AAAR 34th Annual Conference
October 12 - October 16, 2015
Hyatt Regency
Minneapolis, Minnesota, USA
Abstract View
A Rigorous Examination of the Impact of Modern Gasoline Vehicles on the Use of CMB to Determine the Gasoline/Diesel Split
ANDREW HIX, Andrew May, Christopher Hennigan, Allen Robinson, Neil Donahue, Albert A. Presto, Carnegie Mellon University
Abstract Number: 341 Working Group: Source Apportionment
Abstract PM source apportionment results obtained with the chemical mass balance (CMB) model are highly sensitive to the choice of source profiles. Source profiles for gasoline and diesel vehicle emissions have often been constructed from relatively small vehicle fleets (N<10), and in some cases published source profiles represent the emissions of a single vehicle. Additionally, many of the vehicle sampling campaigns in the published literature are dated and do not include modern vehicles.
This study addresses the variability in gasoline vehicle fleet using bootstrap resampling to generate probability distributions of the population mean marker/OC ratios from a variety of sampling campaigns including the EPA’s Kansas City Study and a recent campaign performed in cooperation with the California Air Resources Board (CARB; May et al, Atmos. Environ., 2014). These probability distributions are inputs to a Monte Carlo implementation of CMB to determine the impact of gasoline source profile variability on CMB predictions. While the EPA’s Kansas City Study produced a larger data set (~270 vehicles), the CARB study contains the first large-scale measurements of molecular marker emissions from Tier 2/LEV-2 (e.g., model years 2003-2013) vehicles. Regardless, the gasoline apportionment results, calculated using both data sets, were statistically insignificant.
Historically, due to differences in engine design, elemental carbon was the key species for splitting gasoline and diesel vehicles; diesel engines produced EC/OC ratios significantly greater than 1, and gasoline engines produced EC/OC ratios significantly less than 1. However, the gasoline source profiles calculated using the CARB and Kansas City Campaigns yielded EC/OC ratios ~1, making them “diesel-like,” and causing the gasoline and diesel source profiles to become degenerate. Comparisons by model year suggest that these results are influenced by gasoline vehicles built after 1996. This study will rigorously test this observation.