Author: Weize Song (Tsinghua University) - China’s ambitious carbon-neutral target is challenged by the urbanization, especially the motorized commuting carbon emissions, which occupies more than one-half in the transport sector. Besides, there are substantial disparities in motorized commuting carbon emissions (CCE) across cities. However, few studies have focused on the roles of structural characteristics of road network, especially in the densely populated urban areas. Here, we propose a machine learning-based low emission road optimized model (LEROM) to identify the heterogeneous importance of road network management measures. The results demonstrate that there is a significant correlation between CCE and the entire betweenness, closeness, and flatness centralities of road systems of cities. For example, the national averaged Shapley contribution of betweenness centrality is 17.3% higher than that of density of road intersections. Besides, the synergic benefits of pairwise topological and geometrical measures mitigate CCE in northern China. Our findings provide road network-based regulation measure for the long-term low carbon transition priorities in the global transport sector. We argue that road network should be considered into the policy tool framework of climate mitigation and adaption actions.
