. New Microbiol. 25:51-56. b C.S. Li, C.W. Hsu, and M.L. Tai. 1997. Indoor pollution and sick building syndrome symptoms among workers in day-care centers. Archive of Environmental Health 52:200-207. c Combination of directly counted individual bacterial and fungal cells. d S.J. Reynolds, D.W. Black, S.S. Borin, G. Breuer, L.F. Burmeister, L.F. Guortes, T.F. Smith, M.A. Stein, P. Subramanian, P.S. Thorns, and P. Whitten. 2000. Indoor environmental quality in six commercial office buildings in the Midwest United States. Appl. Occup. Environ. Hyg. 16:1065-1077. e C. Duchaine, Y. Grimard, and Y. Cormier. 2000. Influence of building maintenance, environmental factors, and seasons on airborne contaminants of swine confinement buildings. Am. Ind. Hygiene Assn J. 61:56-63. f D.O. Lis, J.S. Pastuszka, and R.L. Górny. 1997. [The prevalence of bacterial and fungal aerosol in homes, offices and ambient air of Upper Silesia. Preliminary results.] Rocz. Panstw. Zakl. Hig. 48:59-68. g R. Bholah and A.H. Subratty. 2002. Indoor biological contaminants and symptoms of sick building syndrome in office buildings in Mauritius. Int. J. Environ. Heal. Res. 12:93-98. h R.L. Gorny, T. Reponen, K. Willeke, D. Schmechel, E. Robine, M. Boissier, and S.A. Grinshpun. 2002. Fungal fragments as indoor air biocontaminants. Appl. Environ. Microb. 68:3522-3531. i exposure limit 56   K.B. Teeuw, C.M. Vanderbroucke-Grauls, and J. Verhoef. 1994. Airborne gram-negative bacteria and endotoxin in sick building syndrome: A study in Dutch governmental office buildings. Arch. Intern. Med. 154:2339-2345. 57   S.J. Reynolds, D.W. Black, S.S. Borin, G. Breuer, L.F. Burmeister, L.F. Guortes, T.F. Smith, M.A. Stein, P. Subramanian, P.S. Thorns, and P. Whitten. 2000. Indoor environmental quality in six commercial office buildings in the Midwest United States. Appl. Occup. Environ. Hyg. 16:1065-1077. 58   C. Duchaine, Y. Grimard, and Y. Cormier. 2000. Influence of building maintenance, environmental factors, and seasons on airborne contaminants of swine confinement buildings. American Industrial Hygiene Association Journal 61:56-63.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases Other Bioaerosols Many other bioaerosols occur in indoor environments, although most are probably present in concentrations too low to be a problem with respect to false alarms. However, cat and dog allergens (proteins associated with particles ranging from smaller than 1 μm to larger than 20 μm59) could be present in sufficient concentration in homes with these animals to trigger detectors that rely on protein fluorescence.60 The same is likely to be true for other animals, especially rodents kept in animal care facilities associated with research buildings or hospitals. With respect to bioagent detection, natural indoor aerosols may trigger false alarms unless the detectors and their analytic methods have a means for controlling this source of error. Because, in general, the types of aerosols that are natural in indoor air are different from those likely to be associated with biological attacks, some level of particle identification may solve this problem. The specificity of identification necessary depends on the specific types of aerosols actually found to be common in indoor air and the relationships between these organisms and the bioagents of concern. Determining these parameters will require measurement of indoor aerosols using new methods of detection, and then empirically determining the level of specificity needed to prevent excessive false alarms. Predicting the Prevalence of Bioaerosols Understanding the prevalence of biological aerosols requires consideration of all the factors in Table 3.9. It is impossible to map all aerosol concentrations, and predictions are essential to allow extrapolation from relatively limited collected data. If one can document the effects of each of these variables, predictive models could be developed so that extensive sampling becomes less necessary (other than to test the reliability of the models). A number of groups have published models intended to predict pollen and spore concentrations. These models fall into four general categories: Prediction of long-term trends—for example, trends associated with global warming; Prediction of the severity of coming seasons; Prediction of the start date for future seasons; and Prediction of concentrations for the following day. Predictive models are currently being used to study the effects of global warming. In general, pollen and spore concentrations are expected to gradually rise in response to warmer temperatures and higher levels of CO2.61 Both pollen and fungal spores are produced in greater abundance as CO2 levels increase in their environment.62 Emberlin et al.63 evaluated long-term changes in pollen concentrations as possible causes for the increasing prevalence of hay fever symptoms. Land use changes, cumulative temperature, and rainfall were used as predictors for the severity of coming seasons in a single-equation multivariate model that resulted in greater than 95 percent predictive value. Galan et al.64 also used weather variables in a 59   L. Holmquist and O. Vesterberg. 