3506B24 Final Report
Download 11.49 Mb.
|
- Navigate this page:
- 3 BACKGROUND
Figure 2: Annual total acres burned by prescription in comparison with areas and numbers of wildfires occurred on Fort Benning’s military installation since 1985. Proper management may require as much as 1/3 of the land to undergo treatment by fire each year. These activities, however, can contribute significantly to already burdened local and regional air pollutant loads, leading to increased exposure of larger populations to elevated pollutant levels. For example, at the Fort Benning military reservation, which covers a total area of approximately 180,000 acres in the lower part of central Georgia and Alabama, approximately 32,000 acres must be prescribed burned each year, ideally during the growing season (i.e., March through August). However, Fort Benning is located six miles southeast of Columbus, Georgia, and the Columbus Environmental Committee is viewing the prescribed burning operation at Fort Benning as a potential contributor to the ground-level ozone (O3) and particulate matter (PM) pollution levels. This became a public relations concern for Fort Benning and the prescribed burning operation was moved to the fall and winter months [Larimore, 2000]. In order to help minimize the potential contribution from prescribed burning emissions to already burdened local to sub-regional air quality, the Georgia Forestry Commission (GFC) plays the important role of fire weather/smoke management forecasting, training, public relations, and fire permitting. Weather forecasting is provided for 17 geographical locations within the state. Spot weather forecasting is available via the Internet for anywhere in Georgia from http://weather.gfc.state.ga.us/. These valuable resources had been utilized in the evaluations and scientific analyses of the collected data presented below. The GFC administers burn permits and smoke management from 140 locations. Permits are required from the GFC for all burns except agricultural burning and leaf pile burning. Although federal agencies are exempt from state laws pertaining to outdoor burning, it is the policy of the USFS and the Department of Interior agencies located in Georgia to comply with GFC requirements. Table 1 summarizes the total areas burned for agricultural, silvicultural and land clearing purposes as an annual average from a five year data record (1999-2003) representing monthly totals for all 159 GA counties. Figure 3 illustrates the monthly variability of these averages across GA (top) and the variability of the average fractions from each class of burn for each month (bottom). The prescribed burning conducted on the Fort Benning military installation between November 2000 and October 2003 are shown for comparison. The importance of agricultural burns becomes evident as a significant fraction of the openly burned areas in May and June, when between 23,000 and 32,000 acres are being burned state-wide, while forest management (silviculture) burns both on privately owned land state-wide and on the military installations are less intense in the summer months, as mentioned before. Table 1: Annual average and standard deviation of GFC monthly total areas burnt by prescribed fires in all (max. 159) Georgia counties between 1999 and 2003, in comparison with Fort Benning burns between 11/00 and 10/03.
Figure 3: Monthly averages of total areas burned for agricultural, silvicultural and land clearing purposes from monthly county-level data spanning 1999-2003 in GA, with Fort Benning data spanning from 11/00 to 10/03 on a logarithmic (!) scale (top); average distribution of the main three burn categories contributing to the sum total (bottom). Considering the large amounts of areas burned by prescribed fires, and further considering that these areas are vastly distributed within the state of Georgia and the entire South-East, it is conceivable that the prescribed burn activities have a significant impact on regional air quality beyond the state of Georgia. The Georgia Department of Natural Resources Environmental Protection Division (EPD) in close collaboration with representatives from GFC, the Department of Transportation, the Department of Public Safety, the federal military bases located in Georgia, federal land managers associated with the US Fish and Wildlife Service and the US Forest Service, and groups and associations representing environmental interests or private individuals in Georgia, currently develop a Smoke Management Plan (SMP) with purpose to “successfully balance the use of prescriptive fire while providing the citizens of the State of Georgia with