Seasonal variation in the transmission rate of covid-19 in a temperate climate can be implemented in epidemic population models by using daily average temperature as a proxy for seasonal changes in transmission rate

From march 2020 to march 2022 covid-19 has shown a consistent pattern of increasing infections during the Winter and low infection numbers during the Summer. Understanding the effects of seasonal variation on covid-19 spread is crucial for future epidemic modelling and management. In this study, seasonal variation in the transmission rate of covid-19, was estimated based on an epidemic population model of covid-19 in Denmark, which included changes in national restrictions and introduction of the $\alpha$-variant covid-19 strain, in the period March 2020 - March 2021. Seasonal variation was implemented as a logistic temperature dependent scaling of the transmission rate, and parameters for the logistic relationship was estimated through rejection-based approximate bayesian computation (ABC). The likelihoods used in the ABC were based on national hospital admission data and seroprevalence data stratified into nine and two age groups, respectively. The seasonally induced reduction in the transmission rate of covid-19 in Denmark was estimated to be $27\%$, (95% CI [$24\%$; $31\%$]), when moving from peak Winter to peak Summer. The reducing effect of seasonality on transmission rate per $+1{}^{\circ }$C in daily average temperature were shown to vary based on temperature, and were estimated to be $-2.2\%[-2.8\%;-1.7\%]$ pr. 1 ${}^{\circ }$C around $2^{\circ }$C; $2\%[-2.3\%;-1.7\%]$ pr. 1 ${}^{\circ }$C around $7{}^{\circ }$C; and $1.7\%[-2.0\%;-1.5\%]$ pr. 1 ${}^{\circ }$C around a daily average temperature of 11 ${}^{\circ }$C.