Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Office of Planning, Environment, & Realty (HEP)
PlanningEnvironmentReal Estate

HEP Events Guidance Publications Awards Contacts

Handbook for Estimating Transportation Greenhouse Gases for Integration into the Planning Process

Chapter 6 - Alternative GHG Estimation Approaches

PDF files can be viewed with the Acrobat® Reader®

This section reviews two other methodologies for estimating emissions: the first focuses on estimating freight emissions based on commodity flow data; the second is a statewide policy analysis tool developed by FHWA called the Energy and Emissions Reduction Policy Analysis Tool (EERPAT).

6.1 Commodity Flow Based Methods to Estimate Freight Truck Emissions

Table 22. Selection Criteria for Commodity-flow-based Methods


Selection Criteria

Commodity-flow Based Methods

Basic Approach (e.g., use commodity flow data )

Refined approach (e.g., estimate truck VMT)

Analysis Type

Inventory or Forecast

Inventory or Forecast

Geographic Scope

-State
-Some regions

-State
-Some regions

Analysis Precision

Limited - simple method relying on many assumptions

Moderate - depends on levels of refinement

Data Needed

Commodity flow estimates

Commodity flow estimates and truck survey data

Necessary Analytical Capabilities

Limited - only applying emissions factors

Requires some modeling skills

Level of Resources Required (i.e., staff/budget)

Limited - but depends on commodity flow data availability

Depends on existing capabilities - requires development of origin/destination truck trip table

Capable of Addressing Vehicle Technology/Fuels Changes

No

No

Capable of Addressing Changes in Travel Demand

Yes, to the extent accounted for in commodity flow data

Yes, to the extent accounted for in commodity flow data

Capable of Addressing Changes in Vehicle Speeds and Operations

No

No

Description

Emissions from freight truck transportation, as well as other freight modes, can be difficult to forecast, since they are heavily influenced by future economic conditions, affecting both, truck VMT and fleet turnover/technology. This is particularly true since U.S. economic conditions are difficult to forecast, especially over decades, as is needed for GHG analyses that often extend to 2050.

Moreover, freight truck emissions are often driven by factors that are external to a state or region. This is particularly true with pass-through traffic and internal-external trips (truck trips with one end outside of the state or region). Some states have developed statewide truck travel demand models. A small number of MPOs have developed travel demand forecasting models specific to freight trucks; many MPOs simply forecast truck traffic as a fraction of passenger vehicle VMT. While some travel demand models do estimate truck VMT, these estimates often do not adequately address some of the key factors that influence freight truck travel and emissions levels. The strengths and weaknesses of current approaches to modeling freight emissions are explained in greater detail in a 2010 National Cooperative Freight Research Program Report - "Representing Freight in Air Quality and Greenhouse Gas Models. "1

Commodity flow data provides an approach to estimate current and future freight movement, and can be used as a basis for GHG emissions estimates. An advantage of commodity flow data is that it can be linked to underlying economic drivers, expressed in employment data by industry, so forecasts will reflect expected economic changes. Commodity flow data can also be useful for examining shifts between freight modes (e.g., truck vs. rail).

Some states that do not have truck models have used commodity flow data to directly estimate emissions from freight trucks using a simplified approach. The following section provides a methodology for using commodity flow data to estimate emissions.

Strengths and Limitations

Table 23. Strengths and Limitations of Commodity Flow Based Models

Strengths

Limitations

  • Provide a simple means to estimate freight-related emissions, particularly where travel forecasting models for freight are lacking.
  • Commodity flow data and forecasts are linked to economic drivers.
  • While some data is available, there is overall a lack of data about freight traffic within and between regions, particularly with regard to "empty miles. "
  • These methods have limited ability to consider congestion effects on truck GHG, truck driver eco-driving programs, speed limits on trucks, logistics improvements that reduce truck GHG, and other variables.

Key Steps and Data Options

Step 1: Gather or develop commodity flow data. These data are usually expressed as tons of freight transported between origin region and destination region, by mode and commodity type. Data sources include:

Step 2: Calculate ton-miles by mode. Using commodity flow data and estimates of distance between origin and destination regions, calculate the ton-miles of commodity flow, by mode. No additional data sources required.

Step 3: Estimate emissions factors and calculate emissions. Simple ton-mile GHG emission factors can be obtained for freight modes. These factors are usually estimated based on national data. Multiplying the freight truck emission factor by the truck ton-miles produces a GHG estimate. The same approach can be applied for other freight modes.

