Hacker News new | past | comments | ask | show | jobs | submit login
High Temperature and High Humidity Reduce the Transmission of Covid-19 (ssrn.com)
162 points by umeshunni on March 13, 2020 | hide | past | favorite | 96 comments



But not by enough. This epidemic has a growth rate (R) of at least 2. The article claims "One degree Celsius increase in temperature and one percent increase in relative humidity lower R by 0.0383 and 0.0224, respectively." The March to July difference for New York City is only 19C. San Francisco, 9C. For both cities, humidity is relatively uniform year round.

Even assuming this result is meaningful, it still leaves an R way above 1.


It's important to understand the limitations of the paper. the paper simply plotted a set of regions where the outbreak happened, and analyzed the growth rate in these regions vs the average climate in that region. Notably, it did not include areas where there was no growth (which coincidentally, are almost universally hot or tropical regions). If those were included, you'd have seen a much larger effect.

With that said, even with just including areas with the epidemic - it found a statistically significant effect. Just, as you note - not a significant effect (which doesn't mean that a significant effect might not exist)


When cases in northern Italy got to the hundreds, German news were full with the position has been that Coronavirus was like influenza with a strong seasonal component. This was because the national plan for the disease included that idea. However, since Monday March 9th, they have adjusted it. This [1] is a (german-language) statement from Monday by a scientist involved in that effort. He mentions a study from the USA, but doesn't give any details that make the paper easily findable. The only study I could find was this [1] one. It studies the same topic and has a similar conclusion, but might not be the one that the change was based on. Relevant part of the abstract:

> Here, we examine province-level variability of the basic reproductive numbers of COVID-19 across China and find that changes in weather alone (i.e., increase of temperature and humidity as spring and summer months arrive in the North Hemisphere) will not necessarily lead to declines in COVID-19 case counts without the implementation of extensive public health interventions.

[1]: https://www.youtube.com/watch?v=DyK9PSsnCI8

[2]: https://www.medrxiv.org/content/10.1101/2020.02.12.20022467v...


> For both cities, humidity is relatively uniform year round.

There is no way humidity is relatively uniform year round in NYC. It gets very humid during the summer and very dry during the winter.


The relative humidity in NYC does not vary much from 60-65%. Absolute humidity is measured as grams water per cubic meter and is virtually never reported because what is relevant to human comfort is evaporation.

Indoor humidity in the winter is generally low because increasing the temperature without adding moisture is lowers relative humidy.

https://weather-and-climate.com/average-monthly-Humidity-per...


That's indoors. Outdoors, both cities are next to oceans, which tends to stabilize humidity.


In many parts of China, it will be 30C hotter and 50%p more humid in the summer than it is now. The combination might reduce R to virtually nothing.

In Japan, this could mean that it was the right decision to schedule the Olympics in the middle of summer after all.


I'm pretty sure that the Olympics will be cancelled. It's unlikely that we'll have a vaccine before then and Covid19 is well beyond containment at this point.


Abe is dead set on hosting the Olympics one way or another, so he might try to reschedule it... to a time of year when R is higher. :(


There will be all kind of athletes ill. Especially sucks if you're part of a team. I mean, for individuals as well, but as team you rely on each other's strengths even more. Gonna be interesting for sure.


The study uses relative humidity, which is affected by the dew point. While straight humidity is often consistent at least in the US, relative humidity fluctuates wildly. Relative humidity is what gives the muginess feel. Most of the US except the Rockies and most of the West Coast have way higher relative humidity in the summer months. NYC can get especially bad (or good in this case). In the South, that increase has already started.

Admittedly, it's not a huge affect, but taking your NYC example, the temperature alone will decrease the R0 by at least .5. I'll take that.


There are plenty of places that are hotter all year round, or have a larger temperature+humidity difference between seasons. Or was your comment only intended to address large US cities?


I think this serves to better explain the low case totals out of africa, but that might change in the near future.


Also, India which is not yet mentioned in the news, a country of 1+ billion people.


Also in Myanmar which shares the same border as China, 0 cases have been confirmed so far.


Why? Do you realize Africa is a huge continent with all kinds of weather right? https://www.timeanddate.com/weather/?continent=africa


Yes, but the average temperature in vast majority of the areas is higher than the avg temperature of say Europe. When we say Africa is warmer, it is a simplification simply by the virtue that its centroid closer to equator than Europe. Thus, we can say that it's warmer.

Side note: don't use rhetorical questions to respond to someone. It's demeaning. Rephrase "Why? Africa is a huge continent with all kinds of weather."


