There are many methods for preparing local data inputs for emission modeling. The decision to select a particular method may depend on the availability and quality of data, costs, resources, and individual preference. The objective of this task is to review the current state of the practice used by state agencies and metropolitan planning organizations (MPOs) to develop local data inputs for transportation conformity analyses and SIP development. The local data inputs of interest in this task are:
The list of areas whose SIP submission(s) was under the EPA adequacy review was obtained in December 2009 from the website http://www.epa.gov/otaq/stateresources/transconf/currsips.htm. Five areas were then selected from the list based on their size (in terms of population) and geographic location. The areas with very large (more than 10 million) and large (1-10 million) population were selected as they were likely to have more resources to invest in developing local data inputs for emission modeling. The selection was also made in a way that areas from the different parts of the country were represented. The final five areas selected for the state-of-the-practice review include:
In the review, the information was mainly gathered from the areas' SIP and transportation conformity documents. In some cases, personal communication with the areas' staff was also made to acquire more detailed information or clarification. The next section presents a summary of findings from the state-of-the-practice review.
The South Coast air basin in California consists of Orange County, and part of Los Angeles, Riverside, and San Bernardino Counties. The MPO for the region is the Southern California Association of Governments (SCAG). Since South Coast is in California, it uses the California Air Resource Board (CARB)'s EMFAC as the regulatory model in its transportation conformity analyses. In addition to developing and updating the EMFAC model, CARB also plays a significant role in developing many of the data inputs for running the model.
Vehicle population, age distribution, and fleet mix data for each county in California are developed and included in the EMFAC model by CARB. The primary source of these data is the vehicle registration database maintained by the California Department of Motor Vehicles (DMV). The data in the current version of the model (EMFAC 2007) are based on six years (2000 - 2005) of DMV data [California Air Resources Board, 2007]. CARB has committed to update the vehicle fleet data in EMFAC on a 3‐year cycle thereafter.
The estimate of VMT of HDTs is one of the outputs of SCAG's regional travel demand model (TDM), which contains a specific HDT module. SCAG's regional TDM follows a standard four-step modeling approach. The modeling methodologies, parameters, and inputs are periodically updated to reflect current travel conditions and demographic changes. The model is also subject to periodic peer reviews to insure that the model is valid and represents the current state of the practice for transportation modeling [Southern California Association of Governments, 2010].
The HDT module in SCAG's regional TDM consists of two major components: internal truck trip models and external truck trip models. The internal truck trips are generated using a cross-classification method by applying truck trip rates for a two-digit code by the North American Industry Classification System (NAICS) to the number of employees in that category and also the number of households within each zone. The daily truck trip ends are distributed using a gravity model to create daily truck trips for each of the three truck types: 1) light HDT, 2) medium HDT, and 3) heavy HDT. The external truck trips are developed using an econometric model to estimate inbound and outbound commodity flows by counties. The county to county commodity data is allocated to the zonal level based on NAICS employee distribution and then converted to truck trips using observed data collected during model development. Seaport and airport-related truck trips are included as special generator truck trips. The daily truck trips by truck types are allocated to four time periods and merged with the auto trips in trip assignment [Southern California Association of Governments, 2010].
Additionally, in order to maintain consistency of model results with VMT estimates from the Highway Performance Measurement System (HPMS), a set of base year HPMS VMT to model VMT ratios (factors) is developed for each subarea of county by air basin. Separate factors are generated for autos and trucks. These same factors are applied to final network assignments of each model run to yield final network flows and congestion [Southern California Association of Governments, 2010].
SCAG is currently in the process of enhancing the HDT module. The enhancements include an extensive travel survey, an updated external trip estimation methodology, and a more accurate representation of warehouse related trips [Southern California Association of Governments, 2010].
The number of idling hours of HDTs is also developed by CARB and included in the EMFAC model. It was developed from three instrumented studies of a total of 147 HDTs. Idle trips were identified from this combined dataset where an idle trip is a key-on to key-off event with a speed of less than 5 mph and a trip length of less than 5 miles.All other trips with speeds greater than or equal to 5 mph and trip length greater than or equal to 5 miles were considered non-idle trips. Then, the idling duration was calculated from these idle trips [California Air Resources Board, 2000].
The San Joaquin Valley Air Pollution Control District is made up of eight counties in California's Central Valley: San Joaquin, Stanislaus, Merced, Madera, Fresno, Kings, Tulare and the San Joaquin Valley Air Basin portion of Kern.
