Meteorology and the Epidemic

It couldn’t be clearer at this point that CV-19 is highly seasonal, although debate continues on what precise conditions favor or disfavor transmission.  It has taken time, obviously, to be able to observe the patterns and to gather data related to spread.  Along with DD, who has done so much good analytic work with me, I am looking at meteorological factors and case levels in Minnesota to see if there is any clear relationship, now that we have had almost a year of spread in the state.  As that work proceeds I will give you periodic updates.

This study looks again at various factors and purports to find a testable relationship.  (AIP Article)   The authors determined that the concentration of the virus in droplets was determined by a combination of temperature, relative humidity and wind speed, with concentration decreasing with increased temperature and increasing with increased wind speed and relative humidity.  Obviously, higher, more persistent concentrations are more likely to lead to transmission.  This was only a modeling study itself, so limited direct experimental evidence.  The researchers ran a very large number of simulations of viral concentrations in saliva and the consequences of expellation in different meteorological conditions, resulting in a formula.  They suggest that this formula needs to be used as a parameter, called the “airborne infection rate index”, in epidemic modeling. While I am not sure what the exact formula would be, I completely agree that with a seasonal virus, meteorologic factors should be included in modeling.

The authors tested their formula on case rates in several cities globally.  The found good concordance with case rates.  Specific discussion, including a projection forward for a couple months, was given for Paris and New York and Rio de Janeiro in the Southern hemisphere.  The Southern hemisphere has a flatter peak due to different general weather factors in that hemisphere.   This may also account for some different shaping of epidemic curves in the states in the northern versus southern US.  A very interesting paper.

The reasons for the effect of meteorological conditions on transmission is likely fourfold.  One is direct effect on the virus.  It has lipid envelope, the stability of which is affected by factors like temperature.  More sunlight, specifically ultraviolet, likely degrades the viral particle.  The second is an effect on transmission conditions.  Lower temperatures and higher relative humidity may favor the development and persistence of certain sized particles carrying the virus.  Air flow, both indoors and outdoors may also affect persistence and concentration.  The third is the effect on human biochemistry.  Most notable among these hypotheses is that less sunlight affects vitamin D levels, which play a role in immune response strength.  Drier and colder air may also affect the upper respiratory tract tissues, making them more or less susceptible to infection.  And lastly, weather variables can influence human behavior, such as relative time indoors.

There is a great deal of research to be done to sort more carefully through all these possibilities, but it has been a real disservice in this epidemic that very little modeling has incorporated these factors despite the common seasonal patterns to respiratory viruses, including the seasonal coronaviruses.