2002. Direct on air sampling filter quantification of cat allergen. J. Biochem. Bioph. Meth. 51:17-25. H. Ormstad and M. Lovik. 2002. Air pollution, asthma and allergy—the importance of different types of particles. Tidsskr. Nor. Laegeforen 122:1777-1782. 60   H. Ormstad. 2000. Suspended particulate matter in indoor air adjuvants and allergen carriers. Toxicology 152:52-68. 61   C.P. Osborne, I. Chuine, D. Viner, and F.I. Woodward. 2000. Olive phenology as a sensitive indicator of future climatic warming in the Mediterranean. Plant, Cell and Environment 23:701-710. 62   J.N. Klironomos, M.C. Rillig, M.F. Allen, D.R. Zak, M. Kubiske, and K.S. Pregitzer. 1997. Increased levels of airborne fungal spores in response to populus tremuloides grown under elevated atmospheric CO2. Can. J. Bot. 75:1670-1673. 63   J. Emberlin, J. Mullins, J. Corden, S. Jones, W. Millington, M. Brooke, and M. Savage. 1999. Regional variations in grass pollen seasons in the UK: Long-term trends and forecast models. Clin. Exp. Allergy 29(3):347-356. 64   C. Galan, P. Carinanos, H. Garcia-Mazo, P. Alcazar, and E. Dominguez-Vilches. 2001. Model for forecasting Olea europaea L. airborne pollen in south-west Andalusia Spain. Int. J. Biometeorol. 45:59-63.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases TABLE 3.9 Important Factors in Bioaerosol Predictive Models, with Examples Source Characteristics Release Mechanisms Dispersion Decay Physical nature of source (single living organism; pond filled with organisms) Components of populations (single organism, multiple organisms, fragments) Concentrations of populations (concentrations of total or individual particles) Patterns of variability over time Inherent to the organism (forcible spore discharge in fungi) Inherent to the source (dispersal bombs, spray humidifiers) External factors (e.g., air movement, water splash and bubbles; abrasion) Air movement rates and patterns ( wind speed and direction, including pattern of changes, turbulence) Space characteristics (shape and size, connections to other spaces) Particle size factors (settling, impaction, diffusion) Impaction surfaces Dilution (amount of clean air available for dilution) Chemical and biological changes (death of infectious particles; denaturing of allergens) multivariate model to predict total output of pollen (i.e., severity of the season) with good success. However, predicting maxima has been less reliable in this instance and others.65 Threshold temperatures and mean heat accumulation have been used to predict the start of the pollen season.66,67,68 Predictions of next-day counts also have used meteorological parameters, but it is necessary to add variables to account for seasonality and other effects on flowering and pollen release.69 Because of the seasonality of pollen production, flowering time predictions are an important aspect of day-to-day pollen predictive modeling. Local conditions that affect pollen distribution (as well as other bioaerosols) also need to be considered. For example, sea–land breezes play a role in regionally distributing pollen (and probably other local bioaerosols).70 Another important set of pollen prediction parameters is endogenous to the plant. Consideration of possible pathogen effects is also important.71 One group has used neural networks, chaos theory, and fractals to develop pollen predictive models.72 Spore models have also been constructed, especially to predict prevalence of plant pathogens, which reach local concentrations in excess of 100 per liter.73 Factors intrinsic to the fungi (i.e., life cycle 65   S. Kawashima and Y. Takahashi.1999. An improved simulation of mesoscale dispersion of airborne cedar pollen using a flowering-time map. Grana 38(5):316-324. 66   H. Garcia-Mozo, C. Galan, M.J. Aira, J. Belemonte, C. Diaz de la Guardia, D. Fernandez, A.M. Gutierrez, F.J. Rodriguez, M.M. Trigo, and E. Dominguez-Vilches. 2002. Modelling start of oak pollen season in different climactic zones in Spain. Agr. Forest Meteorol. 110:247-257. 67   J.M. Corden, A. Stach, and W.M. Millington. 2002. A comparison of Betula pollen seasons at two European Sites: Derby, United Kingdom, and Poznan, Poland (1995-1999). Aerobiologia 18:45–53. 