Grantee Research Project Results

2017 Progress Report: Evaluating the Timeline of Particulate Matter Exposure from UrbanTransportation and Land-Use Greenhouse Gas Mitigation Strategies Using aNovel Modeling Framework

EPA Grant Number: R835885
Title: Evaluating the Timeline of Particulate Matter Exposure from UrbanTransportation and Land-Use Greenhouse Gas Mitigation Strategies Using aNovel Modeling Framework
Investigators: Rowangould, Gregory
Institution: University of New Mexico
Current Institution: University of Vermont
EPA Project Officer: Keating, Terry
Project Period: January 1, 2016 through December 31, 2018 (Extended to December 31, 2021)
Project Period Covered by this Report: January 1, 2017 through December 31,2017
Project Amount: $335,605
RFA: Particulate Matter and Related Pollutants in a Changing World (2014) RFA Text | Recipients Lists
Research Category: Air , Climate Change , Early Career Awards

Objective:

The aim of this research is to create a clearer picture of how changes in land-use patterns and transportation systems affect population exposure to particulate matter (PM) pollution from vehicle traffic. The research also aims to understand how the timing of land-use and transportation system changes, particularly those strategies intended to mitigate climate change, affect greenhouse gas (GHG) emissions and cumulative exposure to PM emissions. Finally the research will consider how well changes in PM emission inventories, which are widely used to assess improvements in air quality, correspond to changes in PM exposure.

Progress Summary:

During the first year of the project we acquired models and data necessary for completing this project. This includes travel demand and land-use models from the Mid Region Council of Governments (MRCOG – the Albuquerque metropolitan area planning organization) as well as travel demand and land-use model outputs from the Atlanta Regional Commission (ARC – the Atlanta metropolitan area planning organization). The models and data are very complex and required significant effort to setup and ensure their functionality in our research lab.

During the second year of the project we began modeling GHG and fine particulate matter (PM2.5) emissions and PM2.5 exposure in the Atlanta and Albuquerque metropolitan areas using the models and data provided by ARC and MRCOG. Our modeling focused on evaluating the long range regional transportation plans created by each agency and we began work on developing and modeling additional scenarios for the Albuquerque metropolitan area. Each region uses a different, yet innovative, type of modeling system. The different models used by ARC and MRCOG provide an opportunity to understand how various modeling technologies affect our understanding of vehicle emissions and emissions exposure and what can be expected in the future under alternative transportation and land-use planning strategies.

In Atlanta, the ARC uses an activity-based travel demand model. This type of model contains a synthetic population of “agents” that represent everyone in the region. The agents are assigned a daily schedule of activities to complete based on their socioeconomic profile. Each agent attempts to complete their daily schedule of activities which often requires travel to various places throughout the region (e.g., to work, school or retail outlets). Each agent’s travel throughout the region can affect the travel of other agents by creating traffic congestion, which may then cause an agent to seek an alternative route, an alternative mode of transportation, or change to their activity schedule. Activity based travel demand models provide a unique opportunity to understand how each person’s unique daily travel pattern also affects their exposure to vehicle emissions.

Using output provided by ARC for the year 2017, we modeled the average annual concentration of PM2.5 for each period of the day across the entire region. We then estimated the average annual PM2.5 exposure of each agent by tracing their movements throughout the region and through areas of various PM2.5 concentration. We refer to this as “dynamic exposure”. We also estimated each agent’s exposure based on the PM2.5 concentration where they live, which is typically how emissions exposure has been evaluated. We refer to this as “static exposure”. Our results find that the static exposure method overestimates exposure in populations living near major roadways by a moderate amount and severely underestimates exposure for populations living in suburban and rural areas away from major roads. Overall, the dynamic exposure method results in higher PM2.5 exposure estimates. The large difference between the two methods results from the static approach not accounting for the large share of a person’s daily exposure to PM2.5 from vehicle emissions that occurs while they travel and are at work. Our future work will use the dynamic modeling method to understand how changes in land-use and the transportation system affect individual’s exposure to PM2.5.

In Albuquerque, MRCOG uses a very common trip based travel demand model. There are no individual agents seeking to complete a daily activity schedule, instead statistical equations estimate how individuals living in different types of households and with different socioeconomic characteristics travel. One consequence of this modeling method is that it is not possible to track where each individual spends time throughout the day and therefore only static exposure estimates are possible. MRCOG, however, also uses a land-use simulation model to understand how changes in traffic patterns affect land-use and land development. The travel demand and land-use model work together. Traffic outputs from the travel demand model are used as inputs to the land-use model. Parcels of land that are accessed by more congested roadways become less attractive for development (all else being equal). Land-use forecasts from the land-use model are then used as inputs for the travel demand model. Where there is less development, there is less traffic and vice versa. We configured MRCOG’s travel demand and land-use models to interact with each other on an annual basis (i.e., swapping traffic and land-use forecasts with each other annually), something that to our knowledge has not previously been done. This frequent interaction provides a more realistic treatment of how a region grows, and how changes to the transportation system and land-use rules affect growth and travel behavior.

