文献类型：期刊文献

英文题名：Nonlinear and Congestion-Dependent Effects of Transport and Built-Environment Factors on Urban CO2 Emissions: A GeoAI-Based Analysis of 50 Chinese Cities

作者：Chen, Xiao Li, Yubin Li, Xiangyu Zheng, Huang

第一作者：Chen, Xiao

通信作者：Li, XY[1];Zheng, H[2]

机构：[1]Xian Traff Engn Inst, Sch Traff & Transportat, Xian 710300, Peoples R China;[2]Beijing Jiaotong Univ, Sch Traff & Transportat, Beijing 100044, Peoples R China;[3]Zhejiang Univ, Coll Civil Engn & Architecture, Hangzhou 310058, Peoples R China;[4]Guizhou Inst Technol, Sch Transportat Engn, Guiyang 550003, Peoples R China

第一机构：Xian Traff Engn Inst, Sch Traff & Transportat, Xian 710300, Peoples R China

通信机构：corresponding author), Zhejiang Univ, Coll Civil Engn & Architecture, Hangzhou 310058, Peoples R China;corresponding author), Guizhou Inst Technol, Sch Transportat Engn, Guiyang 550003, Peoples R China.|贵州理工学院;

年份：2026

卷号：16

期号：2

外文期刊名：BUILDINGS

收录：;EI(收录号：20260519985885);Scopus(收录号：2-s2.0-105028645493);WOS:【SCI-EXPANDED(收录号:WOS:001672460900001)】；

基金：This research was funded by Xiao Chen, grant number 2024KY-34 and The APC was funded by Xiao Chen, grant number 2024KY-34.

语种：英文

外文关键词：urban CO2 emissions; transport emission; built environment; GNNWR; GeoAI

摘要：Understanding how transport conditions and the built environment shape urban CO2 emissions is critical for low-carbon urban development. This study analyses CO2 emission intensity across fifty major Chinese cities using integrated ODIAC emissions, VIIRS night-time lights, traffic performance indicators, built-environment morphology, population/POI structure, and socioeconomic controls. We develop a GeoAI workflow that couples XGBoost modelling with SHAP interpretation, congestion-based city grouping, and 1 km grid-level GNNWR to map intra-urban spatial non-stationarity. The global model identifies night-time light intensity as the strongest predictor, followed by population density and building density. SHAP results reveal pronounced nonlinearities, with high sensitivity at low-medium levels and diminishing marginal effects as activity and density increase. Although transport indicators are less influential in the aggregate model, their roles differ across congestion regimes: in low-congestion cities, emissions align more consistently with overall activity intensity, whereas in high-congestion cities they respond more strongly to population distribution, motorisation, and built-form intensity, with less stable relationships. Grid-level GNNWR further shows that key mechanisms are spatially uneven within cities, with local effects concentrating in specific cores and corridors or fragmenting across multiple subareas. These findings demonstrate that emission drivers are context-dependent across and within cities. Accordingly, uncongested cities may gain more from activity-related energy-efficiency measures, while highly congested cities may require congestion-sensitive land-use planning, spatial-structure optimisation, and motorisation control. Integrating explainable GeoAI with regime differentiation and spatial heterogeneity mapping provides actionable evidence for targeted low-carbon planning.