healthy air and a vibrant ecology”. Due to the lack of knowledge and very limited understanding of the relationships between prescribed burning emissions, air quality, human exposure and health risks, the SMP is being developed based on the requirements mandated by the US Environmental Protection Agency through its National Ambient Air Quality Standards (NAAQS) for criteria pollutants, with special emphasis here on fine PM, or PM2.5, referring to airborne particles with aerodynamic diameters less than 2.5 microns. Ultimately, this study will provide information towards the validity of the current NAAQS in areas that are impacted regularly by emissions from prescribed burn activities, which will potentially help develop SMPs that more effectively protect human beings. The following section 3 will briefly summarize the current scientific understanding of the role of airborne PM in the more general context of urban air quality, and under special consideration of prescribed burning in particular. Subsequent descriptions of observations made during the multi-year Fall-line Air Quality Study (FAQS) that led to this Prescribed Burn Study (PBS) will follow in section 4, before the data collection, measurement results and interpretations are presented in sections 5 and 6. Conclusions will be drawn in the last section before the list of references cited. Auxiliary data tables and information will be listed under the Appendices section. 3 BACKGROUND Airborne particulate matter (PM) represents a complex mixture of organic and inorganic substances of both solid and liquid phase, which suspended in air, and containing volatile, semi-volatile and non-volatile species and therefore being in equilibrium with reactive gas-phase species, defines the aerosol. PM suspended in air varies in size, composition and origin. Apart from its extensive properties and the individual chemical composition, the aerosol is characterised by its particle number and mass size distributions and the degree of mixing. Therefore, the chemical characterisation of an aerosol involves multi-dimensional challenges. Whereas the bulk composition is dominated by the large particle size modes, size-resolved analysis is necessary in order to address the characteristics of the different modes. These modes differ in their sources, sinks and atmospheric residence times, and, hence, in their chemical composition. The degree of mixing addresses the chemical homogeneity of individual size modes, both the mode as a whole as well as on the single particle level (e.g. internal vs. external mixtures). This distinction is relevant since many properties of the aerosol are not represented by the sum or arithmetic mass-weighted or number-weighted means. Moreover, as a consequence of both different direct emissions and ageing in the atmosphere (condensation, coagulation, cloud processing), shell-like structures with non-homogeneous chemical composition of individual particles may result. This implies a chemical behaviour of the particle which may be dominated by the chemical properties of the surface and which is not represented by the overall mass-weighted particle composition. In addition, the apparent particle size depends on its morphology and on the method by which it is determined (e.g. by an electrostatic, optical, or an inertial particle sizer), and may also vary as a function of thermodynamic conditions, e.g. the relative humidity. Particles rich in carbon are particularly difficult to characterize chemically due to the enormous complexity of their chemical composition (e.g. Rogge et al., 1993, 1996; Schauer et al., 1996; Saxena & Hildemann, 1996). As has been noted quite early (Winkler, 1974; Rogge et al., 1993), apart from significant fractions of the particulate organic matter (POM), which are soluble in organic solvents or in water, there is also a significant fraction which is soluble in both types of solvents. The non-elutable fraction of the POM is also very significant in mass (Schauer et al., 1996; Kubátová et al., 2001). The most significant classes of organic substances in terms of total mass, which have been identified in urban aerosols are alkanes, fatty acids, dicarboxylic acids, and sugars (Rogge et al., 1996; Gelencsér et al., 1998; Röhrl & Lammel, 2001; Zdrahal et al., 2002). Comparatively less mass is contributed by otherwise relevant low-molecular weight substance classes, such as the polycyclic aromatic hydrocarbons (PAHs). The insoluble fraction is dominated by high-molecular weight organic material, which contains aliphatic polyols and polyphenols, proteins, unsaturated aliphatic and aromatic polyacids, and