 |
NARSTO Workshop on Innovative Methods for Emission Inventory Development and Evaluation University of Texas, Austin October 14-17, 2003 |
 |
Poster 8: Fuel-Based Mobile Source Emission Inventory Methods
Donald H. Stedman and Gary A. Bishop
Chemistry Dept., University of Denver, CO 80208
A single on-road remote sensor can be used to provide both public information (a SMART SIGN can be used to motivate voluntary emissions related repairs) and a mobile-source emission inventory. The public information aspect has been quantified and published. The process by means of which one can create a fuel-based mobile source emission inventory for CO, HC and NOx has been carried out in Denver and compared to the MOBILE model. Mobile modeling for the Denver metropolitan area predicts 794 metric tons per day of CO in 2001 reducing to 710 by 2006. We measured 582 in 2000. This is a smaller emission, and has a slower observed rate of decrease between 1996 and 2000 than the model predicts. Fuel-based emission inventory has quantifiable uncertainty. Emission measurements at several locations chosen for high and low socioeconomic factors span the range of possibilities for the urban area. The other source of potential uncertainty is the fuel use. State total data for fuel sales are uniformly available and generally of high quality. Apportionment of the state total to a given urban area or district is more difficult. Two methods which have been investigated are apportioning by vehicle registration and apportioning by population. It turns out that these two methods agree very closely and are probably not a major source of added error. Mobile source, fuel based inventories, with estimated uncertainties, for various cities in Texas will be presented based upon on road emissions data.