68   H. García-Mozo, C. Galán, M.T. Gómez-Casero, and E. Dominguez. 2000. A comparative study of different temperature accumulation methods for predicting the Quercus pollen season start in Córdoba (Southwest Spain). Grana 39:194-199. 69   P.C. Stark, L.M. Ryan, J.L. McDonald, and H.A. Burge. 1997. Using meteorologic data to predict daily ragweed pollen levels. Aerobiologia 13:177-184. 70   J. Gassman, C.F. Perez, and J.M. Gardion. 2002. Sea-land breeze in a coastal city and its effect on pollent transport. Int. J. Biometeorol. 46:118-125. P.V.d. Water and E. Levetin. 2001. Contribution of upwind pollen sources to the characterization of Juniperus ashei phenology. Grana 40:133-141. 71   M. Forenaciai, L. Pieroni, F. Orlandi, and B. Romano. 2002. A new approach to consider the pollen variable in forecasting yield models. Econ. Bot. 56(1):66-72. 72   M.E. Degaudenzi and C.M. Arizmendi. 1999. Wavelet-based fractal analysis of airborne pollen. Phys. Rev. E 59(6):6569-6573 C.M. Arizmendi, J.R. Sanchez, N.E. Ramos, and C.I. Ramos. 1993. Time series predictions with neural nets: Application to airborne pollen forecasting. Int. J. Biometeorol. 37:139-144. 73   O. Carisse and V. Philion. 2002. Meteorological factors affecting periodicity and concentration of airborne spores of Bremia lactucea. Can. J. Plant. Pathol. 24(2):184-193. J. Angulo-Romero, A. Mediavilla-Molina, and E. Dominguez-Vilches. 1999. Conidia of Alternia in the atmosphere of the city of

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases seasonality, spore release mechanisms) are especially important in fungal aerosol models, as are meteorological parameters. Unfortunately, the kind of model that would be useful for developing background information for sensing bioagents has not been developed. Needless to say, any useful model would have to be specific as to particle type and geography, and would have to have a resolution similar to that of the sensors. The question arises as to whether or not modeling will ever replace sampling for estimation of the chances of false alarms by specific bioaerosol detectors. The answer, of course, depends on the acceptable alarm rate and on the accuracy of the model predictions. Until models have been attempted that directly address this question, we will not know. Certainly, short of empirically determining how reliable sensors are in many background situations, exploration of models seems a cost-effective approach. CONTROL OF BIOAEROSOLS Air Cleaning Air cleaning is the primary approach used in large buildings and is also marketed as a pollution control device for small buildings, residences, and schools. Filtration is by far the most common approach to air cleaning. While centrally installed air cleaning systems have proven efficacious in reducing indoor aerosols, room-sized units have not. Room-sized units tend to be either noisy (so that occupants turn them off) or relatively inefficient.74 Filtration Properly used, filtration can reduce some microbial aerosols to virtually unmeasurable concentrations. Among the tested aerosols are fungal spores,75 tuberculosis bacilli,76 and a mouse virus.77 The last study mentioned used two medium-efficiency filters one after the other and demonstrated essentially a zero disease transmission rate in equipped units. Filtration efficiency does not necessarily depend on the efficiency of filters.78 Filtration efficiency is controlled by the pore size of the filter material, whether it is wet or dry, and how securely the filter is installed (i.e., what fraction of the total airstream actually passes through the filter).79 Many studies using actual ventilation systems have achieved less than 100 percent reduction even though the filters themselves were capable of 100 percent capture of bioaerosols, indicating that leakage can be a significant problem.80 While theoretically filtration could provide absolute protection, practically speaking, gaps and other openings in mechanical systems allow some penetration of aerosols. This effect has not been carefully studied in ordinary buildings. While standard new HVAC filters can remove considerable percentages of bacteria and fungi from outdoor air, wet, humid conditions may lead to bacterial growth and subsequent release into the building air.81 This effect has also been documented in a laboratory setting.82     Cordoba, Spain in relation to meteorological parameters. Int. J. Biometeorol. 43:45-49. 74   Holmquist et al., 2001. See note 47 above. 75   R.L. Jacobs and C.P. Andrews. 1989. Hypersensitivity pneumonitis treated with an electrostatic dust filter. Ann. Intern. Med. 110:115-118. 76   R.L. Marier and T. Nelson. 1993. A ventilation-filtration unit for respiratory isolation. Infect. Cont. Hosp. Ep. 14:700-705. 