Models to Enable Estimation of Marginal CO2 Emissions in Electricity Production and Urban Mobility Systems.

Abstract

CO2 produced from the combustion of fossil fuels for energy production in electricity and transportation sectors is the biggest source of climate change causing greenhouse gases (GHG) in the U.S. GHG mitigation policies will affect how the existing systems operate and methods are necessary to examine the marginal effects and resulting change in CO2 emissions to evaluate the effectiveness of these policies. This dissertation develops models of electricity production and commuters’ choice of travel modes to enable the quantification of marginal CO2 emissions.

Electricity production systems constantly balance the demand and supply of electricity while functioning under a set of Operating Constraints (OCs). The model of electricity production developed in this dissertation incorporates major system OCs, which were either excluded or simplified in the previously used models, but are necessary to achieve reliable estimates of marginal CO2 emissions. The model was applied to evaluate the strategy for reducing CO2 emissions through increased utilization of existing Natural Gas (NG) generating units and reduced utilization of more CO2 intensive coal units. The analysis finds that about 27% less reduction in CO2 emissions could be achieved than estimated previously. The role of various OCs in limiting the extent to which CO2 emissions can be reduced is examined to inform future policy decisions.

Reducing the use of personal vehicles and increasing the utilization of public transportation and non-motorized modes such as biking has been considered as a CO2 mitigation measure. The second part of the dissertation develops models of commuting mode choices in Portland, Oregon to examine the potential for reducing vehicle miles traveled. The study compares the effectiveness of two mechanisms through which mode choices can be influenced – by varying the attributes of specific modes and by changing attitudes that determine how individuals value these attributes. The study develops a modeling approach that can predict individual-level mode choices as opposed to population level aggregate choices as done in previous studies. Because people can travel for different distances, the ability to predict individual-level choices is necessary to estimate passenger-miles traveled with specific modes and resulting CO2 emissions in a more deterministic manner.