Vehicle population, age distribution, and fleet mix data for each county in California are developed and included in the EMFAC model by CARB. The primary source of these data is the vehicle registration database maintained by the California Department of Motor Vehicles (DMV). The data in the current version of the model (EMFAC 2007) are based on six years (2000 - 2005) of DMV data. CARB has committed to update the vehicle fleet data in EMFAC on a 3‐year cycle thereafter.
All MPOs in the San Joaquin Valley region use network- based travel models. However, none produces specific estimates of heavy-duty truck VMT. VMT estimates for HDTs are based on CARB calculations from DMV updates and HDT trip statistics derived from the instrumented HDT studies [California Air Resources Board, 2000].
The number of idling hours of HDTs is also developed by CARB and included in the EMFAC model. It was developed from three instrumented studies of a total of 147 HDTs. Idle trips were identified from this combined dataset where an idle trip is a key-on to key-off event with a speed of less than 5 mph and a trip length of less than 5 miles.All other trips with speeds greater than or equal to 5 mph and trip length greater than or equal to 5 miles were considered non-idle trips. Then, the idling duration was calculated from these idle trips [California Air Resources Board, 2000].
The New York Metro Transportation Council (NYMTC) is the MPO for New York City, Long Island and the lower Hudson Valley.
The vehicle age distributions used in MOBILE6 are obtained from the New York State Department of Motor Vehicles (NYSDMV) registration data for the current year at the beginning of each July. Each record is sorted into the 28 vehicle types by county. The 2002 registration distribution was used for 2002 inventories. Diesel fractions are obtained at the same time as the registration distributions.
The NYMTC uses an activity-based travel demand model in conjunction with a post-processing tool called PPSuite. The VMT outputs for all vehicle classes combined from the travel demand model are disaggregate into five vehicle classes based on the vehicle mix fraction data provided by the New York State Department of Transportation.
No information is available.
The Maricopa Association of Governments (MAG) is a Council of Governments (COG) that serves as the regional agency for the metropolitan Phoenix area.
Vehicle class and age distributions are based on vehicle registration data provided by the Motor Vehicle Division of the Arizona Department of Transportation (ADOT). The distributions are developed from only those vehicles registered in the county. The MAG internally develops a scheme to map the vehicle classes in the vehicle registration database to MOBILE6 vehicle classes. Then, a conversion tool provided by the EPA is used to convert MOBILE6 vehicle class fraction to MOVES source type fraction.
The MAG performs regional transportation modeling in TransCAD software. The VMT outputs for five vehicle classes from the transportation model are distributed to different vehicle classes based on local survey data.
The MAG develops local HDT population data as an input and uses the default calculation of extended idling hours of HDT supplied by MOVES.
The Denver Regional Council of Governments (DRCOG) is an MPO for the nine-county Denver region. The DRCOG provides vehicle activity results from travel demand model runs to the Colorado Air Pollution Control Division for air quality analyses.
Vehicle class and age distributions are based on vehicle registration database.
The DRCOG uses an activity-based travel demand model called Focus. The Focus travel model was estimated based on detailed data from multiple surveys, including a commercial vehicle survey that gathered complete travel information from more than 800 commercial vehicles on an assigned day. The VMT fraction by vehicle type including heavy-duty trucks is obtained from the permanent traffic counter data recorded by the Colorado Department of Transportation. This VMT fraction is used for both baseline and future year scenarios.
The default values in MOVES are used.
Based on the limited review of the current state of the practice conducted in this study, it is found that the data sources and methodology used to develop local data inputs for emission analyses vary from one area to another.
For vehicle class and age distributions, although all five areas reviewed in this study rely on the use of vehicle registration databases, the vehicle class definition used in the registration database for each state is different. Therefore, the responsible agencies have to develop a tool to map the vehicle classes in the registration database to MOBILE6 vehicle classes or MOVES source types. In addition, it is found that none of the five areas currently accounts for out-of-area vehicles in the development of vehicle class and age distributions.
For vehicle miles traveled of HDTs, although there are differences among the areas such as the choice of and the level of details in the travel demand models used, the general trend is that VMT outputs from travel demand models are distributed to different vehicle classes including HDTs based on some sorts of vehicle class fraction derived from surveys or traffic counters. Only the SCAG's travel demand model has a dedicated HDT modeling capability.
Lastly, for idling hours of HDTs, the two areas in California use the data derived from instrumented studies of a total of 147 HDTs in the state. For Maricopa County, AZ, and Denver metro, CO, the default data in MOVES for extended idling of HDTs are used.
