Correlation and Causation Analysis Between COVID-19 and Environmental Factors in China

Abstract

Coronavirus disease 2019 (COVID-19) is seriously threatening and altering human society. Although prevention and control measures play an important role in preventing the transmission of severe acute respiratory syndrome coronavirus, signals of climate impact can still be detected globally. In this paper, the data of 265 cities in China were analyzed. The results show that the correlations between COVID-19 and air quality index (AQI) and PM2.5 concentration were very weak and that the correlations between COVID-19 and meteorological factors were significantly different in different climate backgrounds. So, a fixed model is not enough to describe the correlations. Overall, high humidity, low wind speed, and relatively lower air temperature are conducive to the spread of COVID-19. The climate background suitable for the spread of COVID-19 in China is air temperature 0~15°C, specific humidity <3 g kg−1, and wind speed <3 m s−1. The Granger causality test shows that there is a causal relationship between daily average air temperature and the number of COVID-19 confirmed cases in some cities of China, and air temperature is indicative of the number of confirmed cases the next day. However, this phenomenon is not universal due to regional climate differences.