EPA's Proposal for MOBILE6 Facility-specific Speed and Non-FTP Correction Factors March 7, 1997 Abstract In MOBILE5, speed adjustments were made based on emission results from the Federal Test Procedure (FTP) and on a number of cycles with varying average speeds. For MOBILE6 we propose to adjust for differences in roadway (facility) type as well as speed and non-FTP effects. EPA has developed new facility-specific inventory cycles, based on real-world driving studies, to address these purposes. Background EPA's highway vehicle emission factor model, MOBILE, which is used for inventory modeling, has historically been based primarily on testing of the FTP certification cycle. Correction factors for various conditions (e.g., average speed, temperature, fuels) are applied to emissions measured at the FTP "standard" conditions. The speed correction factors are based on test results for vehicles tested on both the FTP driving cycle and several other cycles, each having a different average speed. "Real-World Driving" The Clean Air Act Amendments of 1990 mandated a closer look at "real-world driving" - that is, driving modes that are not covered by the FTP. The FTP is used for certification of all vehicles sold in the US. A new Supplemental FTP rule was finalized in October 1996. This rule specifies new certification cycles and associated standards with more aggressive driving. For the Federal Test Procedure Review Project, EPA collected both chase car data and instrumented vehicle data in Baltimore, MD, and Spokane, WA, which was supplemented by an instrumented vehicle study by Research Triangle Park (RTP) in Atlanta, GA, and a chase car study by California Air Resources Board (CARB) in Los Angeles, CA. The Federal Test Procedure Review Project intent was to develop driving cycles for certification purposes, therefore inventory issues were not taken into consideration. For certification purposes, worst case conditions were included in the new test procedure cycles. The new certification cycles are not intended to represent the variety of in-use fleet driving patterns. Transportation Models The speed correction factors currently contained in MOBILE were not designed to estimate emissions for individual segments of the roadway system. Available transportation models represent the roadway system as a network of "nodes," which are usually intersections, connected by "links," which represent a particular type of roadway or "facility." Since transportation models generate link-specific estimates of speed and traffic volume, local officials have been using MOBILE to generate link-specific emissions estimates. The driving patterns in the instrumented vehicle studies show that some types of facility-specific driving contain more frequent and more extreme acceleration and deceleration than others, which reach a similar speed but remain at a steady cruise. There is a need to quantify the emission differences for facility-specific speed related traffic control measures in inventory modeling. This approach requires facility-specific speed correction factors because, at a particular average speed, the pattern of vehicle operation is expected to be substantially different for different types of roadways. At an average speed of 35 mph, for example, travel over surface streets is likely to be dominated by cruising in the vicinity of the speed limit at a low level of traffic congestion, while travel on a freeway would involve a high congestion level and much less cruise operation. Facility-specific Inventory Cycles Facility-specific inventory cycles have recently been developed for use in the MOBILE model. These cycles will better represent actual fleet driving and will also include more aggressive "real-world driving". In a work assignment for EPA, Sierra Research developed eleven facility-specific and one non-freeway area-wide driving cycles. The cycles are based on chase car and instrumented vehicle data from Baltimore, Spokane and Los Angeles, which was collected during the FTP Revision Project for use in developing the supplemental certification cycles and new standards. Cycle Construction Methodology Sierra Research constructed the facility-specific cycles using randomly selected microtrips to match the speed-acceleration frequency distribution (SAFD) of all vehicle operation occurring under the conditions of interest (e.g., a particular facility type and congestion level). This approach ensures that the speed-time profile of the cycle is constructed from real speed-time profiles that reflect the proper proportion of a broad range of vehicle operation. Sierra did a separate assessment of the highest load points (i.e. the highest combined speed/acceleration points), to make sure the cycles have a representative sample of the high load points. A secondary criteria for developing the cycles was to match the total proportion of specific power values in two groupings: between 200-299 mph/sec (moderate high load points) and 300+ mph/sec (extremely high load points). Because microtrips begin and end at rest, a modification to this methodology was required to develop cycles representative of uncongested freeway operation. Sierra used appropriate trip segments (in lieu of microtrips) that were driven under the target levels of congestion on freeways. More thorough descriptions of the cycle development methodology can be found in the final report by Sierra Research, which will be available by April 1997. Testing There are two test programs in progress for generating modal emissions data from the 12 new cycles, plus the LA92 and the NYCC (a low speed cycle which has previously been used for speed correction factors in the MOBILE model). The expected samples are: 1. 50 vehicles by Automotive Testing Laboratories (ATL) in East Liberty, Ohio, under contract to EPA; 20" twin roll dynamometer. 2. Approximately 25 vehicles by National Vehicle and Fuels Emission Laboratory in Ann Arbor, Michigan; 48" dynamometer; tested both with air conditioning in operation (at 95 degrees Fahrenheit) and without (at 75 degrees Fahrenheit) . In addition to the modal tests on the 20"/48" dynamometers, an FTP on the Clayton 8.65" dynamometer is also being collected to relate the cycles to the EPA historical Emission Factor database of FTPs. Plans for Modeling in MOBILE6 We are proposing separate speed correction curves by facility-type. Specific curves will be generated for freeways and for arterial/collectors. There are only single cycles to represent local roadways and ramps. EPA proposes to apply the speed corrections to the temperature and fuel corrected base running emissions, which will be based on bag 3 of the FTP minus the hot start effect. Emissions due to engine start will be determined separately. The proposal is for MOBILE6 to output emission factors by facility-type. We are exploring ways to provide a weighted running emission factor that will include all roadway types. The methodology for weighting together the speed-corrected facility-type emissions into a single running emissions rate is under development. We plan to utilize the data samples from Los Angeles, Baltimore, and Spokane for this procedure. We will also utilize the LA92 and non-freeway area-wide cycles emissions data to validate the weighting procedure. Issues 1. The disaggregation of emission factor by facility-type will result in an emission factor most appropriate for short time periods, instead of the traditional average daily emission factor given in earlier versions of the MOBILE model. We are requesting input from MOBILE model users as to the implications of this. How much effort should be put into generating additional average daily emission factors in the MOBILE output? What are the purposes for these average daily emission factors? 2. The user input requirements would change significantly with these proposals. There would be no more need for operating mode inputs. Instead, the user would input average speed per %VMT combinations for freeway driving, arterial/collector driving, and %VMT for ramp driving (the remaining fraction would be considered local driving). The VMT fractions would be used to weight the facility-specific emission factors at each speed to an area-wide running emission factor which would be representative of a specific urban area. How many various conditions are sufficient for an area-wide estimate? How much effort should be put into area-wide estimates? 3. Another issue we are investigating is whether it will be possible to analyze the arterial/collector modal data to include signal density in the model. Detailed information on the arterial/collector cycle microtrip relationship to signals and types of signals has been documented by Sierra Research and it will be included in the final report describing the new cycles.