COVID19 update, April 20, 2020: sensitivity to sunshine and humidity strongly suggests seasonality

(1) So how seasonal is COVID19? A preliminary technical report from the Department of Homeland Security’s scientific and technical division offers clues. Yahoo News obtained a copy https://news.yahoo.com/sunlight-destroys-coronavirus-very-quickly-new-government-tests-find-but-experts-say-pandemic-could-still-last-through-summer-200745675.html

To cut a long story short: the researchers exposed virus samples to artificial sunlight of varying intensity. The half-life of the virus in the equivalent of midday sun at mid-US latitudes was two minutes: that meant in practice that just 20 minutes, or ten half-lives, would kill 99.9% (or to be precise, 1023/1024) of the viruses. In weaker sunlight, the 

For influenza viruses, a 2009 PNAS paper (http://doi.org/10.1073/pnas.0806852106; editorial commentary at http://doi.org/10.1073/pnas.0900933106 ) showed an inverse relationship between absolute humidity and virus survival/transmission. From the preliminary findings of the DHS group, it appears that the same applies to coronaviruses. Cold and dry weather is best for virus ‘survival’, hot and humid worst; dry but abundant sunshine will still whack it.

These findings suggest that the COVID19 epidemic likely will exhibit similar seasonality as influenza, and that it will be less virulent in sunny climates. I notice Australia and South Africa got off pretty light this round, as these countries were in their antipodal summer and early fall: I assume many are bracing there for an antipodal winter resurgence at the same time as the epidemic might die out in the north. Hopefully, by the time we might see a second winter-spring wave of COVID19 in the Northern hemisphere, there should be a vaccine available.

(2) was the virus already in California in November? Via Erik Wingren, here is a Twitter thread by viral geneticist Trevor Bedford (Fred Hutch and U. Of Washington) that appears to debunk this theory, based on analysis of patient samples from the Seattle Flu Study. 

https://threadreaderapp.com/thread/1249414291297464321.html

“We confirmed that these samples from acute respiratory infections from Oct 2019 through Feb 2020 contained a variety of different viruses including influenza, RSV, rhinovirus, metapneumovirus and seasonal coronavirus. […S]easonal coronaviruses are responsible for ~30% of common colds and are easily distinguished from #SARSCoV2 (the virus responsible for COVID-19) in molecular assays. There is no chance of confusion between these in our assay.[…] If we restrict to viruses sampled in California (highlighted here as larger yellow dots) we see that they fall in with the rest of the US epidemic. There is no chance SARS-CoV-2 was circulating in California in fall 2019. Circulation in CA started in Jan or Feb 2020.Estimating total number of infections is difficult without serology[…] but I’d guess that we’re catching between 1 in 10 to 1 in 20 infections as a confirmed case.  

This would give 5-10 million infections in the US or about 2-3% cumulative prevalence. This is a long way from the 50% (R0 of 2) to 66% (R0 of 3) we’d need for herd immunity. I see #TestTraceIsolate as the only real solution to the problem we’re facing, alongside non-economically disruptive distancing and broad use of masks.”

(3) The “positivity rate” as a metric. This article https://www.theatlantic.com/technology/archive/2020/04/us-coronavirus-outbreak-out-control-test-positivity-rate/610132/ argues in favor of using the percentage of tested people who test positive as a metric for the severity of an epidemic. Seems a little bass-ackwards at first, since normally this will be influenced by how many test kits are available (if they are scarce, normally only people strongly suspected of being infected will get tested), but he does note an intriguing correlation between the positivity rate and the severity of an epidemic.