Global Prediction System of the COVID-19 Pandemic
This prediction model is a modified epidemiological SIR model incorporating real global pandemic data, the meteorological factors, and quarantine measures. It is assumed that the total population (N) in various regions remains unchanged during the outbreak; COVID-19 is only spread via human-to-human infection; there is no difference in immunity among individuals. The total population of each country (N = S + I + R) is divided into three categories, namely S (susceptible), I (infected) and R (cured & dead). The SIR infectious disease model is described as the following equations:
where r is the number of persons contacted with the infected person; β is the infection rate;μ is the withdrawal rate.
Based on the classic SIR model defined above, we developed a novel model that includes the impact of temperature, humidity, urban population density, and control intensity on COVID-19 infection. Our model is defined as follows:
 |
(4)
|
|
(5)
|
We found that environment temperature and the variation of NO2 concentration in the atmosphere are two important indicators to predict COVID-19 pandemic. The optimum temperature for the spread of COVID-19 is around 5-15 °C, where 70% of confirmed COVID-19 cases worldwide occurred (please see previous research results). Furthermore, the variation of NO2 concentration derived from satellite can well reflect the effectiveness of prevention and control measures implemented by governments. The decline in NO2 concentration is associated with reduced automobile exhaust and restricted industrial activities. When NO2 concentration falls, usually significant declines in traffic flows and increase in social distance are expected. After about 14 days, the pandemic curve will decline remarkably (please see previous research results).
To introduce temperature, humidity and government controls, we assume that：
 |
(6) |
|
|
(7)
|
|
where F1(T2m) and F2(RH2m) are the functions of the local temperature, relative humidity and daily confirmed new cases respectively; NO2 is the variation of local NO2 concentration, which reflects the intensity of quarantine measures. Strict quarantine measures help to increase the social distance and lessen the possibility of infection.
