Elements of Change 1995: Session I Ward

The International Global Atmospheric Chemistry Program (IGAC), a core project of the International Geosphere Biosphere Program (IGBP), has emphasized experiments in tropical regions of the world to reduce the uncertainties regarding the contribution of biomass burning to the buildup of greenhouse gases in the atmosphere. Most of the research discussed by Ward was done in response to the objectives for the IGAC Program and relates to the global carbon balance. Biomass burned globally is estimated at 6,366 teragrams per year at combustion efficiencies ranging from 0.80 to 0.97. Combustion efficiency is the ratio of carbon released as carbon dioxide (CO₂) to the total carbon released by a fire. This is basically the efficiency with which a fire converts carbon to CO₂.

According to the FAO, deforestation has increased from 11 million hectares in 1980 to nearly 17 million hectares in 1991. It is more difficult to quantify CO₂ emissions from biomass burning than from fossil fuel use. In per capita terms, some developing countries may now be generating more CO₂ through land use change than developed nations release from fossil fuel combustion.

The Vegetation Fire Information (VFI) System is a proposed method, accepted by IGAC, for assessing the temporal and spatial extent of fires in different ecosystems in different regions of the world and their contribution to smoke emissions [Malingreau et al., 1993]. The VFI System relies on satellite methods of detecting fires. AVHRR and other satellite systems do not typically detect all individual fires. Generally, very little information is provided regarding the size of fires, making additional data necessary. The temporal effect of change in area burned per lit-up pixel is adjusted based on the application of fire behavior models that respond to weather. The VFI System incorporates the effect of land-use change and succession models, to develop knowledge about fuel loading and the structure of fuel. It also makes use of fire behavior modeling coupled with land cover modeling, and emissions modeling to provide estimates of source strength for regional areas. The results from the VFI System are proposed to be checked by flying aircraft along transect lines to estimate the flux of emissions. Ultimately, through an iterative process, the predictability of source strength for a region can be improved to an acceptable level of accuracy.

When biomass combines with oxygen in a heated environment, combustion occurs. The completeness of combustion in producing CO₂ is dependent on the chemical mixture of fuel and air. Products of combustion, many of which technically are products of incomplete combustion, can be described by the combustion efficiency. The combustion efficiency ranges from 0.80 to as high as 0.97. Variables affecting smoke emissions include fuel type, fuel load, fuel chemistry, fuel moisture content, packing ratio for the fuel particles, surface area to volume ratio, and method of ignition. Correlating fuel parameters and other factors of combustion can help predict combustion efficiency and emission factors for fires.

The average release of carbon from a fire is approximately as follows: 88% CO₂, 9.5% Carbon Monoxide (CO), 0.5% Methane (CH4), 0.5% non-methane hydrocarbons (NMHC), and 1.5% in particles of less than 2.5 microns diameter (PM2.5).

Vegetation consumed by fire is about 50% carbon (ranging from 47.5% in savanna vegetation to 51-52% for coniferous ecosystems). The measured carbon in a unit volume of emissions above background levels is used in estimating the amount of fuel consumed in producing the emissions and expressed as an emission factor (g/kg).

In per capita terms, some developing countries may now be generating more CO₂ through land use change than developed nations release from fossil fuel combustion.

The Fire Atmosphere Sampling System (FASS) has been used extensively by Ward et al. [1992] for characterizing smoke emissions on areas with measured fuel characteristics (see Figure 20.1 ). The FASS is an instrument de veloped specifically to study smoke emissions from a range of fire conditions as a function of the rate of carbon release. It is computer controlled and measures CO₂, CO, and NO concentrations, vector wind components, temperature, and other parameters that are logged once each second. Grab samples of particulate matter are collected concurrently with samples of the combustion gases by phase of combustion. A light absorption meter and integrating nephelometer measure the light absorption and scattering by particles by phase of combustion, respectively. Samples of the PM2.5 are collected on filters at the top of a tower and the gases are sampled through a line running from the top of the tower to the ground. The main part of the FASS is placed underground near the base of the tower. The FASS is completely automatic once it is armed and goes through background sampling, calibration, and phase of combustion sampling automatically.

The FASS is an instrument developed specifically to study smoke emissions from a range of fire conditions as a function of the rate of carbon release.

