Regulation of Fuels and Fuel Additives: Renewable Fuel Standard ...

in terms of the distillation temperatures, due to the relatively low boiling point of ethanol.
Other fuel parameters are expected to be very similar. For this analysis we have assumed
no changes to fuel parameters other than ethanol and MTBE content for RFG.
b. Emissions from Motor Vehicles
We use the EPA Predictive Models to estimate the impact of gasoline fuel quality
on exhaust VOC and NOx emissions from motor vehicles. These models were developed
in 2000, in support of EPA’s response to California’s request for a waiver of the RFG
oxygen mandate. These models represent a significant update of the EPA Complex
Model. However, they are still based on emission data from Tier 0 vehicles (roughly
equivalent to 1990 model year vehicles). We based our estimates of the impact of fuel
quality on CO emissions on the EPA MOBILE6.2 model. We base our estimates of the
impact of fuel quality on exhaust toxic emissions (benzene, formaldehyde, acetaldehyde,
and 1,3-butadiene) primarily on the MOBILE6.2 model, updated to reflect the effect of
fuel quality on exhaust VOC emissions per the EPA Predictive Models. Very limited
data are available on the effect of gasoline quality on PM emissions. Therefore, the
effect of increased ethanol use on PM emissions can only be qualitatively discussed.
In responding to California’s request for a waiver of the RFG oxygen mandate in
2000, we found that both very limited and conflicting data were available on the effect of
fuel quality on exhaust emissions from Tier 1 and later vehicles. 82 Thus, we assumed at
the time that changes to gasoline quality would not affect VOC, CO and NOx exhaust
emissions from these vehicles. Very little additional data has been collected since that
time on which to modify this assumption. Consequently, for our primary analysis for
today’s proposal we have maintained the assumption that changes to gasoline do not
affect exhaust emissions from Tier 1 and later technology vehicles.
There is one recent study by the Coordinating Research Council (CRC) which
assessed the impact of ethanol and two other fuel properties on emissions from twelve
2000-2004 model year vehicles (CRC study E-67). The results of this program indicate
that emissions from these late model year vehicles may be at least as sensitive to changes
to these three fuel properties as Tier 0 vehicles on a percentage basis. 83 However,
because this study is the first of its kind and not all relevant fuel properties have yet been
studied, in our primary analysis we continue to assume that exhaust emissions from Tier
1 and later vehicles are not sensitive to fuel quality. Based on the indications of the CRC
E-67 study, we also conducted a sensitivity analysis where the exhaust VOC and NOx
emission impacts for all vehicles were assumed to be as sensitive to fuel quality as Tier 0
vehicles (i.e., as indicated by the EPA Predictive Models).
82 The one exception was the impact of sulfur on emissions from these later vehicles, which is not an issue
here due to the fact that renewable fuel use is not expected to change sulfur levels significantly.
83 The VOC and NOx emissions from the 2000-2004 model year vehicles are an order of magnitude lower
than those from the Tier 0 vehicles used to develop the EPA Complex and Predictive Models. Thus, a
similar impact of a fuel parameter in terms of percentage means a much smaller impact in terms of absolute
emissions.
- 156 -