so-called humic-like substances, HULIS, which were also associated with direct biomass burning emissions [Graham et al., 2002]. These polyacidic compounds are thought to consist predominantly of high-molecular-weight “air polymers” with properties resembling those of HULIS [Havers et al., 1998; Zappoli et al., 1999; Decesari et al., 2000; Facchini et al., 2000; Krivácsy et al., 2000; Krivácsy et al., 2001; Fuzzi et al., 2002; Franze et al., 2003). As has been hypothesized only very recently (Gelencsér et al., 2003; Limbeck et al., 2003) such polymeric material can be generated in the aqueous phase of airborne particles. This hypothesis is relevant for the potential health and climate effects of particles rich in carbon. It has also been shown that neutral, high-molecular-weight oligo-saccharides are primary thermal degradation products of plant poly-saccharides, derived either from the incomplete breakdown of polymeric chains [Köll et al., 1990] or from recondensation of the initial monomeric anhydrosugar products [Shafizadeh and Fu, 1973]. Thus it is likely that a significant fraction of the water soluble fraction (WSOC) is composed of a range of high-molecular-weight species, both acidic and neutral, which would be expected to remain undetected by GC analysis because of their low volatility. However, by employing solid phase extraction (SPE), WSOC can be further fractionated into hydrophobic (mainly humic and fulvic acid-like compounds) and hydrophilic organic compounds (e.g., neutral, mono- and di-carboxylic acids) [Thurman and Malcolm, 1981; Mukai and Ambe, 1986; Havers et al., 1998; Mao et al., 2000; Fuzzi et al., 2001; Hoffer et al., 2004]. Biomass burning includes wild fires (natural), prescribed burning, burning of agricultural wastes, and domestic fuel-wood (anthropogenic) burning, and has raised considerable concern because it is a significant source to regional and global particulate matter (PM) in the atmosphere. In addition, it is reported to contribute up to 40% of CO2, 32% of CO, 10% of CH4, 24% of non-methane hydrocarbons (NMHC), 21% of NOx, 38% of tropospheric O3, 39% of organic particulate matter, and 86% of elemental carbon globally [Levine et al., 1995]. Biomass emissions impact global climate by perturbing solar radiation (e.g. CO2, CH4, PM) and by destroying ozone in the stratosphere (CH3Cl, CH3Br) [Levine et al., 1995; Cicerone, 1994]. On regional and local scales, biomass emissions contribute to photochemical production of ozone in the troposphere due to the release of precursors (e.g. NOx, NMHC). The burning of wild land (wild fires and prescribed burning) is the largest single source of directly emitted PM2.5 in the United States, especially in the southeastern area [US EPA, 2000]. Up to 80% of PM2.5 from biomass fires are POM, which includes significant amounts of solvent-extractable organic compounds such as carcinogenic, mutagenic polycyclic aromatic hydrocarbons (PAH) and polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/F) [Hays et al., 2002; Freeman and Cattell, 1990; dos Santos et al., 2002; Bacher et al, 1992; Nestrick and Lamparski, 1982; Chagger et al., 1998; Gullett and Touati, 2003]. Many studies have been conducted to characterize emissions from the burning of different biomass materials and combustion conditions (e.g., residential fireplaces, wood stoves or open burning), and to determine impacts of biomass burning on the atmospheric system. While many emissions were characterized in laboratory set-ups, our understanding of the atmospheric effects and impacts on local to regional air quality is still limited. Furthermore, identifying specific “fingerprint” biomarkers for different sources of burning, remains an important task for researchers as well. In addition, little has been done to assess human exposure to biomass burning and the health impact on humans. In the United States, exposure to wood smoke is seen to cause a negative effect on respiratory and pulmonary function in children [e.g. Larson and Koenig, 1994]. While reducing the threat for wildfires, and despite a much reduced fuel load, prescribed fires are still a form of biomass burning: they produce combustion byproducts that are potentially harmful to human health and welfare [Hardy and Leenhouts, 2001; Battye and Battye, 2002; Crutzen and Andrea, 1990; Cheng et al., 1998; GCVCT, 1996; NWCG, 2001]. Biomass burning is among important sources for atmospheric aerosols, emitting a variety of specific tracers. For example, particulate potassium (K+) often serves as a diagnostic tracer, because the combustion of plant matter, which contains K+ as a major electrolyte within