Freight ton-mile emissions factors can be obtained from a variety of different sources. EPA's SmartWay Transport Partnership has estimated illustrative ton-mile emission factors for freight trucks. 2 EPA's Climate Leaders Program3 has also estimated ton-mile emission factors for medium and heavy-duty trucks. The World Resources Institute and the World Business Council for Sustainable Development GHG Protocol Initiative provides factors for use in preparing corporate GHG inventories.

Outputs: GHG emissions factors per ton-mile for freight modes, GHG emissions for freight modes.

Example: Massachusetts State Freight Plan analysis

Massachusetts DOT used a commodity flow approach for the Massachusetts State Freight Plan. Freight ton-miles were obtained from the Global Insight Transearch database to estimate freight ton-miles transported by truck in the state. Ton-mile emissions factors from EPA were then used to calculate emissions. Global Insight provided forecasts of freight traffic through 2035, allowing the state to develop a baseline forecast for emissions analysis. See http://www.massdot.state.ma.us/planning/Main/StatewidePlans/FreightPlan.aspx.

A more complex application of commodity flow data involves converting ton-mile data into truck trips, to estimate truck VMT. This approach has been used by some State DOTs for statewide freight analysis, as well as by a few MPOs to analyze external truck trips (those with one trip end within the metro area). In this approach, estimates of truck average payload by commodity type are used to convert commodity flow data to truck trips. Average payload factors have been estimated by researchers using the Census' Vehicle Inventory and Use Survey (VIUS) and other sources, and are available from FHWA. FHWA has recently updated VIUS estimates of average payloads for trucks by commodity. 4 Additional truck trips may need to be estimated to account for empty truck trips. Once the state or MPO has an origin/destination (O/D) table of truck trips, the trips can be assigned to the roadway network as part of the travel modeling process. Truck VMT and emissions can then be estimated using the approaches discussed in Section 5 above.

6.2 Energy and Emissions Reduction Policy Analysis Tool (EERPAT)

Table 24. Selection Criteria for EERPAT

Selection Criteria

EERPAT

Analysis Type

Scenario/strategy analysis

Geographic Scope

State

Analysis Precision

Screening level analysis - not suitable for project or detailed plan-level analysis

Data Needed

Extensive demographic, land use, strategy-related data required as inputs.

Necessary Analytical Capabilities

Tool is readily available but requires understanding of data inputs

Level of Resources Required (i.e., staff/budget)

May be significant depending on data collection required

Capable of Addressing Vehicle Technology/Fuels Changes

Yes

Capable of Addressing Changes in Travel Demand

Yes.

Capable of Addressing Changes in Vehicle Speeds and Operations

Limited to some congestion impacts

Description

The Energy and Emissions Reduction Policy Analysis Tool (EERPAT), which is based on Oregon's GreenSTEP Model, is designed specifically for GHG analysis. The EERPAT is a statewide policy analysis tool for providing rapid analysis of many scenarios that combine effects of various policy and transportation system changes, including those that are often difficult to analyze using traditional transportation system analysis tools. 5 EERPAT is sensitive to a large number of factors such as land use, transportation demand, vehicle technology, fuels, price and other inputs. The model is an open source tool and is designed to be adapted and used by other states.

Strengths and Limitations

Table 25. Strengths and Limitations of EERPAT

Strengths

Limitations

  • EERPAT provides policy sensitivity for different GHG mitigation measures, including carbon taxes, technology solutions, transit, and demand management.
  • It can evaluate future changes in land use and it is sensitive to external changes in the price of fuel, as well as other pricing strategies.
  • EERPAT can incorporate changes in tailpipe emissions associated with changes in technology such as increased use of electric vehicles or plug-in hybrids.
  • The model can be used to assess the overlapping effects of bundles of GHG mitigation strategies.
  • VMT estimates are attributed to the regions where the households are located instead of where the travel occurs. The model does not include trips originating outside of the state.
  • There are a large number of model inputs and some may be difficult to obtain. For example:
    • Battery range of electric vehicles,
    • Percentage of workers paying for parking
    • Percentage of employers with strong employer-based programs and percentage of households subject to strong TDM programs.

Key Steps and Data Options

Step 1: Collect demographic data to generate synthetic households. The model allows the user to generate a set of synthetic households for each forecast year that represents the likely household composition for each county, given the county-level forecast of persons by age. Each household is described in terms of the number of persons in each of six age categories residing in the household. A total household income is assigned to each household, given the ages of persons in the household and the average per capita income of the region where the household resides. Sources for this type of data include:

EERPAT model inputs are shown below.