We use our highly integrated version of MRCOG’s travel demand and land-use models to accomplish two unique types of emission analysis. First, we model how the Albuquerque region grows year over year, including where people live and work and the resulting traffic volume and speed on each roadway. We use these outputs to then make annual GHG and PM2.5 emission estimates and PM2.5 emission exposure estimates. Additionally, we can use our annual modeling outputs to estimate the cumulative impacts of the region’s long range transportation and land-use plans. Typically, plans are evaluated by considering the change from the base year (2012) to the final year in the planning period (2040) without any understanding of what occurs during the period of time in-between. Since the accumulation of GHG emissions in the atmosphere is what matters for evaluating climate change impacts and negative health outcomes associated with PM2.5 exposure are not reversed by cleaner air in the future, understanding annual and cumulative emissions and air quality impacts is important.

During year 2 we completed modeling for MRCOG’s currently adopted long range regional transportation plan. We find that GHG and PM2.5 emissions vary significantly over time. GHG emissions are expected to increase at first, then decrease before finally increasing again during the final years of the planning period. We find that PM2.5 emissions and PM2.5 exposure will decrease rapidly at first, and then slowly increase in the final years of the planning period. These trends overtime have not previously been documented and may have important implications on how long range plans are evaluated. We also find that results obtained by interacting MRCOG’s models on an annual basis vary significantly from when the travel demand and land-use models are only interacted once (many regions do not conduct land-use modeling at all and those that do typically conduct a few intermediate year interactions at most). When interacting the modeling annually, the average concentration of PM2.5 emissions is similar to when the models are only interacted once; however, PM2.5 emissions are concentrated much more in the central part of Albuquerque and are less concentrated in outlying suburban and rural areas. Analysis of traffic and land-use outputs indicate that the greater interaction between the two models results in the region being modeled to grow more densely, which results in more traffic congestion in central Albuquerque and hence a greater concentration of vehicle emission.

Our future work will use our highly integrated land-use and travel demand modeling method to evaluate a wide range of transportation and land-use scenarios that we develop for the Albuquerque region to seek out strategies that can mitigate both GHG emissions and PM2.5 emissions exposure over the entire planning period. We have begun this phase of our project by developing 17 land-use and transportation scenarios that aim to produce significant GHG emission reductions. We have modeled these scenarios using MRCOG’s travel demand model but have limited the interaction with the land-use model so that we can quickly evaluate them (a more complete analysis can take up to 3 weeks per scenario). This method provides a good estimate of how effective various strategies might be in reducing GHG emissions. Over the remainder of the project period we will evaluate the most effective scenarios using our highly integrated modeling system and we will also evaluate disparities in exposure by land-use and the neighborhood level socioeconomic characteristics.

Future Activities:

We now have developed, tested, and used all the modeling methods required to complete the project’s 7 main tasks and answer the 5 research questions we posed in our research proposal. We plan to complete all remaining project tasks; however, to do so we plan to request a one year no-cost extension.

Over the remaining period of this project we plan to use our dynamic modeling approach to evaluate year 2020 and 2040 planning scenarios in the Atlanta region and we also hope to obtain several alterative planning scenarios from the ARC. The alterative planning scenarios would allow us to evaluate tradeoffs in GHG emission reductions and PM2.5 emissions exposure as originally planned. We also plan to complete a more in-depth analysis of the results we have already obtained, as well as future modeling results, focusing on disparities in exposure by land-use, activity, and socioeconomic characteristics.

We also plan to model a wide range of alterative land-use and transportation plans for the Albuquerque region as outlined in our research proposal. We will begin by modeling a range of strategies aimed at maximizing GHG emission reductions. We will then evaluate alternatives that also seek to maximize PM2.5 emission reductions while maintaining GHG emission reductions. We will also be completing more in-depth analysis of our current and future modeling results to understand exposure disparities by land-use and socioeconomic characteristics.

Journal Articles on this Report : 2 Displayed | Download in RIS Format

Publications Views
Other project views:	All 16 publications	6 publications in selected types	All 6 journal articles

Publications
Type	Citation	Project	Document Sources
Journal Article	Tayarani M, Poorfakhraei A, Nadafianshahamabadi R, Rowangould G. Can regional transportation and land-use planning achieve deep reductions in GHG emissions from vehicles? Transportation Research Part D:Transport and Environment 2018;63:222-235.	R835885 (2017) R835885 (2018) R835885 (Final)	Full-text: Science Direct-Full Text HTML Exit Abstract: Science Direct-Abstract Exit Other: Science Direct-Full Text PDF Exit
Journal Article	Tayarani M, Nadafianshahamabadi R, Poorfakhraei A, Rowangould G. Evaluating the cumulative impacts of a long range regional transportation plan:particulate matter exposure, greenhouse gas emissions, and transportation system performance. Transportation Research Part D:Transport and Environment 2018;63:261-275.	R835885 (2017) R835885 (2018) R835885 (Final)	Full-text: Science Direct-Full Text HTML Exit Abstract: Science Direct-Abstract Exit Other: Science Direct-Full Text PDF Exit

Supplemental Keywords:

Dispersion modeling, mobile sources, travel demand

Progress and Final Reports:

Original Abstract

The 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.