77   M. Mrozek, U. Zillmann, W. Nicklas, V. Kraft, B. Meyer, E. Sickel, B. Lehr, and A. Wetzel. 1994. Efficiency of air filter sets for the prevention of airborne infections in laboratory animal houses. Lab Animal 28:347-354. 78   S.C. Miller-Leiden, C. Lobascio, W.W. Nazaroff, and J.M. Macher. 1996. Effectiveness of in-room air filtration and dilution ventilation for tuberculosis infection control. J. Air Waste Manage. 46:869-882. 79   F.S. Rhame. 1991. Prevention of nonscomial aspergillosis. J. Hosp. Infect. 18(Suppl. A):466-472. 80   C. Cundith, C. Kerth, W.R. Jones, T.A. McCaskey, and D.L. Kuhlers. 2002. Microbial reduction efficiencies of filtration, electrostatic polarization, and UV components of a germicidal air cleaning system. J. Food Sci. 67:2278-2281. 81   M. Moritz, H. Peters, B. Nipko, and H. Ruden. 2001. Capability of air filters to retain airborne bacteria and molds in heating, ventilating, and air conditioning (HVAC) systems. Int. J. Envir. Heal. 203:401-409.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases Electrostatic Precipitation Electrostatic precipitation has long been considered one of the best approaches to air cleaning, especially for residential environments. One case report documents a significant reduction in aerosols of Aspergillus spores (2-4 μm) and improvement in symptoms of hypersensitivity pneumonitis using electrostatic precipitators in the return duct of a home.83 Air Treatment Ultraviolet Light Ultraviolet light can damage or kill many microorganisms, although the kill rate is rarely 100 percent. Gram-negative bacteria are especially sensitive to ultraviolet light, while acid-fast organisms and spores are very resistant.84 A combination of filtration and intensive ultraviolet light reduced bacterial aerosols by 90 to 92 percent, an insufficient reduction to protect fully from large bioagent releases85 but still providing a useful mitigation. The change in ultraviolet energy susceptibility of bacteria due to ambient humidity and the phenomenon of photoreactivation also should be considered when attempting to control airborne bacteria using ultraviolet light. Both humidity and visible light protect cells from ultraviolet light-induced damage.86 Installed ultraviolet light sources in central ventilation systems can keep surfaces free of microbial growth and may significantly reduce viability in deposited organisms (including resistant ones) over time.87 Steam Condensation Steam condensation has been used to clear aerosols from occupied spaces and could be an approach for increasing the rate of clearance in rooms in which releases have occurred.88 Local Exhaust Remediation of fungal contamination in large buildings is generally done under local exhaust ventilation. Done properly, this approach essentially prevents the spread of contamination outside the containment unit. This approach could be used for rooms where releases have occurred if a facility for exhausting air is in place. Reservoir Removal Generally, the first step in remediating any interior contaminated with fungi and/or bacteria is to remove all reservoirs of the organism. For biowarfare agents, this may mean removing carpeting, vacuuming then washing all surfaces, and disinfecting the space.89 Carpet is particularly a problem since one of its intended purposes is to trap dirt and make it invisible. Organisms are more likely to accumulate 82   P.C. Kemp, H.G. Neumeister-Kemp, G. Lysek and F. Murray. 2001. Survival and growth of micro-organisms on air filtration media during initial loading. Atmos. Environ. 35:4739-4749. 83   R.L. Jacobs and C.P. Andrews. 1989. Hypersensitivity pneumonitis treated with an electrostatic dust filter. Ann. Intern. Med. 110:115-118. 84   S. Miller and J. Macher. 2000. Evaluation of a methodology for quantifying the effect of room air ultraviolet germicidal irradiation on airborne bacteria. Aerosol Science and Technology 33:274-295. 85   Cundith et al., 2002. See note 80 above. 86   J. Peccia and M. Hernandez. 2001. Photoreactivation in airborne Mycobacterium parafortiutum. Appl. Environ. Microbiol. 67:4225-4232. 87   D. Menzies, J. Pasztor, T. Rand, and J. Bourbeau. 1999. Germicidal ultraviolet irradiation in air conditioning systems: Effect on office worker health and well being: A pilot study. Occup. Environ. Med. 56:397-402. 88   J.H. Edwards, D.M Trotman, and O.F. Mason. 1985. Methods for reducing particle concentrations of Aspergillus fumigatus conidia and mouldy hay dust. Sabouraudia 23(4):237-243. 89   Paraformaldehyde or chlorine dioxide may be used as disinfectants.