Impact of Daily Weather on COVID-19 outbreak in India

Gupta, A.; Pradhan, B.

The COVID-19 pandemic has outspread obstreperously in India. As of June 04, 2020, more than 2 lakh cases have been confirmed with a death rate of 2.81%. It has been noticed that, out of each 1000 tests, 53 result positively infected. In order to investigate the impact of weather conditions on daily transmission occurring in India, daily data of Maximum (TMax), Minimum (TMin), Mean (TMean) and Dew Point Temperature (TDew), Diurnal Temperature range (TRange), Average Relative Humidity, Range in Relative Humidity, and Wind Speed (WS) over 9 most affected cities are analysed in several time frames: weather of that day, 7, 10, 12, 14, 16 days before transmission. Spearman rank correlation (r) shows significant but low correlation with most of the weather parameters, however, comparatively better association exists on 14 days lag. Diurnal range in Temperature and Relative Humidity shows non-significant correlation. Analysis shows, COVID-19 cases likely to be increased with increasing air temperature, however role of humidity is not clear. Among weather parameters, Minimum Temperature was relatively better correlate than other. 80% of the total confirmed cases were registered when TMax, TMean, TMin, TRange, TDew, and WS on 12-16 days ago vary within a range of 33.6-41.3 deg C, 29.8-36.5 deg C, 24.8-30.4 deg C, 7.5-15.2 deg C, 18.7-23.6 deg C, and 4.2-5.75 m/s respectively, hence, it gives an idea of susceptible weather conditions for such transmission in India. Using Support Vector Machine based regression, the daily cases are profoundly estimated with more than 80% accuracy, which indicate that coronavirus transmission cannot be well linearly correlated with any single weather parameters, rather multivariate non-linear approach must be employed. Accounting lag of 12-16 days, the association found to be excellent, thus depict that there is an incubation period of 12-16 days for coronavirus transmission in Indian scenario.