It's not demeaning, it actually adds color to the answer and makes it more engaging. I don't know what's up with HN's fetishizing of impassionate, humorless and snarkless style. Not everything has to be written like the IEE 754 specification.


Snarky comments explicitly discouraged in the official guidelines, which you can find at the bottom of this page.


You can be passionate and nice to people. This was mean.



Those temperatures are normalized to sea-level. If you go to say La Paz or Macchu Pichu expecting scorching weather based on that graphic you are going to have a bad time.


I am eagerly awaiting more studies into this subject. Something doesn't feel right about the relative lack of local transmission in e.g. Vietnam (temperatures over 30C at least in the south right now) compared to e.g. Italy (temperatures, what, 15C or lower in the north right now?).

Cambodia's extremely low case count can easily be explained away by them not testing anyone, but Vietnam appears to be quite proactive with tracing and testing, yet still has a relatively low count and minimal transmission.


Vietnam has taken this epidemics 10x more seriously than Italy, that could just be the difference.

They are testing as much as they can, everybody who had contact with infected people are tracked down and quarantined, streets where you had infected people are closed down and disinfected, almost everybody wear masks, you have temperature checks & hydroalcoholic gel available in public areas, schools are fully closed since early January.


Yes. However, there's also no travel restriction between Cambodia and Vietnam, significant cross-border traffic, and Cambodia is not "taking it nearly as seriously" as Vietnam is.

If Vietnam had the transmission going on that Italy does, the case-tracking and quarantining would be a lot harder, and I just don't buy that the other things (masks, forehead temperature checks[0], hand sanitizer that hardly anyone knows how to use properly) have much of an impact.

There has to be some much larger factor than those at play (and this study suggests what that factor could be).

[0] I encourage people to ask or look what the reading is when getting these checks. About a third of the time when I get a forehead temperature check, the reading is totally nonsensical, like 31C, but the person just waves me through since it's under the threshold. I'm skeptical of the efficacy of a test that is this inaccurate.


I live in Vietnam, maybe you too. It is true that Government here is being very restrictive, controlling flights, quarantine people, etc... But its true that during early January, still many people coming from China came to Vietnam, also people comng from other south east asians. During second half of february there were no restrictions for visa exemption for tourists, thousands came from Spain, Italy, Germany, UK... What I mean if this was Italy, Austria, Norway, the virus would have been unstoppable. We would have case everywhere , thousadns, same should be Singapore, Thailanda, laos, etc... Its very clear that the virus can be transmited here? YES, of course but only thorugh very direct contact, what is that is happening. I am sure the weather played a big role and virus is not stable for long time when exposed. I am sure also, that all this here is creating a huge oportunity of propaganda for the socialist government, they make popultaion feel fear all the time, all the people here think they will die if they catch the virus so they see like sheeps the government as an angel. They are taking too strict measurements and the damage here economically speaking is already HUGE. An example how to manage this is Singapore. They always had imported cases, and few direct contact cases from this imported cases. But they still run the country as normal as possible with not too strict measures. Here they are literally crazy. Of course, personal opinion, cheers


What doesn't feel right?

The abstract states that high temperature and high humidity reduce transmission rates relative to colder climates, which corresponds with the high transmission rates during the northern hemisphere winter and early spring.

And yes, lack of testing has resulted in some eyebrow raising numbers.

Up in Da Nang now, both locals and foreigners are aware and taking it seriously -- scary times.


Sorry if I was unclear. I was trying to say that it seems clear to me intuitively that something other than "taking it seriously" is reducing the transmission rates in VN. It's just a "feeling" -- but this study seems to provide some (thin) evidence supporting it.



Really hoping this helps in India which so far has had really low reported cases.


Sure, but what about when it gets cold again? It may come back even worse


By then, we should have enough testing kits, and authorities should be better prepared to handle it.


Not to mention a better baseline immunity within the population, and more preparedness as a whole.

Testing kit shortages, lack of information, no vaccine, etc. are all a function of how little time we've had to respond. If we can buy 3 months, that's a lot of bonus time to prepare.


Europe had about 3 months compared to China and is handling the situation much worse.


I think this thing is so contagious that all the weak people will already be dead.

The second wave might kill a few of the survivors that had their immune system weakened since the first wave, but it shouldn't be as much people as the first time.


Surely surviving the first wave will be a plus against a second wave as you already have some immunization?


IF this is the case, we should be way more prepared including possibly having a vaccine available.


Still no vaccine for SARS 18 years later, what makes you think we'll have a vaccine for this?


SARS hasn't been seen in 15 years. Exactly whom did you expect would spend the time and money to develop a vaccine for it?


