We investigate the climatic influence on COVID-19 transmission risks in 228 cities globally across three climatic zones. The results, based on the application of a Boosted Regression Tree algorithm method, show that average temperature and average relative humidity explain significant variations in COVID-19 transmission across temperate and subtropical regions, whereas in the tropical region, the average diurnal temperature range and temperature seasonality significantly predict the infection outbreak. The number of positive cases showed a decrease sharply above an average temperature of 10°C in the cities of France, Turkey, the US, the UK, and Germany. Among the tropical countries, COVID-19 in Indian cities is most affected by mean diurnal temperature, and those in Brazil by temperature seasonality. The findings have implications on public health interventions, and contribute to the ongoing scientific and policy discourse on the complex interplay of climatic factors determining the risks of COVID-19 transmission.
Keywords: Boosted Regression Tree; COVID-19 transmission; SARS-CoV-2; climatic association; stochastic model.