Given accelerated urbanization and increasing residential mobility, optimizing urban environments to encourage low-carbon travel has attracted considerable attention. However, within complex urban systems, further research is needed to clarify how urban street-scape visual features interact with factors such as residents’ travel distances and origin-point characteristics—thereby influencing low-carbon travel behavior and associated carbon emissions. Taking Shenzhen—a megacity in China—as an example, this research integrates multi-source data to explore the underlying mechanisms influencing residents’ low-carbon travel behaviors. Specifically, we analyzed survey questionnaires collected from 3,976 urban park visitors and utilized deep learning algorithms to extract urban street-scape visual features from 137,000 street-view images. Moreover, localized carbon emission intensities for different transportation modes were applied to estimate the probabilities of residents’ travel mode choices and associated carbon emissions at a spatial grid scale across various urban locations. Our findings indicate that socio-demographic attributes, urban street-scape environmental features, neighborhood environmental characteristics, and travel distance significantly increase the likelihood of residents adopting low-carbon travel behaviors. Additionally, per capita carbon emissions of urban park visitors show a significant negative correlation with the green view index among urban street-scape visual features. Based on these findings, improving street-level green view index can effectively contribute to carbon emission reductions, and prioritizing enhancements in densely populated urban areas can yield even greater emission reduction benefits. This study offers a novel methodology and empirical evidence to support low-carbon development strategies in megacities.
扫码关注我们
求助内容:
应助结果提醒方式:
