Car use creates significant externalities for urban residents worldwide. City characteristics such as the configuration of the urban landscape and street network likely influence the use and attractiveness of automobiles, especially in rapidly urbanizing areas such as Latin America. The understanding of factors associated with motorization can inform planning measures to reduce car usage, and to promote healthier, safer, and more sustainable urban lifestyles. We harmonized official passenger vehicle data from 300 cities with >100,000 inhabitants in Brazil, Chile, Colombia, and Mexico, and we calculated urban landscape metrics from satellite imagery and street network metrics from OpenStreetMaps. Analyzed cities had an average of 273.3 cars per 1,000 residents in 2015 and showed an average car rate increase of 30 % between 2010 and 2015. We used negative binomial regression to examine the association between car rates and urban landscape and street network characteristics, and linear regression to examine the association between the same characteristics and car rate increases. Car rates in the 300 cities analyzed showed a partial positive association with development fragmentation, and a consistent positive association with urban form complexity and circuity of the street network. In addition, the increase in car rates between 2010 and 2015 showed a negative association with population density. Implementing regional policies to reduce development fragmentation, to promote compact urban forms and less circuitous street networks may help reducing motorization in Latin American cities. Special attention needs to be paid to low density areas, where the increase in vehicle rates has been more pronounced.

汽车使用给全球城市居民带来了显着的外部性。城市景观和街道网络等城市特征可能会影响汽车的使用和吸引力，尤其是在拉丁美洲等快速城市化地区。了解与机动化相关的因素可以为减少汽车使用的规划措施提供信息，并促进更健康、更安全和更可持续的城市生活方式。我们协调了来自巴西、智利、哥伦比亚和墨西哥的 300 个城市的官方乘用车数据，人口超过 100,000，我们根据卫星图像和 OpenStreetMaps 的街道网络指标计算了城市景观指标。分析的城市在 2015 年平均每 1,000 名居民拥有 273.3 辆汽车，并显示 2010 年至 2015 年期间汽车的平均增长率为 30%。我们使用负二项式回归来检验汽车率与城市景观和街道网络特征之间的关联，并使用线性回归来检验相同特征与汽车率增加之间的关联。分析的 300 个城市的汽车率显示出与发展碎片化的部分正相关，以及与城市形态复杂性和街道网络迂回的一致正相关。此外，2010 年至 2015 年间汽车价格的增长与人口密度呈负相关。实施区域政策以减少发展碎片化、促进紧凑的城市形态和减少迂回的街道网络可能有助于减少拉丁美洲城市的机动化。需要特别注意低密度区域，