Grantee Research Project Results
2008 Progress Report: A Long Term Integrated Framework Linking Urban Development, Demographic Trends and Technology Changes to Stationary and Mobile Source Emissions
EPA Grant Number: R831841Title: A Long Term Integrated Framework Linking Urban Development, Demographic Trends and Technology Changes to Stationary and Mobile Source Emissions
Investigators: Anas, Alex , Hewings, Geoffrey
Institution: The State University of New York at Buffalo , University of Illinois Urbana-Champaign
EPA Project Officer: Chung, Serena
Project Period: November 1, 2004 through October 31, 2007 (Extended to October 31, 2010)
Project Period Covered by this Report: November 1, 2007 through October 31,2008
Project Amount: $675,000
RFA: Regional Development, Population Trend, and Technology Change Impacts on Future Air Pollution Emissions (2004) RFA Text | Recipients Lists
Research Category: Climate Change , Air
Objective:
The project is developing an integrated modeling package, a tool for testing trends and policies that together determine future emission levels. The precise aims of the model integration effort targeted by the project are as follows:
The researchers seek a tool capable of producing long-term (25-50 year) projections of stationary- and mobile-source emissions in a metropolitan area. The way emission models are used is at best accurate for short horizons, taking as given local projections of local economic activity and population change. Instead of simply extrapolating these local trends, the researchers seek to explain them by modeling the fundamental behavioral relationships among individuals and firms and by linking these underlying economic relationships to secular national and international trends in population, economic development and technological changes relevant to emissions. Among the specific demographic trends to be examined are the graying of the population, reductions in household size, and international immigration. The researchers also will examine the possibility of the continued deindustrialization of U.S. manufacturing and its impact on a metropolitan area with considerable manufacturing using the Chicago MSA as an example. New technologies likely to impact emissions, such as electric vehicles and hydrogen fuel-cell vehicles, and electricity/energy production with a higher renewable fuel mix under higher sustained energy prices will be considered.
The proposed modeling framework is a dynamic computable general equilibrium model (DCGEM) for a metropolitan area based on principles of economic behavior that can be calibrated well from existing data. The sub-models are: 1) the Regional Economy and Land Use (RELU) model developed at the University at Buffalo, that treats the location, production, and transportation decisions of both businesses and consumers and the land development decisions of developers under a variety of government taxes and subsidies; 2) TRAN, a model of highway-transit modal choice and congested travel conditions on highway networks developed at the University at Buffalo; 3) a model of the energy/electricity sector and its relation to demand from industrial production, residential use, and vehicular transportation to be developed in this project; and 4) models of emissions from mobile and stationary sources including those from production outside of the electricity sector. The researchers will interface these models to establish feedbacks among them consistently with microeconomic theory to allow simulation of the workings of the Chicago regional economy.
The research will provide a tool that is portable from one metropolitan area to another. It will help identify the policies or trends that are the most effective in reducing emissions, and specify how quickly emissions reductions can be achieved and at what cost. It will determine whether the benefits of two policies or planning actions are sub-additive or super-additive. Hence, it could help direct future research and provide guidance to future policy initiatives that are economically and politically viable. Application of the model to other regions in the future would increase understanding of whether the policy actions should vary among metropolitan regions or whether the same policies are effective everywhere.
Progress Summary:
The accomplishments are reported for the previous report period and then for the current report period.
The project’s accomplishments prior to the period of this report included:
RELU-TRAN: Development of the TRAN model and integration of the RELU model with the TRAN model for a representation of the Chicago MSA consisting of 15 geographic zones spanning the city, the suburbs, and the exurban area, four industries (agriculture, manufacturing, business services, and retail trade), four consumer income quartiles, four building types (single-family residential, multiple-family residential, commercial, and industrial), undeveloped land, and an aggregated representation of the highway network. The total number of equations solved by RELU-TRAN is 656. The model is capable of predicting the effects of changes in transportation capacities, land use policies, a variety of taxes as well as in exogenous trends of population increase, export expansion and improved industrial productivity on the regional economy, transportation, and land use changes at the metropolitan level.
As explained in the “Objectives of Research” section above, a model such as RELU-TRAN that simulates the working of the regional economy, its land use, and travel is the centerpiece in an effort to correctly forecast emissions in urban areas. Because emissions are a byproduct of economic activity, they cannot be forecast well unless the underlying economic activity is forecast properly. As RELU-TRAN is a model that relies on the state-of-the-art theory in the urban economics and transportation fields, it plays a key role in the assembly of the models that the project aims to develop.
There are many examples of practical applications of the model assembly that is being developed. One is the use of aggressive land development restrictions to limit urban sprawl and the expansion of low-density suburban areas. Such policies would, in the long run, result in higher density developments in the built-up areas, resulting also in fewer vehicle miles traveled, more traffic congestion, and more transit use. This, in turn, would alter energy/electricity consumption by residences and by vehicles (depending also on the types of vehicles being used as the vehicle fleet changes via car ownership decisions). Emissions, in turn, also will be affected. The assembly of models would trace all of these effects, but it also would be able to quantify other costs and benefits of the aggressive land use policies, such as higher rents and other price distortions in the markets for labor and urban products.
