This paper examines how fragmentation of critical infrastructure impacts the spread of the coronavirus outbreak in New York City at the neighbourhood level. The location of transportation hubs, grocery stores, pharmacies, hospitals and parks plays an important role in shaping spatial disparities in virus spread. Using supervised machine learning and spatial regression modelling we examine how the geography of COVID-19 case rates is influenced by the spatial arrangement of four critical sectors of the built environment during the public health emergency in New York City: health care facilities, mobility networks, food and nutrition and open space. Our models suggest that an analysis of urban health vulnerability is incomplete without the inclusion of critical infrastructure metrics in dense urban geographies. Our findings show that COVID-19 risk at the zip code level is influenced by (1) socio-demographic vulnerability, (2) epidemiological risk, and (3) availability and access to critical infrastructure.
本文研究关键基础设施的碎片化对纽约市冠状病毒社区传播爆发的影响。交通枢纽、杂货店、药房、医院和公园的位置在决定病毒传播的空间差异方面起着重要作用。我们利用监督机器学习和空间回归模型，研究了在纽约市该突发公共卫生事件期间，新冠肺炎 (COVID-19) 发病率的地理分布如何受到建筑环境中四个关键要素的空间排列的影响，这四个要素即医疗保健设施、移动网络、食物和营养以及开放空间。我们的模型表明，如果不将密集城市地理中的关键基础设施指标包括在内，对城市健康漏洞的分析将是不完整的。我们的研究结果表明，邮政编码级别的新冠肺炎风险受 (1) 社会人口脆弱性、(2) 流行病学风险和 (3) 关键基础设施的可用性和可访问性的影响。.
Keywords: built environment; health; inequality; infrastructure; machine learning; social justice; spatial regression analysis.
© Urban Studies Journal Limited 2021.