Emission factors for the release of emissions of CH4 for savanna fires are about 12% of the emission factors for fires used to burn debris during conversion of primary forested areas. Emissions of CH4 from burning of cut second-growth forest sites is intermediate to savanna and primary forest sites. Emission factors for pasture burns in the Brazilian Amazon have been found to be about 20% larger than for cerrado area savanna burns.

For savanna ecosystems ranging from semi-arid to humid savanna ecosystems, models have been developed for predicting combustion efficiency using the ratio of the amount of grass to the sum of grass and litter. Models have also been developed for estimating emission factors for a range of compounds and particulate matter as a function of combustion efficiency for these same savanna ecosystems. There is a need to incorporate net primary productivity models into the system for smoke production for the African savanna ecosystems.

Carbon released from fires used for converting primary forested areas to agriculture in the Amazon can last for several days with over 50% of the biomass consumed through smoldering combustion. Areas of savanna burned by fires usually have smoldering combustion carbon released that is 10 to 15% of that during the flaming phase of combustion. Carbon models that are coupled with land use data by region are needed.

Emissions of hydrocarbons and other carbon-containing compounds were studied for charcoal making in Zambia and Brazil. A preponderance of the hydrocarbons are oxygenated for the two types of kilns studied: brick style for Brazil and earthen construction for Zambia. Both kiln types were operated using wood cut from open forested savanna ecosystems. The Zambian earthen kilns converted 23-25% of the wood to charcoal whereas the Brazilian brick kilns were of a higher efficiency, converting approximately 35% of the wood to charcoal. It is practical in Brazil to transport wood to the brick kilns, whereas in Zambia, this may be impossible. Fire management programs are needed to sustain the production of charcoal and recovery of the forested areas used in the production of wood for fire wood and the making of charcoal.

The FASS packages have been used to develop carbon release and emission factors for fires used with shifting cultivation in southern Africa. The chitemene style of agriculture is used to concentrate nutrients on large areas of 2 to 5 hectares on garden spots of 0.5 to 1 hect are. The tops and limbs from trees are harvested and used to create a uniformly deep pile for the garden spot. The pile is burned immediately prior to the onset of the rainy season, typically in October. Agriculture is practiced on each garden spot for 5-7 years before abandonment. The active pile burns and burns of the fallow chitemene sites have been studied for smoke emissions. The results for the fallow chitemene sites follow the same trends found by Ward et al. (1995) for burns of miombo and dambo savanna ecosystems. Fire management practices are needed to decrease the time of recovery of forested areas to sustain the chitemene style of agriculture.

The VFI System will require good emission factors and biomass consumption models for the major ecosystems on a global scale. This research needs to be extended to other ecosystems, especially those in Asia, to complete the development of models for predicting emissions from fires.

The Zambian earthen kilns converted 23-25% of the wood to charcoal whereas the Brazilian brick kilns were of a higher efficiency, converting approximately 35% of the wood to charcoal.

Ward, D. E., R. A. Susott, J. B. Kauffman, R. E. Babbitt, D. L. Cummings, B. Dias, B. N. Holben, Y. J. Kaufman, R. A. Rasmussen, and A. W. Setzer. 1992. Smoke and fire characteristics for Cerrado and deforestation burns in Brazil BASE-B Experiment. J. Geophys. Res. 97:14,601-14,619.

Malingreau, J.-P., F. A. Albini, M. O. Andreae, S. Brown, J. S. Levine, J. M. Lobert, T. A. Kuhlbusch, L. Radke, A. Setzer, P. M. Vitousek, D. E. Ward, and J. Warnatz. 1993. Group Report: Quantification of fire characteristics from local to global scales, Fire in the Environment: The Ecological, Atmospheric, and Climatic Importance of Vegetation Fires . edited by P. J. Crutzen and J. G. Goldammer, John Wiley & Sons Ltd., pp. 329-343.

Ward, D. E., W. M. Hao, R. A. Susott, R. A. Babbitt, R. W. Shea, J. B. Kauffman, C. O. Justice. 1995. Effect of fuel composition on combustion efficiency and emission factors for African savanna ecosystems. Accepted for publication J. Geophys. Res.

Susott, R. A., S. P. Baker, G. Olbu, D. E. Ward, and J. B. Kauffman. 1995. "Carbon, hydrogen, and nitrogen content of tropical ecosystem fuels," in Biomass Burning and Global Change, edited by J. S. Levine, in preparation.