New accomplishments in the period 11/1/2007 through 10/31/2008 are as follows:
RELU-TRAN 1 (Simulations for Chicago): In the report period, the researchers completed extensive simulations with RELU-TRAN 1. Using this version, three types of simulations were run:
Cost of congestion delays: The purpose of these simulations was to quantify the economic cost of delays caused in all personal travel in Chicago by the traffic congestion in the period 2000 through 2030. A well-known method used by the Texas Transportation Institute relies on extremely simple assumptions to quantify the cost of congestion because a lot of the needed information is not available. RELU-TRAN 1 generates much of the needed information. For example, the value of time to be used for a traveler in the TTI method is a constant number that TTI uses for the entire United States, whereas RELU-TRAN 1 generates within the model values of time for different consumers depending on the particular simulation and how it affects the urban economy.
Congestion and sprawl: Using the regional population growth forecasts for the Chicago MSA (2000-2030), the researchers showed that as the region grows, urban sprawl in the suburbs will continue. The projected population growth and the resulting sprawl increases congestion per mile of travel on average as expected and this can be calculated with and without highway capacity additions. RELU-TRAN, however, captures the simultaneous suburbanization of jobs and of residences. As a result, the results obtained show what happens to total travel times per traveler when both jobs and residences suburbanize.
Congestion pricing policies: The researchers simulated the effects of congestion pricing in the period from 2000 to 2030, using two sets of policies. One of these is a cordon around the Chicago CBD, the other a Pigouvian toll levied on all major roads in the region. The costs of congestion reductions that would be accomplished by these policies as well as the revenues raised by these policies and the effects they would have on consumer relocations and, in particular, on whether they contribute to consumers returning to central cities in significant numbers, were calculated.
All of these simulations will be written up in the form of two papers for publication in economics or environmental economics journals. A tentative preview report of these results will be presented on October 27, 2008, at the project review meeting sponsored by EPA in Research Triangle Park, North Carolina.
RELU-TRAN 2: In the report period, the project team focused on the development of RELU-TRAN 2. This version extends the model by adding a complete automobile sector in personal travel. Consumers in the model are extended so that they incur acquisition and maintenance costs for vehicles of a particular fuel efficiency, and gasoline expenditures as they operate their chosen vehicles. Meanwhile all of the other features of RELU-TRAN 1 (see above) remain in RELU-TRAN 2.
RELU-TRAN 2 was developed entirely during the report period. Initially, the researchers attempted to treat the choice of fuel efficiency as a continuous variable. Although considerable time and effort were spent on this approach, the researchers were not successful because two things became apparent: a) The researchers ran into technical problems that could not be overcome in the solution of the simultaneous equations that treated the choice of automobile efficiencies. b) The researchers also learned that when these equations could be solved, the results were not defensible as the model generated large changes in the vehicle fuel efficiencies of some consumers. As a result, the researchers shifted strategy and started implementing an alternative treatment of vehicle fuel efficiency as a discrete variable. This approach, although less elegant and more time consuming on the computer, did prove successful, and development of RELU-TRAN 2 was completed.
The new version predicts vehicle efficiency choice as a function of a consumer’s location and housing choices, the consumer’s income, etc. so that a fully microeconomic explanation of vehicle ownership is achieved. Random effects among consumers of the same type are also treated. Electric vehicles and hybrids with different purchase costs and fuel consumption characteristics can be entered into the model. The new version generates the following outputs that were not present in RELU-TRAN 1 (unless otherwise indicated):
- Vehicle miles traveled for work and non-work trips by consumer type and vehicle efficiency;
- Travel time for work and non-work trips by consumer type and vehicle efficiency;
- Aggregate fuel consumption in work and non-work trips by consumer type, vehicle type, and location of consumers;
- Aggregate CO2 emissions in work and non-work trips by consumer type, vehicle type, and location of consumers;
- Congestion delays in work and non-work travel by consumer type, vehicle type, and location of consumers; and
- Average fuel efficiency of consumer owned and operated vehicles.
Armed with these outputs, in addition to congestion tolls (Pigouvian tolls or cordon tolls), the new version can model taxes per gallon of gasoline, taxes on car ownership graded according to fuel inefficiency, and taxes proportional to the carbon emissions, in addition to congestion tolls which were modeled in version 1. Simulations can then be run to see how the effects of these taxes compare with each other when the alternative taxes are adjusted to raise the same aggregate revenue. This is known as a “revenue neutral comparison” in economics.
The researchers have almost completed the calibration of RELU-TRAN 2. The researchers are tinkering with the calibration, so that it fits known data from Chicago on vehicle miles, traffic speeds, fuel consumption, travel times, etc. In a few weeks, the researchers will be ready to start running simulations with the completed calibration of RELU-TRAN 2.