Table 26. EERPAT Model Inputs

Input Data and Assumptions

Description

Demographics

County population projection by age cohort
State average per capita income growth
Statewide population projection

Land Use

Urban growth boundary expansion rates
Growth proportions in metropolitan, other urban and rural areas
Urban mixed use assumptions

Transportation Characteristics

Rate of transit revenue mile growth
Rate of freeway & arterial lane mile growth

Mitigation Strategies

Households affected by travel demand management, vehicle operations and maintenance strategies
Car sharing deployment assumptions
TDM travel reduction assumptions

Vehicle Fleet

Light weight vehicle ownership and use assumptions
Vehicle type percentages
Average fleet MPG by type and model year
Electric Vehicle (EV) & Plug-in Hybrid Electric Vehicle (PHEV) travel range, market penetration
Vehicle use optimization

Cost

Travel, parking, carbon, VMT, etc.

Fuel

Fuel type, carbon lifecycle, emissions per Kilowatt of electricity

Other

Incident reduction assumptions, truck deadhead percentage

Step 2: Collect input data to apply land use and transportation system characteristics. EERPAT includes models to estimate density and land use characteristics at a Census tract level based on more aggregate policy assumptions about metropolitan and other urban area characteristics. Each household is assigned to a metropolitan, other urban, or rural development type in the county where it is located based on policy assumptions about the proportions of population growth that will occur in each type. The number of lane miles of freeways and arterials is computed for each metropolitan area based on base-year inventories and policy inputs as to how rapidly lane miles are added relative to the addition of metropolitan population. In addition, growth in transit revenue miles is also input, including the revenue mile split between electrified rail and buses.

Step 3: Collect data on mitigation strategy assumptions, vehicle fleets, costs, and other inputs. The model assigns each household as being a participant or not in a number of travel demand management programs (e.g. employee commute options programs, individualized marketing) and/or to vehicle operations and maintenance programs (e.g. eco-driving, low rolling resistance tires) based on policy assumptions about the degree of deployment of those programs and household characteristics. Input assumptions about the market penetration of plug-in hybrid electric vehicles (PHEVs) and probability models are used to determine future shares of PHEVs and EVs based on input assumptions about the range of these vehicles. Total variable costs are determined for vehicle travel based on fuel economy, electric power consumption and policy variables (carbon taxes, parking fees, etc). Data sources for these inputs include:

Step 4: Calculate fuel consumption and estimate GHG emissions. The model estimates vehicle usage and vehicle fuel economy based on the travel behavior of the synthetic households. Each household is assigned the number of vehicles it is likely to own based on the number of persons of driving age in the household, the income of the household, the supply of transit and freeways and whether the household is located in an urban or mix-use area. This behavior is sensitive to a range of factors, including the price of fuel, the range of electric vehicles, the cost of parking, the impacts of congestion on fuel economy, the availability of other modes, etc. The model incorporates the overlapping effects of multiple policy strategies and considers how household budgets would respond to transportation costs.

Output: GHG emissions based on travel behavior and vehicle technologies.

Example: Oregon GHG scenario analyses

The model was developed first in Oregon as the GreenSTEP Model. GreenSTEP is currently being used to test various scenarios for the Oregon Statewide Transportation Strategy (STS) for reducing transportation sector greenhouse gas emissions. In the first round of modeling, a total of 144 scenarios were modeled. Policies were organized into six general categories:

Based on the GreenSTEP model, EERPAT was developed for application to other states. Its use was piloted by FHWA in Florida and documentation for the model is available on the FHWA website at: http://www.planning.dot.gov/FHWA_tool/.


1 National Cooperative Freight Research Program Report 4, "Representing Freight in Air Quality and Greenhouse Gas Models," prepared by Browning, L. et al. , 2010, http://onlinepubs.trb.org/onlinepubs/ncfrp/ncfrp_rpt_004.pdf, p. 8.

2 http://www.epa.gov/smartway/documents/partnership/shipper/partnership/420b12005.pdf.

3 http://www.epa.gov/climateleadership/documents/resources/commute_travel_product.pdf.

4 FHWA, Office of Freight Management and Operations. "Estimation of 2007 VIUS Variable. " October 15, 2009.

5 Available online at: http://www.planning.dot.gov/FHWA_tool/default.asp.

Updated: 03/27/2014
HEP Home Planning Environment Real Estate
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000