An alternative methodology for collection of local vehicle start information for use in MOBILE6

Where would I be today without the endless concern and support from my loving family and sincere friends? Through many years of school, these people have given to me more time and energy than I could ever quantify. I wish to first and foremost thank them all for always listening to me, caring about my welfare and progress, and offering me the support and insight that I needed to be as successful as I am today.At the University, I am grateful and fortunate to have been placed under the direction of Jerry Everett, who trusted me with the space and flexibility to accomplish my responsibilities as a Graduate Research Assistant, as well as a student attending class and in pursuit of completing a thesis. Certain wording from him is reprinted in this thesis with his permission. His availability was only matched by that of Dr. FrederickWegmann, professor and advisor, who enthusiastically guided me along the path to successful completion of the graduate program and my thesis. He signed me up lastAugust, and with hope, in a few weeks, he will sign me out.The challenge of data collection was substantially made easier by the assistance ofLoma Wren, Carine Penne, Scott Kiser, and Phil Montgomery. Sandra Rodriguez atNuStats was helpful in sending us daily updates of study participants. Mike Newmanwas patient and energetically involved in helping me assimilate and analyze the data.David Brzezinski was a valuable inside source for information from the EPA.Also, I would like to thank Dr. Arun Chatteijee for his care and guidance, and Dr.Bruce Torn from the Department of Urban and Regional Planning, for his interest and support in my interdepartmental studies.MOBILE is the regulatory mobile source emission factor model that the U.S.Environmental Protection Agency (EPA) requires all states except California to use in developing emission inventories and for performing conformity analyses. EPA currently plans to release MOB1LE6, the latest version of the model, in 2001. MOBILE6 will be dramatically different from previous versions of the model and will allow users the option of providing local input data for several new and revised variables.EPA has decided to include a mechanism for producing separate start emission rates in MOB1LE6 by hour of the day and vehicle class. The default data included in the model are based primarily on start data collected from a series of instrumented vehicle studies in two metropolitan areas. However, EPA explicitly warns of the limitations such"national averages" have on effectively representing individual metropolitan areas.Moreover, the resources utilized in the methodology to acquire the default data may not be readily available or economically feasible for many metropolitan planning organizations to implement. This paper outlines a recently implemented alternative procedure for collecting vehicle start profile data, and offers a preliminary comparison of these profiles to the default values proposed for M0B1LE6.Data were collected and profiles representing vehicle start characteristics inKnoxville, Tennessee were developed. The methodology of the data collection process varied considerably from EPA's practices yet produced comparable results. The profiles developed were compared against the EPA default data to determine any significant differences. Preliminary analysis of the data show that the Knoxville profiles forAverage Starts per Day per Vehicle were somewhat lower than the default "national average" values included in MOBILE6. While the profiles for Daily Distribution ofVehicle Starts, Soak Distribution, and Hot Soak Activity followed the general pattern ofM0BILE6 default data, some differences were found that support the importance of data collection for individual MPO's.