|BOTLAGUDURU, V - Texas A&M University|
|MCGEE, R - Texas A&M University|
|PARNELL JR, C - Texas A&M University|
Submitted to: National Cotton Council Beltwide Cotton Conference
Publication Type: Proceedings
Publication Acceptance Date: 2/12/2010
Publication Date: 4/26/2010
Citation: Botlaguduru, V.S., Wanjura, J.D., Mcgee, R.0., Parnell Jr, C.B. 2010. Comparison of AERMOD and ISCST3 emissions factors for PM from cotton harvesting. In: Proceedings of the Beltwide Cotton Conferences, January 4-7, 2010, New Orleans, LA. 2010 CDROM. p. 611-621.
Interpretive Summary: Regulation of particulate pollution continues to intensify across the US. Agricultural operations once exempt from many particulate regulations are now required to comply with air quality permitting requirements and, in some states, submit management plans that detail the actions the operation will undertake to help reduce stationary and fugitive source emissions. Cotton harvesting operations were identified as a major source of particulate matter pollution in the San Joaquin valley of California and as a result, cotton producers in CA have been targeted by regulators to reduce emissions. Accurate emission factors must be used in the regulatory process to ensure fair regulation for all industries. However, the characterization of particulate emissions from agricultural operations is complex due to factors such as the shape and size of emission sources, size and speciation of the particulate matter emitted, and the interaction of the source and natural environment. This complexity is not well understood by many regulators and has resulted in the inequitable regulation of agricultural sources. Instead of in situ particulate matter concentration measurement, a tool regulators use in the permitting and emissions inventory process that simplifies the analysis is dispersion modeling. The objective of this work was to investigate the influence of dispersion modeling techniques on resulting emission factors for cotton harvesting operations. Two methods were investigated, one modeling the harvesting operation as a single, continuous area emission source, and the second modeling the operation as a series of line sources emitting particulate matter. Results indicate that the methods tested produced similar emission factors. Additionally, a comparison of emission factors developed using different dispersion models showed the emission factors to be significantly different indicating that emission factors are model specific. These findings help to elucidate the complexity of characterizing agricultural particulate matter emissions and will help regulators avoid the use of inappropriate emission factors in dispersion modeling.
Technical Abstract: Agricultural production and processing operations, including cotton harvesting, are facing increasing regulatory pressure from state and federal regulators. In order to regulate these operations equitably, accurate emission factors (EF) representative of the true level of particulate matter emissions are needed. The objective of this study was to evaluate the influence of dispersion modeling techniques on resulting emission factors for cotton harvesting operations. EFs for TSP, PM10, and PM2.5 from cotton harvesting were determined using the EPA recommended dispersion model, AERMOD. Modeling results for two different methods were analyzed. Method 1 modeled the operation as a single area source while Method 2 modeled the operation as a series of line sources. Method 1 EFs for true PM10 were 154 +/- 43, 425 +/- 178 kg/km2 for six-row and two-row harvesters, respectively. Method 1 EFs for true PM2.5 were 5.46 +/- 1.42, 15.4 +/- 6.46 kg/km2 for six-row and two-row harvesters, respectively. The results of this study indicate that EFs developed using Method 1 and Method 2 were not statistically different. Contrary to our hypothesis, the results lead to the conclusion that modeling method (Method 1 or Method 2) would not cause difference in EFs. This is an important finding and it suggests that the protocol developed at Texas A&M for developing EFs for area sources (Method 1) can be used for harvesting operations. This would save valuable time in the modeling phase of projects aimed at developing EFs. Additional work comparing EFs from AERMOD and the formerly recommended regulatory dispersion model ISCST3 were conducted. This comparison observed that, for a six-row harvester, AERMOD EFs were 1.8 times higher than ISCST3 EFs. This leads to the conclusion that EFs developed with dispersion models are model specific. These EFs should be used in conjunction with the same model with which they were developed. If used with a different model, the results would lead to incorrect estimates of downwind concentrations.