its cytoplasm, releases large amounts of submicron particles rich in K+, whereas soil- or sea-spray derived submicron aerosol usually is low in K+ [Andreae et al., 1996; Cachier et al., 1991; Gaudichet et al., 1995]. Combustion processes release primary PM (soot), i.e. an aerosol containing elemental carbon (EC) formed by the pyrolysis of biomass. It is of graphitic nature, but contains aromatic hydrocarbons, functional groups of various type as well as chemisorbed water on its surface (Smith et al., 1989; Chughtai 1999a & b). As already mentioned above, there is now growing evidence from field and laboratory studies that the ageing of soot may lead to not only hygroscopic (e.g., Kotzick & Niessner, 1999), but even to water soluble POM (Decesari et al., 2002). The contribution of EC to PM, however, is highly variable and can range from 25 to 4 %, depending on the fire’s level of flaming to smoldering, respectively [Khalil and Rasmussen, 2003]. Various organic particulate compounds are also uniquely found in biomass burning emissions, and have been used for source apportionment [Schauer et al., 1996, Zheng et al., 2003]. Furthermore, certain gaseous species are considered useful tracers, including acetonitrile (CH3CN), methylchloride (CH3Cl), and less uniquely, but especially in conjunction with the sum of odd nitrogen oxides (NOy), carbon monoxide (CO) [Blake et al., 1996, Blake et al., 1999, Reiner et al., 2001]. While the above tracers have atmospheric lifetimes of several days and weeks, even up to two months in the case of CO, a great variety of reactive gaseous species with much lower lifetimes, e.g. NH3, sesqui-terpenes, oxygenated volatile organic compounds (OVOC) and other organic and inorganic gases are being emitted, that act as potentially effective precursors for the formation of new particles, the so-called Secondary Organic Aerosol (SOA). Due to their reactivity, these gaseous precursors are hard to measure in the field and have been detected mainly in benchmark-type laboratory set-ups [Lobert et al., 1991; Yokelson et al., 1996; Goode et al., 1999; Hays et al., 2002; Christian et al., 2003; Christian et al., 2004], or in isolated intensive field campaigns in different locations for a reduced amount of species attempting to distinguish emissions from flaming versus smoldering stages [Andreae et al., 2001; Friedli et al., 2001, Goode et al., 2000; Griffith et al., 1991; Hobbs et al., 2003; Holzinger et al., 1999; Nance et al., 1993; Worden et al., 1997; Yokelson et al., 1997; and 1999]. Christian et al. [2003] detected direct biomass burning emissions of acetaldehyde, phenol, acetol, glycolaldehyde, methylvinylether, furan, acetone, aceto-nitrile, propene-nitrile, and propane-nitrile, most of which are OVOC, which further reinforces the importance of these reactive compounds in the atmospheric formation of SOA. SOA formation by gas/particle partitioning of semi-volatile products from the photo-oxidation of reactive organic species has been of enormous scientific interest recently, because of the potential implications on global climate, visibility, and human health. Besides the above mentioned OVOC, other potential SOA forming precursors primarily include mono-terpenes (C10H16 ringed structures, e.g. α-, and ß-pinene, limonene) and sesqui-terpenes (C15H24 multi-ringed, e.g. α-cedrene, α-copaene, α-humulene, ß-caryophyllene) from biogenic sources (Hull, 1981; Altshuller, 1983; Hoffmann et al., 1997; Odum et al., 1997; Wangberg et al., 1997; Jang and Kamens, 1999; Kamens et al., 1999; Noziere et al., 1999; Yu et al., 1999; Glasius et al., 2000; Jaoui and Kamens, 2001), and aromatics from anthropogenic sources (Jeffries,1995; Odum et al., 1996). Atmospheric oxidation reactions of these organic compounds create multi-functional oxygenated or nitrated semi-volatile organic compounds (SVOC), i.e. organic acids, diacids, and aldehydes from the oxidation of terpenes, and result in SOA formation via either a self-nucleation process or gas/particle partitioning on pre-existing PM. Due to their reactivitiy and extremely short life-time, the sesqui-terpnes, which are believed to have 2 to 3 times higher SOA yields than mono-terpenes, are very difficult to measure directly in the atmosphere. Therefore, PM formation and growth linked to sesqui-terpenes has only been observed in laboratory chamber studies thus far. A heterogeneous hemiacetal/acetal formation was observed from the reaction of aldehydes with alcohols in the presence of light and photo-oxidants within a few hours [Jang and Kamens, 2001]. The studies showed that on time-scales between several minutes to few