ABSTRACT
Globally, the spread and severity of COVID-19 have been distinctly non-uniform. Seasonality was suggested as a contributor to regional variability, but the relationship between weather and COVID-19 remains unclear and the focus of attention has been on outdoor conditions. Because humans spend most of their time indoors and because most transmission occurs indoors, we here, instead, investigate the hypothesis that indoor climate-particularly indoor relative humidity (RH)-may be the more relevant modulator of outbreaks. To study this association, we combined population-based COVID-19 statistics and meteorological measurements from 121 countries. We rigorously processed epidemiological data to reduce bias, then developed and experimentally validated a computational workflow to estimate indoor conditions based on outdoor weather data and standard indoor comfort conditions. Our comprehensive analysis shows robust and systematic relationships between regional outbreaks and indoor RH. In particular, we found intermediate RH (40-60%) to be robustly associated with better COVID-19 outbreak outcomes (versus RH < 40% or >60%). Together, these results suggest that indoor conditions, particularly indoor RH, modulate the spread and severity of COVID-19 outbreaks.
Subject(s)
COVID-19 , Humans , COVID-19/epidemiology , Humidity , Weather , TemperatureABSTRACT
The spread of the COVID-19 pandemic is mainly due to the direct transmission routes of SARS-CoV-2 virus-carrying aerosols in indoor environments. In this study, the effect of indoor relative humidity (RH∞) on the number concentration, size distribution, and trajectory of sneeze droplets was studied in a confined space experimentally and numerically. Computational fluid dynamics (CFD) simulations using the renormalization group k-ε turbulence model by considering the one-way and two-way (humidity) coupling models were performed to assess the effects of humidity fields on the propagation of droplets. Number concentration profiles indicated that the RH∞ affected the dispersion modes of droplets differently for the puff, droplet cloud, fully-dispersed, and dilute-dispersed droplets phases identified by the shadowgraph imaging technique. The two-way (humidity) coupling model led to a close agreement with the experimental data in all phases. In particular, the two-way coupling provided better agreement with the data in the puff phase compared to the one-way coupling model. However, the one-way coupling model was sufficient for studying the motion of airborne droplets in the other phases. The velocity fields in the droplet cloud were more sensitive to RH∞ than the puff and fully-dispersed droplets phases. Also, the effect of RH∞ on the maximum spreading distance of droplets, dmax,sp, in the puff was insignificant, while its effect became dominant in the dilute-dispersed droplets phase. A dynamic change in the velocity profile of the sneeze jet was seen at a critical relative humidity RH∞,crit of about 48%. At RH∞< RH∞,crit, the number concentration of aerosolized droplets increases, significantly affecting the size distribution and the velocity of droplets. At RH∞≥ RH∞,crit, the effect of evaporation time on the number concentration, and diameter of droplets was negligible. At RH∞ of 24 and 64%, dmax,sp was 2.14 m (7 feet) and 3.05 m (10 feet), respectively. However, a dry indoor environment led to an increase in evaporation rate and more than four times number concentration of aerosolized droplets compared to a humid environment. Thus, the risk of direct transmission of Covid-19 in a humid indoor environment was higher than the dry conditions, which suggested the requirements for incorporating the RH∞ effect in the social distancing guideline.