Examining the population flow network in China and its implications for epidemic control based on Baidu migration data

This paper examines the spatial pattern of the population flow network and its implications for containing epidemic spread in China. The hierarchical and spatial subnetwork structure of national population movement networks is analysed by using Baidu migration data before and during the Chinese Spring Festival. The results show that the population flow was mainly concentrated on the east side of the Hu Huanyong Line, a national east-west division of population density. Some local hot spots of migration were formed in various regions. Although there were a large number of migrants in eastern regions, they tended to concentrate in corresponding provincial capital cities and the population movement subnetworks were affected by provincial administrative divisions. The patterns identified are helpful for the provincial government to formulate population policies on epidemic control. The movement flow from Wuhan (the city where the covid-19 outbreak) to other cities is significantly and positively correlated with the number of confirmed cases in other Chinese cities (about 70% of the population was constituted through innerprovincial movement in Hubei). The results show that the population flow network has great significance for informing the containment of the epidemic spread in the early stage. It suggests the importance for the Chinese government to implement provincial and municipal lockdown measures to contain the epidemic spread. The paper indicates that spatial analysis of population flow network has practical implications for controlling epidemic outbreaks.

Sheng Wei, Lei Wang