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases and survive in carpeting than on bare floors. In high-risk environments, it may be appropriate to replace carpeting with hard surface flooring and install a mechanism for fast local exhaust to prevent spread of released microorganisms. Applicability for Controlling Releases Good filtration will lessen the impact of building releases, providing systems are properly designed and installed and air exchange rates are high. Obviously, all return air must go through filters if releases within the occupied space are to be controlled. HEPA filtration is not necessary to reduce aerosol concentrations significantly. Viruses, in particular, are often considered to be able to pass through most filter media. This assumption is based on the name "filterable viruses," which refers to the fact that viruses in liquid will pass through very fine filters (even some HEPA filters). However, in aerosol form, viruses are unlikely to be in the form of single virions, and dry filters are far more efficient than wet ones at capturing small particles. While electrostatic precipitation is considered a method of choice for preventing intrusion of outdoor particles, particle collection efficiencies are not likely to be high enough to reduce agent aerosols to a nonhazardous level. FINDINGS AND RECOMMENDATIONS Consistent with the organization of this chapter, the committee offers findings and recommendations related to either the outdoor or indoor backgrounds, as well as the types of organisms potentially present in background bioaerosols. Outdoor Aerosols Finding 3-1: Pollen concentrations are apparently low relative to the large bioagent releases that are likely outdoors. However, pollen concentrations are well known only as 24-hour averages and from relatively elevated sites (i.e., tops of buildings). Recommendation 3-1: Research is needed to determine concentrations of pollen plumes as they are released from plants and concentration variations on shorter time scales in the ambient air. Finding 3-2: Fungal aerosol plumes in outdoor air are likely to trigger false alarms in bioaerosol detectors, given the acute nature of the factors leading to massive releases. Fungal aerosols well in excess of 1,000 spores per liter have been documented, and with more accurate analysis, concentrations are likely to prove much higher. Recommendation 3-2: Fungal aerosols need to be monitored on a much shorter time scale using traditional approaches as well as those similar to the approaches that will be used in bioagent detectors. Finding 3-3: Virtually no good data are available in the literature on outdoor bacterial aerosols, with respect to either total bacterial counts or the types of bacteria in aerosols. These aerosols have been measured in concentrations exceeding 104 per liter and can trigger false alarms in nonspecific detectors. Recommendation 3-3: Studies need to be conducted evaluating concentrations of bacterial aerosols and their variability over relevant time scales using methods similar to those likely to be used in bioagent detectors. Finding 3-4: The types and concentrations of viral particles are unknown in outdoor air. Recommendation 3-4: Viral particles need to be assessed in outdoor air, along with parameters controlling releases and variability in the aerosols.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases Finding 3-5: Clouds of other particles have been reported sporadically in air. For example, soy aerosols have caused widespread asthma epidemics during unloading of ships in harbors, indicating clouds of soy particles that are transported to distant sites in concentrations high enough to cause illness. Other kinds of bioaerosols probably occur in a similar manner but have yet to be reported. Recommendation 3-5: An assessment of the likelihood of these other kinds of aerosols needs to be performed. Indoor Aerosols Finding 3-6: Pollen is probably not of great concern indoors, as long as outdoor concentrations are well studied. Recommendation 3-6: Studies of indoor pollen may safely be limited to understanding outdoor-to-indoor transport mechanisms. Finding 3-7: Fungal spore aerosols indoors will be problematic in buildings where active growth is occurring. Understanding the release parameters from such sources is a necessary precursor to predicting the frequency and magnitude of releases. Recommendation 3-7: Indoor fungal spore aerosols need to be studied using both traditional methods (which allow detection of unexpected organisms) and new methods. Finding 3-8: Bacterial concentrations are essentially unknown indoors in spite of the fact that many studies (using culture) have been performed. Recommendation 3-8: Indoor bacterial aerosols need to be mapped using methods that are less selective than culture methods. Finding 