Energy/Electricity Sectors: In the report period, this work was significantly progressed by the team at the University of Illinois (UI) headed by Professor Geoffrey Hewings. The UI team produced a model of the generation and distribution of electricity in the Chicago region summarized in the progress report (see publications). The team assisted Alex Anas in the modeling of the demand for electricity by buildings and the calibration of such demand functions. The team provided the researchers with an input-output and social accounting matrix for the region, that includes utilities and energy as separate sectors. The researchers are now in discussions with the UI team about the checking of these models/data and about how they can be integrated with RELU-TRAN 2.
Future Activities:
There are three areas of work that need to be completed:
- The calibration of RELU-TRAN 2 needs to be completed and tested.
- The following articles will be written based on simulations with the calibrated RELU-TRAN 2:
- “Incorporating Vehicle Choice, Fuel Efficiency, and Emissions in Urban Economy Models”
- “Urban Sprawl, Fuel Economy, Congestion, and Emissions: Simulations for Chicago with RELU-TRAN 2”
- Integration of the electricity demand and electricity production sectors into RELU-TRAN 2, which will be named RELU-TRAN 3. In this part of the work, the electricity production and transmission sector model developed by REAL-UIUC will be integrated into RELU-TRAN 2 together with the electricity demand generated from buildings, inter-industry demands, and other sources (e.g., electric vehicles).
- Calibration of RELU-TRAN 3 for Chicago so that the whole modeling assembly meets empirically known relationships in land use, transportation, production, electricity, and emissions.
- Testing of the calibrated RELU-TRAN 3 for convergence and stability of the algorithm.
- Simulations with the calibrated RELU-TRAN 3 to test, especially, the relationships between urban sprawl, congestion, and various pricing policies on the one hand and emissions on the other.
- A book may be written describing the development and simulations made with RELU-TRAN. (all versions).
Journal Articles on this Report : 2 Displayed | Download in RIS Format
Other project views: | All 24 publications | 13 publications in selected types | All 8 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Anas A, Liu Y. A regional economy, land use, and transportation model (RELU-TRAN©): formulation, algorithm design, and testing. Journal of Regional Science 2007;47(3):415-455. |
R831841 (2007) R831841 (2008) R831841 (2009) R831841 (Final) |
Exit Exit |
|
Anas A. A unified theory of consumption, travel and trip chaining. Journal of Urban Economics 2007;62(2):162-186. |
R831841 (2007) R831841 (2008) R831841 (2009) R831841 (Final) |
Exit Exit |
Supplemental Keywords:
Air, air quality, analytical, behavioral models, business location, clean technologies, consumer behavior, cost-benefit analysis, decision making, demand for electricity, demographic change, dynamic computable general equilibrium (DCGE) modeling, economic behavior, economics, electric vehicles, electricity production, emissions, EMME/2, environmental regulation, future emissions, general equilibrium, global change, greenhouse gas, greenhouse gases, housing location, hydrogen vehicles, industrial location, infrastructure, land use, land use modeling, land use policies, long run, long-run effects, long term, long-term impact, long-term effects, manufacturing, mobile source, mobile sources, modeling, motor vehicle emissions, nitrogen oxides, ozone, particulate matter, particulates, preferences, public policy, regional development, residential location, sensitivity analysis, smart growth policies, social science, socioeconomic, stationary source, stationary sources, technological change, traffic congestion, transportation, transportation infrastructure, transportation modeling, travel demand, travel pattern, uncertainty, urban development, vehicle emissions, vehicle miles traveled, vehicle ownership, vehicular technology, VMT, VOC, Illinois (IL), midwest, Chicago MSA,, RFA, Scientific Discipline, Air, Air Quality, Environmental Chemistry, climate change, Air Pollution Effects, mobile sources, Environmental Monitoring, Ecological Risk Assessment, Ecology and Ecosystems, Urban and Regional Planning, Atmosphere, ecosystem models, infrastructure systems, traffic, engine exhaust, modeling regional scale ozone, vehicle emissions, human activities, motor vehicle emissions, Emissions Inventory Modeling System, air quality models, ozone, automotive emissions, traffic patterns, automobiles, automotive exhaust, green house gas concentrations, modeling, mobile source emissions, emissions impact, atmospheric pollutant loads, tropospheric ozone, regional emissions model, global warming, predicting ecological response, climate variability, community structure, Global Climate Change, ambient air pollutionRelevant Websites:
The Web site of Alex Anas contains some preliminary information about the project and will be updated. The project articles cited above will be posted there. To go there, type “Home Page of Alex Anas” in Google, or use the URL address: http://www.acsu.buffalo.edu/~alexanas/
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.
Project Research Results
- Final Report
- 2009 Progress Report
- 2007 Progress Report
- 2006 Progress Report
- 2005 Progress Report
- Original Abstract
8 journal articles for this project