Nonlinear impacts of urban built environment on freight emissions

Abstract

Truck activities are the primary source of NOx and PM2.5 emissions in urban environments. This study harnesses a vast truck GPS dataset from Chengdu, employing the International Vehicle Emissions (IVE) model to estimate freight-related emissions and utilizing Gradient Boosting Regression Trees (GBRT) for an in-depth analysis of the impact of urban built environment factors—such as land use, traffic design, and density—on these emissions. It reveals significant correlations, highlighting that main road density, built-up area, industrial land ratio, and electricity consumption play pivotal roles in affecting freight pollution. Specifically, regions with main road density ranging from 5890 to 9560 m/km² exhibit a marked increase in emissions. These insights provide robust support for the formulation of urban planning and traffic management strategies aimed at mitigating freight-related pollution, emphasizing the critical need to consider the urban built environment in policy-making processes.