Maybe we should develop responses to pathogens that have the ability to wipe humanity off the galaxy.

Given how close SARS and SARS-CoV-2 is, maybe the SARS vaccine could be applied to the current case.


> Maybe we should develop responses to pathogens that have the ability to wipe humanity off the galaxy.

Not a bad idea, but isn't that a near-infinite set?


I think that's likely, but it's understandably difficult to convince anyone to spend a few hundred million dollars to develop vaccine that might never be used.


Seems like a pretty good deal right now. Our economic system has structural pathologies that make it extremely difficult to meaningfully organize whole populations.


Yeah, ok, I think I agree with this. I have to believe that a sustained effort to develop coronavirus vaccines would bear fruit, whether actual vaccines or merely processes and knowledge, even if the particular motivating disease was no longer in circulation.


It can and will. For that matter, it's an RNA virus that is morphing as a function of the error factor of replication, so vaccines may be ineffective (for the same reason we continue to get the "common cold" again and again over our lives, as well as the flu -- the signature keeps changing).

We will see.

Though it's worth noting that there are a lot of very contrary signals on whether it attenuates with heat. Some find that it doesn't, while some do. The very origin of this -- Wuhan -- had very temperate conditions when the outbreak began, and transmission seems to be much worse in the US than in Canada.


Maybe that's why Singapore hasn't been badly impacted?


Singapore is included in how they determined this result, therefore you shouldn't use this result to explain why singapore is doing better.


No, Singapore is not in China and is thus not in the set of "100 Chinese cities". It's only mentioned in the paper in graphs of temperature and humidity of a whole bunch of countries.


Ah fair enough. I guess we'll have to observe areas with similar climates and see if it holds. If so, I'm taking that tropical cruise after all. ;)


Here's the study on arxiv [1], if you're more inclined to use that system.

[1] https://arxiv.org/ftp/arxiv/papers/2003/2003.05003.pdf


This isn't peer-reviewed.

Maybe I have bad information, but cold plus high humidity caused it to persist on fomites longer (9+ days). Wouldn't heat plus low humidity be better than heat plus high humidity? What is/are the breakdown mechanism(s)?


High humidity reduces aerosol transmission, which is the more important vector.

(By essentially washing down the vapor and additionally making people exhale less water. At least that's the theory behind it.)


Spain seems to be an exception to this theory.


Madrid has cold weather up until Tuesday - https://www.accuweather.com/en/es/madrid/308526/march-weathe...

It’d be interesting to see if the warmer weather since Tuesday will reduce the number of new cases. If we assume a week between exposure to getting diagnosed, we would start seeing lowering rate of new infection starting from mid next week.


23C, thought it would be warmer now, what about Thailand?


Spain is not hot at all this time of year.


They say it “reduces” the spread. There may be other factors which affect the situation in Spain. Also it’s not that hot now in most places there ;p


Weather in Spain in the en of february is cold most fo the times. March is a changeable weather, can have very cold weeks, some few days with 25C and very sunny. But cold also. May will be a hot month.


I checked with a bunch of biologists and all of them find the results here to be merely correlational and misleading. Have asked to wait for more evidence


Excuse my ignorance but would a colder temperature but with rain help prevent due to the high humidity or does it require heat AND humidity?


In the developed world a lot of people spend time indoors with air conditioning most of the summer anyway? Especially more frequently now thanks to climate change. Not sure how this changes a lot once it’s circulating already?


Taiwan is doing good in containing Covid-19 despite its proximity to China


Looking at figure 4 in the paper gives me some comfort since the majority of the population centers in the northern hemisphere could approach a R of 1 in July.


This sounds great, but I don't know anything about SSRN. Can someone shed light on the credibility here?


same as arxiv


Have the authors considered increased sunlight/UV radiation in warmer areas?


105 data points (patient pairs). Linear regression. Not conclusive.


To be pedantic, COVID-19 is the disease, SARS-CoV-2 is the virus. It's like HIV vs AIDS.


If you reduce the transmission of the virus, do you not reduce the transmission of the disease?


I mean, that's why it's pedantic.

Technically you can't "transmit" the disease. The disease consists of the effects on the body.

Also, you can transmit the virus without having the disease, which is what makes it dangerous since you have no idea. It's like you can transmit HIV without having AIDS for years:

>Acquired immunodeficiency syndrome (AIDS) is defined in terms of either a CD4+ T cell count below 200 cells per µL or the occurrence of specific diseases in association with an HIV infection

That doesn't happen for a long time (or today, for a lot of people, never).


