“Increase in COVID-19 Numbers in Downwind Regions Following a Wind Change” – Alice Hailwood , Catriona Graham , Sandeep Ramalingam
University of Edinburgh – NHS Lothian – Edinburgh (United Kingdom)

Background: Social distancing guidelines for COVID-19 concentrate on virus spread by large droplets/direct contact. While aerosol transmission is considered important in health care setting its importance in community transmission is not emphasized in guidelines. The distance a droplet can travel is affected by its size and wind speed. Evidence suggests that droplets <100┬Ám evaporate and become droplet nuclei before they settle down. These droplet nuclei could potentially be carried long distances by the wind. We investigated the change in trends in case numbers in areas/regions tens of kilometres downwind of a hotspot of COVID-19 cases.

Methods : We used wind directions and speed data (Met Office/Ventusky) to identify periods of sustained change in wind direction across 3 large COVID-19 hotspots (Birmingham, London, Newcastle areas), and a smaller hotspot (Ashford, Kent) during the first wave of COVID-19. Based on official daily reports, cases/100,000 population in hotspots and surrounding areas were plotted against time. We looked at the change in case numbers within 14 days from the start of the wind change (i.e., the incubation period) in downwind areas (Worcestershire and Warwickshire for the Birmingham area; Surrey for the London area; County Durham and Northumberland for the Newcastle area and the areas around Ashford). Regression lines were fitted to the 7-day rolling average case rates in hot-spots compared to adjacent down-wind areas

Results: Periods of sustained change in wind direction were identified in all regions after UK national lockdown. When compared to the hotspot, the trend in case rates in downwind areas plateau or decrease at a slower rate and in one instance showed an increase. The difference in trends in case rates between Birmingham/London/Newcastle hotspots and their corresponding downwind areas were significantly different (p<0.0001).

Conclusions: The clear change in trends in case rates in downwind areas point to an increase in case rates in those areas. As this change happens within the incubation period (i.e., within 14 days), it suggests that viral particles are carried downwind over 50-100 kilometres leading to an increase in infections in these areas. Measures to reduce aerosol transmission in the community hence need to be strengthened.