Technology-based past and future emission inventories for the transportation sector

Emission inventories are necessary inputs to atmospheric models in determining air quality, health, and climate impacts. Transportation sector contributes greatly to emissions of all kinds of pollutants, thus it is crucial to understand its emissions and impacts. This dissertation improves the accounting of past and future emissions mainly in the transportation sector. This work develops an integrated modeling framework to forecast emissions and impacts from freight truck and rail transportation in the United States during 2010-2050. Macroeconomic scenarios are linked with a commodity input-output model, freight flow forecasting models, a dynamic vehicle fleet model, and a reduced-form air quality model to generate emissions and impacts under multiple future pathways. Emissions and impacts are projected under four macroeconomic scenarios featuring different economic growth rates and climate policies, as well as three scenarios of urban spatial forms. While freight activities more than double during 2010-2050, emissions of all pollutants decrease by more than 50% under all scenarios, as older vehicles built to less-stringent standards retire. In the business-as-usual scenario, mortalities decrease from 5500 to 3055, and the total integrated forcing increases from 14 to 22 TW from 2010-2050. Climate policy, in the form of carbon tax, increases oil fuel prices, causes a modal shift from truck to rail, and reduces mortalities and forcing by 20% and 30%, respectively. Eliminating high-emitting truck conditions reduces mortalities by 35% but has little impact on forcing. Polycentric and compact spatial forms reduce urban freight activity and emissions but increases population exposure, with a slight net benefit of 90 avoided deaths in 2050. This work also updates a previous historical emission inventory (1850-2010) of black carbon (BC) and primary organic carbon (OC). The mismatch between the BC concentration trend and the previous emission trend during 1960s-1980s in the United States indicates misrepresentations of past emission characteristics. A dynamic fleet vehicle model is applied to determine historical on-road vehicle emissions. Unlike previous inventories, the technology transitions in the vehicle fleet are explicitly modeled, including vehicle growth, retirement, and transitioning to high-emitting conditions. The emission factors and technology splits for on-road vehicles, off-road diesel engines, residential combustion, and shipping are updated per the observational constraints. Emissions from traditional brick and cement kilns are included. The updated increase almost linearly from 1270 Gg in 1850 to 5650 Gg in 1990, and slowly decrease till 2000, and finally increase again reaching 5580 Gg in 2010. Global OC emissions increase almost linearly from 1850-1975, and remain relatively constant until 1995 before emissions start to increase again, totaling 4950 Gg in 1850 and 11860 Gg in 2010. BC emission trend from on-road vehicles shows a rapid increase from 1960 to 1985, a relatively constant period for about 15 years, and then a slow decrease from 2000 to 2010, different than the previous continuous increasing trend. The updated global BC and OC emissions are 15-30% and 15-45% higher than the previous inventory, respectively.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Environ Engr in Civil Engr

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2018

Author dc:creator

• Liu, Liang

Contributors dc:contributor

• Bond, Tami C.
• Ouyang, Yanfeng
• Riemer, Nicole
• Koloutsou-Vakakis, Sotiria
• Minjares, Ray

• Emission inventory
• Transportation
• Technology
• Freight

Statement dc:rights

• Copyright 2017 Liang Liu

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/99168

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/99168