An early outbreak of COVID-19 took place in Wuhan city in later December 2019 which been diffused to other areas through population flow (Kraemer et al. 2020; Shen 2021a; 2021b). Among the total population of 12.33 million in Wuhan, 3.95 million were migrants from rest of Hubei (68.29%) and rest of Mainland China (31.71%). The COVID-19 risk was not well known to public in China before 20 January 2020. Wuhan city was locked down on 23 January, 2 days before the Lunar New Year on 25 January 2020. Majorities of migrants travelled back to their hometowns during the spring transport period from 10 to 23 January 2020.
To assess the role of migrant flow in the spatial diffusion of COVID-19, an agent-based spatial SEIR model is developed consisting of 32 areas including Wuhan, rest of Hubei and other 30 provinces in Mainland China. The model reconstructs the visitor and migrant flows and COVID-19 diffusion in the simulation period from 1 December 2019 (day 1) to 1 March 2020 (day 92). The accumulated number of exported exposed visitors from Wuhan was 373 on day 41 when 40 exported exposed returning migrants from Wuhan occurred. By day 48, the accumulated number of exported exposed returning migrants from Wuhan (372) surpassed the accumulated number of net exported exposed visitors from Wuhan (371). By day 92, the exported exposed returning migrants from Wuhan (2017) was greater than the accumulated number of net exported exposed visitors from Wuhan (1378).
Thus considering the by-directional nature of visitor flow, the migrant flow exported more exposed population to rest of Mainland China than the net visitor flow from day 48. The migrant flow plays an equal if not more important role than visitor flow in the spatial diffusion of COVID-19 from Wuhan city to provincial regions in China. The results of this research have important policy implications to respond to new pandemic in future (Valsecchi and Durante 2021; Shen 2021c).
Mots clés : spatial diffusion|migrant flow|visitor flow|COVID-19 pandemic|China
A103033JS