Estimating the non-linear effects of urban built environment at residence and workplace on carbon dioxide emissions from commuting

• DOI: 10.3389/fpubh.2022.1077560
• Publication Date: 2022-11-29T06:11:15Z
• Authors: Qingchun Liu; Peixiong Zhao; Yingying Zhang; Zhihui Zhang; Jun Yang
• Keywords: Gradient boosting

Understanding the relationship between CO2 emissions from commuting (CEC) and built environment is crucial for sustainable transportation land-use policymaking during process of constructing a low carbon city. Previous studies usually assume that linear, which may lead to inaccurate CEC prediction ineffective policy. Using daily travel survey data residents in central city Jinan, this study adopted gradient boosting decision tree model explore threshold effect non-linear environments CEC. Our findings suggest 40% related workplace environment, higher than residential other socioeconomic variables. The five most important variables are road density within 1 km radius (13.493%), distance center at residence (10.908%, 10.530%), population (9.097%) bus stop (8.399%). Distance plays role its reflects influence urban spatial structure Jinan on Furthermore, thresholds relationships provide planning guidelines support development policies