3-9: Many potential biothreat agents (e.g., Bacillus anthracis) are probably part of the natural indoor aerosol in some form or other (possibly not infective). These could become a source of false alarms in some detection systems. Recommendation 3-9: Surveys for specific organisms that are considered potential bioagents need to be conducted. Finding 3-10: Viral aerosols have not been studied indoors, nor has the presence or absence of specific agents been evaluated. Recommendation 3-10: Studies need to be done and should use the same methods as are expected to be used in bioagent detectors. Predictive Models Finding 3-11: If the high variability—both geographically and over time—in microbial aerosols is not understood and accounted for, it is likely to lead to an unacceptable level of false positive alarms in bioaerosol detection systems, thereby limiting the smallest attack they can confidently detect. Recommendation 3-11: The feasibility of using predictive models for bacterial and fungal aerosols to allow the development of algorithms that can normalize and differentiate the signal from these natural clouds from the bioaerosol detector signal should be studied.

OCR for page 23
Sensor Systems for Biological Agent Attacks: Protecting Buildings and Military Bases Sampling Strategies to Obtain Critical Missing Data Finding 3-12: Sampling in outdoor air represents a major effort but needs to be done in representative areas of concern, since geographic differences in bioaerosol concentrations and populations are profound. Air sampling is probably the method of choice for outdoors, although source sampling in particularly at-risk locations would be advisable. The primary use of source sampling is to allow for the extremely difficult problem of collecting air samples that are representative in time and space. Sources that could be evaluated outdoors include living and dead vegetation, soil, water, effluents from cooling towers, and composting materials. Some of these data may already be available in the plant pathology literature. Recommendation 3-12: A widespread outdoor air sampling network needs to be developed that will accumulate data for a wide variety of environments. In the United States, sampling at the EPA air monitoring sites would be a useful start. Some more intensive sampling (i.e., multiple sample sites within one community) should be done to determine the representativeness of any single sample site and to contribute to the development of predictive models. Sampling needs to be continuous and should cover several years, especially in view of current climate variability, and it should use not only the same detection methods but also the same types of sampling equipment that are likely to be used for bioagent detection. One possible approach to broad area monitoring would be to coordinate efforts of federal agencies—e.g., the Environmental Protection Agency (EPA), the Transportation Security Administration (TSA), and the National Institute of Standards and Technology, to develop protocols for periodically collecting air samples from various locations throughout the continental United States—and perhaps neighboring skies—for standardized testing. However, equipment would have to be modified so as to allow collection of samples on a time-discriminated basis (minutes). Finding 3-13: The major problem with indoor air sampling is the variety of buildings and conditions that occur and the sporadic nature of bioaerosol releases. For example, in one building sampled over 2 weeks, 1-minute average peak concentrations of actinomycetes were in excess of 10,000 cfu per cubic meter on a Monday morning and near zero on Thursday afternoon.90 Likewise, in the same building, more than 5,000 Sporobolomyces cells were recovered on only one out of 850 samples collected in the building over a week. Source samples are especially important indoors. Most indoor bioaerosol sources release sporadically, so that massive aerosols may only be produced during a few minutes once or twice over periods of several months. The most common source sample used indoors is dust vacuumed from floors or surfaces. While culture drastically underestimates concentrations and misrepresents population composition, still many organisms have been recovered in high concentrations (concentrations of 107 to 108 per gram of vacuumed material are not uncommon). For ducted buildings, air sampling in the return air ducts is relevant. Recommendation 3-13: Longitudinal studies should be conducted of indoor bioaerosols associated with specific sources and activities in the types of buildings expected to be susceptible to attack. 90   H. Burge, Harvard University. 2003. Unpublished data.