Spatial network connectivity of population and development in the USA; Implications for disease transmission

Zipfs Law states that rank-size distributions of city populations follow a power law with an exponent of -1. The assertion of a universal power law is controversial because the linearity and slope appear to vary over time and among countries. We compare census enumerations and night light luminance as proxies for population density and intensity of development in the contiguous United States. Treating population density and development intensity as continuous quantities allows for the definition of spatial networks based on the level of spatial connectivity. The resulting distributions of spatial network components (subsets of connected nodes) vary with degree of connectivity, but maintain consistent scaling over a wide range of network sizes. At continental scales, spatial network rank-size distributions obtained from both population density and night light brightness are well-fit by power laws with exponents near -1 for a wide range of density and luminance thresholds. However, the largest components (10,000 - 100,000 sq.km) are far larger than individual cities and represent spatially contiguous agglomerations of urban, suburban and periurban development. Projecting county-level numbers of confirmed cases of SARS-CoV-2 for the US onto spatial networks of population and development allows the spatiotemporal evolution of the epidemic to be quantified as propagation within networks of varying connectivity.. The results show an abrupt transition from slow increases in confirmed cases in a small number of network components to rapid geographic dispersion to a larger number of components before mobility reductions occurred in March 2020.
Comment: 21 pages, 9 figures