Estimating Emissions from Sources of Air Pollution

6.2 Estimating Emissions from On-Road Mobile Sources

6.2.4 Widely Used Vehicle Emission Models
6.2.4.1 Introduction

There are many on-road vehicle emission models developed for various purposes, from analyzing exposure near intersections or on the roadway, to understanding emission rates in a particular region. Some of the most widely used models for inventory purposes are discussed below. These are: the MOBILE6 model developed by the USEPA, the MOVES model developed by the USEPA, the EMFAC model developed by the California Air Resources Board, the German Handbook of Emission Factors and TREMOD model, the COPERT model developed by the European Environment Agency, the Comprehensive Modal Emissions Model (CMEM) developed by the University of California at Riverside, and the International Vehicle Emissions (IVE) model developed by ISSRC and the University of California at Riverside. The IVE model will be discussed in a section by itself. The other models will be discussed in this section.

6.2.4.2 The US EPA Mobile Model

The purpose of the MOBILE model is to estimate the emission rates from on-road motor vehicles operating in the 49 states such as passenger vehicles, motorcycles, trucks, buses, and motorhomes. The state of California in the United States has separate emissions standards and therefore developed a separate emissions model. The vehicle fleet use in the MOBILE model are primarilly diesel and gasoline fueled, with a very small fraction of natural gas and electric vehicles that are also incorporated into the model. The USEPA developed the first MOBILE model in 1978 to replace the AP-42 emission factor table that was used at the time, for a quicker and easier method for performing computations. This first version only predicted emissions from three pollutants. Since then, six major iterations of the model have taken place, resulting in the latest release in 2002 of MOBILE6. MOBILE6.2 is the most current version and is recommended for all new modeling efforts in the US States other than California. The outputs of the MOBILE model are emissions per unit time or distance of a fleet or vehicle type (i.e. grams/mile or grams/hour) of HC, CO, NOx, CO2, PM, NH3, SO2 and six toxic air contaminants such as lead. MOBILE estimates emissions of both exhaust and evaporative emissions, and particulate emissions from brake and tire wear. One of the differences between the MOBILE model and some of the other models, such as the EMFAC model, is that MOBILE does not apply the vehicle operation such as distance traveled and number of starts. The user will need to apply this separately. Once the distances are obtained, however, the emission rates can be multiplied by the total driving to estimate an overall inventory. One of the abilities of the MOBILE model and most emissions models is that it is designed to be able to predict emission rates from a future fleet to understand how emissions will change over time as vehicles age, overall travel increases, congestion increases, and new vehicles with lower emissions standards enter the fleet. This makes these models a very critical tool for understanding the impacts of potential policy changes to fuel, maintenance programs, growth in the region, and regulation of new vehicles.

The MOBILE model can be downloaded for free from the EPA’s website at http://www.epa.gov/oms/m6.htm. The EPA also conducts MOBILE6 training seminars and provides user guides as well as policy guidance documents for developing State Implementation Plans those can also be accessed from the website. However, this model is geared toward the US vehicle and policy situation, and is most likely not very applicable to any other locations. The MOBILE model was not designed to be easily modified and significant changes to some of the parameters would be required to apply it to a different country. For example, MOBILE uses model years of US vehicles for input. A model year will include a mixture of various types of technologies, and could not easily be decoded into a distribution of another country. Furthermore, MOBILE is a FORTRAN based software that requires modifications to the source code of the model to make many of the changes necessary for areas outside of the US. However, before the availability of more international models, such as the IVE model, there have been instances, such as in Mexico, where the effort was made to modify the MOBILE model for its own region. However, this was an extensive exercise and is now based on an outdated version of the MOBILE model.

6.2.4.3 The California EMFAC Model

The California Air Resources Board developed its own vehicular emissions model that performs a similar function as the MOBILE model, but predicts emissions from on-road vehicles operating specifically in California. The development of this separate model is for several reasons, but one obvious reason is that California has different emission standards from the rest of U.S. and other unique vehicle conditions, such as a relatively large influx of out –of –country vehicles entering the state. The EMFAC model was developed more recently than the MOBILE model, about 10 years ago, and has undergone five major revisions since then. EMFAC2007 is the most current version of the model and should be used for all estimates of on-road vehicular emissions in California. The EMFAC model estimates both the emission rate and the overall emissions from any area in California because it has built into it information about the activity of the vehicles located in the state, such as how much each vehicle drives on a daily basis and how many times it is started. These parameters were developed in coordination with the local transportation authorities and extensive studies on the amount of travel on roadways, the speeds on roadways, the number of starts per day, and the deterioration and emission rates of vehicles operating in California. Even with the vast amount of data and research that has taken place to develop this current version, new information that substantially affects the outcomes of the emissions predicted from the model are being discovered each year. For example, the heavy duty truck emissions and truck activity used in EMFAC is now being updated because the previous estimates did not explicitly account for differences in age and travel of the trucks that come from out of state. This point is made to illustrate that even the most complex and state of the art emissions models are not perfect and can have significant errors in representing the actual emissions that are occurring. The result is that the models are constantly undergoing revisions to improve the overall estimates.

