Identifying the comprehensive metropolitan urban form is important to propose effective policies to mitigate transportation carbon emissions. A publicly accessible night-time light dataset was used to identify urban centers and develop two polycentric indices to compute the composition and configuration of urban form, respectively. We used the most populous 103 U.S. metropolitan statistical areas (MSAs), with their corresponding transportation carbon emissions, polycentric indices, population sizes, gross domestic product (GDP) per capita, and road network densities. We first explored the typology of urban form and classified MSAs into six types based on two polycentric indices. We then introduced correlation analysis and statistical models to test the relationships between polycentric urban form and transportation carbon emissions. We found: (1) more urban centers lead to more emissions (compositional dimension), (2) more spatially distributed urban centers result in less emissions (configurational dimension), and (3) population and GDP per capita are positively related to carbon emissions. These findings suggest the importance of measuring two polycentric dimensions separately but using them together. Urban planners should consider mixed strategies that combine the traditional intra-center-based smart growth principles and the metropolitan-level inter-centers spatial plan to effectively counteract climate change.

确定综合性大都市城市形态对于提出减少交通碳排放的有效政策非常重要。一个可公开访问的夜间灯光数据集用于识别城市中心并开发两个多中心指数来分别计算城市形态的组成和配置。我们使用了人口最多的 103 个美国大都市统计区 (MSA)，以及它们相应的交通碳排放量、多中心指数、人口规模、人均国内生产总值 (GDP) 和道路网络密度。我们首先探索了城市形态的类型学，并根据两个多中心指数将 MSA 分为六种类型。然后我们引入相关分析和统计模型来检验多中心城市形态与交通碳排放之间的关系。我们找到：（1）更多的城市中心导致更多的排放（组成维度），（2）更多空间分布的城市中心导致更少的排放（配置维度），以及（3）人口和人均GDP与碳排放正相关。这些发现表明分别测量两个多中心维度但同时使用它们的重要性。城市规划者应考虑将传统的基于中心内的智能增长原则与大都市级的中心间空间规划相结合的混合战略，以有效应对气候变化。这些发现表明分别测量两个多中心维度但同时使用它们的重要性。城市规划者应考虑将传统的基于中心内的智能增长原则与大都市级的中心间空间规划相结合的混合战略，以有效应对气候变化。这些发现表明分别测量两个多中心维度但同时使用它们的重要性。城市规划者应考虑将传统的基于中心内的智能增长原则与大都市级的中心间空间规划相结合的混合战略，以有效应对气候变化。