Figure 1
From: Understanding the mesoscopic scaling patterns within cities
Spatio-temporal dynamics of population. (a) Illustration of the temporal dynamics of population. The office building is of high population density in the daytime and of low density in the nighttime. The residential building, however, is the opposite. We assume that for each building the infrastructure volume (blue lines) is proportional to total building areas, which equals the footprint area times the number of floors. (b, c) The spatial distributions of daytime (b) and nighttime (c) populations for Beijing. These maps were created with the Datamaps tool (https://github.com/ericfischer/datamaps). The base map is OpenStreetMap, which is licenced under the Creative Commons Attribution-ShareAlike 2.0 licence (CC BY-SA). (d, e) The daytime, nighttime, and active population density gradients from the city center to the periphery for Beijing (d) and Chengdu (e). As expected, the daytime population density is much higher than the nighttime density around the urban center in all studies cities. Statistically, the exponential decay function has a higher \(R^2\) for both daytime and nighttime populations in Beijing. The power decay function achieves higher for daytime density, and exponential function has the better performance for nighttime density in Chengdu.
