The objectives of this project are to improve the PM2.5 (particles smaller than 2.5 micrometer diameter) emission factors (EFs) for burning of common southeastern U.S. vegetation types, test for effects of fire conditions on EFs, and evaluate improvements in the method for estimating EFs. The hypotheses to be tested are that current methods over-estimate EFs and that climate and fire fuel conditions are major factors controlling PM2.5 EFs. The methodology will involve collection of PM2.5 samples followed by measurements of composition using a Multi-Element Scanning Thermal Analysis (MESTA) technique developed recently by one of the principal investigators that can be used to distinguish biomass burning aerosol (BBA) from background aerosol. The BBA mass measurements will be combined with measurements of the stable carbon isotope composition of carbon dioxide generated by the fire. The carbon isotope ratios can be measured in real-time using a commercially available laser-based system and can be used to quantify carbon dioxide emitted by the fire. The ratio of BBA to fire carbon dioxide is the EF.
The results will be used to assess the validity of the current fire PM2.5 EF database, which is the basis for all fire PM2.5 emission prediction and modeling. The research will likely establish a new standard for future PM2.5 EF measurement, which would improve calculation of EF and estimates of PM emission nationally and globally. Products generated from the project will be delivered directly to other scientists and to state and federal regulators, which will lead to improved prescribed fire application and air quality policy in order to protect the health and safety of the general public. At least two workshops will be held to provide direct training and information transfer regarding the developed methods of PM2.5 and fire carbon dioxide emissions and calculation of improved EFs. A post-doctoral associate and minority graduate and undergraduate students from Florida A&M University, a Historically Black University, will have the opportunity to be trained and mentored in areas of air quality research.
Prescribed burning is a beneficial forest management tool commonly practiced in many parts of the country. Smoke (particulate matter or PM, in a technical term) emitted from prescribed burning is an air quality and public health concern in the affected areas. PM is also a source of black carbon, which plays a role on the global heat budget and climate change. PM2.5 (nano-sized PM diameter <2.5 µm) is of particular concern because of its longer suspension time and traveling distance in the air. Modeling PM2.5 emission from prescribed burning is entirely dependent on the use of emission factors (EFs), which estimate the amount of PM2.5 produced per weight of biomass burned. Almost all air quality professionals rely on published EFs, but their accuracy is largely unknown because few studies have been conducted to evaluate and validate them. We initiated this study to examine the accuracy of the current method of PM EF determination. Specifically, we examined the two principle assumptions that the current PM EF determinations are based on: 1) the fire emitted CO2 is the difference between the total CO2 concentration measured and the pre-fire ambient CO2 concentration (the excess concentration assumption); 2) CO2 and PM2.5 have similar diffusion rates in the air such that their simultaneously sampled concentration ratio is relatively the same regardless of where we sample them (no differential diffusion assumption). To examine the first assumption, we used the natural distinctive carbon isotopic ratio between the biomass burning emitted CO2 and the ambient CO2 to quantify biomass CO2 directly without the excess concentration assumption. To examine the second assumption, we determined CO2/ PM2.5 ratio of fires in two places simultaneously: one directly in the convection columns where diffusion was minimal and the other outside the convention columns where diffusion was significant. The results of our study show that excess concentration assumption overestimates 11-13% of the biomass emitted CO2, which leads to an underestimation of PM EF values by 7-13% with this factor alone. We suggest a general correction factor of 10 % for the underestimation of PM EF in the current database, which uses the excess concentration assumption. The no differential diffusion between CO2 and PM2.5 assumption was also proven to be wrong. The no differential diffusion assumption may cause a PM EF measurement to be biased as much as ten times more depending on where and how much of the diffusion process was allowed to take place before a sample was taken. A general correction factor for the differential diffusion effect cannot be suggested because of its highly variable nature. Our study suggested, however, that this differential diffusion effect between CO2 and PM2.5 could be minimized, if a sample is taken directly from the convection column of a fire before significant diffusion can take place. The results of our study can be applied to improve the accuracy of determining PM EF, which is the basis for estimating PM emission from biomass burning in regional, national and global scales. This study also improves our understanding of the factors that affect PM emission in prescribed burning. Gaining this knowledge enables us to minimize PM emission from a prescribed burning and mitigate its effect on air quality and public health. This study also showed that the multi-element scanning thermal analysis (MESTA) is a simple and effective tool to characterize PM. Through MESTA, we can differentiate PM emitted between flaming and smoldering combustion phases. This study is the first to identify a PM combustion phase indicator, which is complementary to the existing combustion phase indicator of gases (modified combustion efficiency or MCE). This PM combustion phase indicator could be very useful in the study of fire emitted PM in the future. The development of MESTA technology in this study shows that MESTA can be a powerful tool of analyzing many other environmental samples, which have been difficult due to their heterogeneous and complex nature such as in soil, sediment, coal and petroleum samples.
