Spatial differentiation of transport carbon emissions under the 5Ds framework: implications for sustainable spatial planning in Guangzhou, China

Sustainable spatial planning is critical for reconciling rapid urbanization with low-carbon transitions, yet the spatial differentiation of transport carbon emissions (TCE) and its moderating factors remain insufficiently integrated into planning frameworks. This study examines how built environment factors influence TCE under the 5Ds framework in Guangzhou, China, using mobile phone signaling data, multi-source spatial data, and the GeoDetector model. We reveal three key findings. First, TCE exhibits a polycentric agglomeration pattern, reflecting finger-like urban expansion. Second, development intensity (FAR) is the strongest explanatory factor (q = 0.309), with the FAR–employment accessibility interaction showing the strongest synergistic effect (q = 0.323), validated as genuine nonlinear enhancement, indicating that density regulation and jobs-housing coordination must be planned jointly. Third, building height and housing price stratifications reveal monotonic gradient effects—FAR’s explanatory power increases from low-rise to high-rise areas and from low-price to high-price zones—demonstrating that uniform policies may exacerbate spatial inequities. Land surface temperature shows an inverted U-shaped moderating effect, highlighting the need for climate-adaptive planning. These findings support a differentiated, context-sensitive planning framework that integrates density regulation, jobs-housing coordination, and climate-adaptive design, moving beyond the conventional assumption that uniform-built environment interventions yield uniform carbon outcomes.