The EMFAC model can be downloaded for free from the ARB website: http://www.arb.ca.gov/msei/onroad/latest_version.htm. The website also contains many reports documenting the technical studies used to gather the data and the user’s guide on how to operate the model. Again, the EMFAC model is designed to be very site specific and includes information only on the counties located in California. For example, if a user chose to model San Diego County, automatically information on the ambient temperature, humidity, types of vehicles, fuels, and amount of travel will be loaded into the model. Unlike MOBILE, the model platform is Java-based and has a relatively user-friendly method for modifying most of the input parameters such as the vehicle speeds, travel, and temperatures. However, still like MOBILE, ARB depends on model year based classifications that make extrapolation to areas outside of the US difficult. However, with very few other options out there, and such a state of the art model existing in California, there has been at least one organization that has taken on the task to modify EMFAC to reflect site specific information elsewhere. Hong Kong has done this and now uses a modified EMFAC model for its emissions estimations, which can be found at http://www.epd.gov.hk/epd/english/environmentinhk/air/guide_ref/air. The modified version includes different emission rates and different technologies of vehicles, as well as local conditions of Hong Kong. The user should specify the fleet mix and the driving parameters and should not rely on the default data in the model.

6.2.4.4 The European COPERT Model

The European Environmental Agency (EEA) has developed an on–road motor vehicle emissions model that is specific for the countries that make up the European Union. COPERT stands for “COmputer Program to calculate Emissions from Road Transport” and is similar to the other models in that it estimates emissions from on-road vehicles for the purposes of inventory development and projection. The first version of COPERT was released in 1989 and has subsequently been revised 3 times. COPERT4 is the most current version and has been revised based on new measurement data in Europe. COPERT estimates emissions of CO, NOX, HC, PM, VOCs, CH4, CO2, NH3, SO2, N2O and heavy metals and other toxics from vehicle activities. It classifies vehicles into various size and age groups as well as categories for highway, urban and rural driving situations. The light duty diesel and moped and motorcycle categories in the COPERT model may have improved information compared with the US models, since there are more of these types of vehicles on the road in Europe. The COPERT model can be downloaded for free from http://lat.eng.auth.gr/copert/.

6.2.4.5 The German Handbook of Emission Factors/TREMOD

The Handbook Emission Factors for Road Transport (HBEFA) provides emission factors for all German, Swiss, and Austrian on-road vehicle categories based on emissions testing and extensive vehicle operation profiles. The Handbook was first released in the 1990s, but the latest version, 2.1, was released in February 2004. The Handbook is based in Microsoft Access and allows users to select a fleet profile, congestion situations, and then calculate overall emissions rates. The handbook can be found at http://www.hbefa.net/. The Handbook is similar to other models in that it contains new measurement data to support both the emission factor development and the driving behavior in the region. It also accounts for start-up and evaporative emissions. The most recent version of the Handbook incorporates sophisticated tools such as incorporation of road grade into the emissions effects, and a wide range of cycles that represent driving on various road and congestion types. The Handbook is used as input into the TREMOD model, the German Traffic Emissions Estimation Model, which calculates overall emissions from on and off-road operations as well as overall fuel use.

6.2.4.6 The US EPA MOVES Model

In addition to the continual update of the MOBILE model, the US EPA’s Office of Transportation and Air Quality has been designing an eventual replacement for the MOBILE model, which they call the MOtor Vehicle Emissions Simulator, or MOVES. The MOVES model takes into consideration recent advances in on-road measurement technology (as opposed to using solely dynamometer data) as well as many of the recommendations in the National Research Council’s Mobile source modeling committee for improving on the shortcomings of the MOBILE model (National Research Council, 2000). This new system is designed to be more flexible than the MOBILE model and to be able to effectively change scales from a very fine-scale analysis to national inventory estimation. This has been a continual problem with the models originally designed to perform regional calculations. The MOVES model is being populated with very fine-scale information, such as second by second resolution emissions and driving behavior that can now be collected with on-board instrumentation. The new system is not a single piece of software, but is instead a toolbox of algorithms, data, and guidance necessary for use in all official US analyses associated with regulatory development, compliance with statutory requirements, and national/regional inventory projections. In addition to the multi-scale resolution of the model, the other state-of-the art feature of the MOVES model will take advantage of more dynamic real-world approach to applying driving patterns as compared to the ‘speed’ correction factor using a defined set of driving cycles.

6.2.4.7 The CMEM Model

The Comprehensive Modal Emissions Model, or CMEM, was developed at the University of California, Riverside as a fine-scale emissions predictions model, that was specially designed to improve the prediction of the variation of the vehicle’s operating conditions (i.e. the various driving ‘modes’, such as idle, steady-state cruise, various levels of acceleration/deceleration, etc). The latest version is CMEM2.0, released in 2000. More than the other models described above, this model was designed more as a research model, and was originally designed only for passenger vehicles operating in California but has now been expanded to include heavy duty diesel vehicles. The model was one of the first to use a very different approach to understanding the emissions variations, which is now becoming a useful component in models such as the MOVES and IVE. This is the change from using a driving cycle with a set average velocity and ‘speed correction factors’ to a more fundament understanding of the variability of fuel consumption and power to predict emissions. Another unique feature of the CMEM model is its special attention to modeling separately various types of malfunctioning vehicles, such as an engine running rich, or running with deteriorated catalyst. The model is now complete and capable of predicting second-by-second tailpipe emissions and fuel consumption for a wide range of vehicle/technology categories. The CMEM model can be ordered for $20US from http://pah.cert.ucr.edu/cmem/. A drawback of the CMEM model for macroscale applications or areas without a lot of resources is the extensive amount of input data required, such as knowing the frontal area of a vehicle, which is required for proper calibration. However, the physical-based concept of the model is useful and is applied in a more moderate fashion in the IVE model described in the next section.