hours, aldehydes, which can be either photochemically produced in the atmosphere or directly emitted by open burning, undergo heterogeneous reactions accelerated by acid catalysts, like sulfuric acid, leading to higher aerosol yields than when the acid compound is absent. A number of more recent investigations supported these findings, and furthermore found that isoprene, a compound emitted in large quantities by natural vegetation and previously thought to be uninvolved in producing atmospheric aerosols, plays a potentially major role in that process [Limbeck et al., 2003; Claeys et al., 2004]. This is especially important in the context of biomass burning, as isoprene not only accounts for up to 50% of the non-methane hydrocarbons (NMHC) in the atmosphere, but is emitted in significant quantities during the open burning of biomass as shown below. Isoprene like many other volatile organic compounds (VOC) emitted by biomass burning are the primary “fuel” for the atmospheric formation of ozone (O3) in the presence of nitrogen oxides (NOx = NO + NO2) and sunlight. The VOC oxidation, like nearly all atmospheric oxidation processes, is initiated by the hydroxyl radical (OH), which itself is very reactive, and therefore, the path on which it becomes recycled is of critical importance. Under “clean” conditions it directly oxidizes carbon monoxide (CO) and methane (CH4) to convert into HO2 which in turn is recycled back to OH via reaction with either NO or O3. In typically polluted urban environs, the NO-recycling path is favored forming nitrogen dioxide (NO2). NO2 is known to be the only source for ozone production via its photolysis in the 320-430 nm spectral range, dissociating into NO and O. However NO immediately reacts with O3, so that no net O3 is formed. The oxidation path of OH with alkanes (e.g. CH4) and most unsaturated, biogenic or anthropogenic NMHC (e.g. alkenes, aromatics, alkynes) leads to alcyl radicals Rn with n-2 C atoms which in turn react explicitly with atmospheric oxygen to form organic peroxy radicals RO2 (here short for RnO2, and sometimes referred to as ROx = RO2 + HO2 as the total peroxy radicals). In polluted urban areas, NO is oxidized by RO2, representing the source for a net ozone production since NO2 is formed without consuming O3. Thus NOx are the catalyst for the VOC (fuel) consumption in the process of photo-chemical O3 production. The variety and amounts of VOC released into the atmosphere from biomass burning is quite large, and so is their reactivity, i.e. individual VOC have different impacts on the chemical formation of ozone, which are discussed here briefly. Differences in the reactivity of individual VOC to ozone formation have been documented in laboratory chambers and other controlled conditions. Consider the function P(O3)j defined as the rate of ozone production from VOC species j. Since VOC oxidation is usually initiated by reaction with OH, P(O3)j can be approximated by  (1)
where Cj is the concentration of VOC species j, COH is the OH atmospheric concentration, kOH(j) is the rate constant for the reaction between species j and OH, and (j) is the ozone yield, defined as the number of ozone molecules produced for each carbon atom of species j that is oxidized. Equation (1) shows that the relative importance of one VOC species to another will depend upon the relative magnitudes of the product of three variables: Cj, kOH, and . For this reason, a species with a large concentration will not necessarily be an important ozone precursor if it is un-reactive with OH or ineffective in producing ozone. Conversely, a species with a small concentration may still be an important ozone precursor if it is extremely reactive. Because of the large variability of Cj and kOH compared to these two variables are the dominant factors that determine the relative contribution of VOC species to ozone production. To account for this combined effect the propylene-equivalent method as defined by Chameides et al.,(1992) can be applied. This reactivity-based method is based on the variable Propyl-Equiv(j) defined as Propyl  (2)
Equation (2) is a measure of the concentration of species j on an OH-reactivity-based scale normalized to the reactivity of propylene C3H6. Therefore if a VOC species has an ambient concentration of 10 ppbC and it is twice as reactive as propylene, it will have a Propyl-Equiv of 20 ppbC. If, on the contrary, the species is half as reactive as propylene, it will have a Propyl-Equiv of 5 ppbC. This method will be applied to the emission measurements in section 6. Download 11.49 Mb. Share with your friends:
| |||||||||||||||||||||||||||||||||||||||||||||