Well, fair. My assessment was based on the assumption that the distribution of outcomes for a patient isn't affected by the source of the infection. If you had some hypothetical way of transmitting the virus only to hosts who would then develop symptoms of the disease, I guess you could argue that you could reduce transmissions of the virus while maintaining transmissions of the disease.

That'd be impressively evil though.


This study fits a two-variable linear model to a high-dimensional nonlinear process.


Exactly how does one take this body of scientific knowledge and act on it?


this abjectly fails to deal with heteroscedasticity


How is the virus doing in Singapore, an island on the equator?


Here's a likely related paper from 2007:

https://jvi.asm.org/content/82/4/1899

In this study, a human immunodeficiency virus (HIV)-based pseudovirus system was employed to address these issues. Our results indicated that the SL-CoV S protein is unable to use ACE2 proteins of different species for cell entry and that SARS-CoV S protein also failed to bind the ACE2 molecule of the horseshoe bat, Rhinolophus pearsonii. However, when the RBD of SL-CoV S was replaced with that from the SARS-CoV S, the hybrid S protein was able to use the huACE2 for cell entry, implying that the SL-CoV S proteins are structurally and functionally very similar to the SARS-CoV S. These results suggest that although the SL-CoVs discovered in bats so far are unlikely to infect humans using ACE2 as a receptor, it remains to be seen whether they are able to use other surface molecules of certain human cell types to gain entry. It is also conceivable that these viruses may become infectious to humans if they undergo N-terminal sequence variation, for example, through recombination with other CoVs, which in turn might lead to a productive interaction with ACE2 or other surface proteins on human cells.

Look at the authors:

Zhengli Shi

https://en.wikipedia.org/wiki/Shi_Zhengli

She is a researcher at the Wuhan Institute of Virology (WIV), which is part of the Chinese Academy of Sciences (CAS). Shi and her colleague Cui Jie found that the SARS virus originated in bats.


Can someone with a better understanding of molecular biology and virology explain what this means? I don't want to jump to conclusions.


The paper describes how virologists in Wuhan engineered a "SARS-like" virus -- previously limited only to bats -- specifically to make it infective in humans via the ACE2 receptor. That's the same receptor used by the SARS-CoV-2 bug we're fighting now.

Note that the paper was submitted 12 years ago. They've had a lot of time to work out the kinks, if they were so inclined.

EDIT:

I want to repeat what I've said in other comments, because what I initially said was inaccurate. These researchers didn't actually create a new coronavirus, but rather a much simpler and safer virus purpose-built to test some receptor-binding structures.

This is a well-established way to study viruses safely, and this kind of work is critical if we're to learn how to develop vaccines faster in response to epidemics.


Why on earth would they do this? What sort of application does this have or serve?

If it's a bioweapon, why was it published?

Is there any concrete genetic or documented evidence to show COVID-19 is engineered? This is a big claim and has severe ramifications for China, if true.


According to the paper, they did it in order to figure out why SARS-CoV was able to infect humans, while numerous other SARS-like viruses are not. Their hyposthesis was that the critical feature is a particular sequence in the protein found on the "spike" that the virus uses to bind to host receptors. By constructing a new SARS-like virus with the desired spike protein, they were able to validate that hypothesis.

It seems to me that if there was any malign intent behind this work, they would not have published a paper on it. Obviously, that's doesn't preclude the possibility of an accidental release from the same lab twelve years later. As far as I know, there's no way to prove whether 2019's coronavirus was engineered in any way.


I want to repeat what I've said in other comments, because what I initially said was inaccurate. These researchers didn't actually create a new coronavirus, but rather a much simpler and safer virus purpose-built to test some receptor-binding structures. This is a well-established way to study viruses safely.

This kind of work is critical if we're to learn how to develop vaccines faster in response to epidemics.


From

> "The evidence we have is that the mutations [in the virus] are completely consistent with natural evolution".[19] Bedford further explained, "The most likely scenario, based on genetic analysis, was that the virus was transmitted by a bat to another mammal between 20–70 years ago. This intermediary animal—not yet identified—passed it on to its first human host in the city of Wuhan in late November or early December 2019" [1]

"20-70 years ago"! [2] Also important: "was transmitted by a bat to another mammal". That could imply it wasn't from bat soup or bats directly.

[1] https://en.wikipedia.org/wiki/Misinformation_related_to_the_...

[2] https://archive.is/X3VGB


SARS infected 8000 humans in 2002. It was a coronavirus. (That's why the current one gets the "-2" suffix.) Nature beat the researchers by a number of years (and they were presumably trying to figure out how).


According to the paper, you're exactly right.


What I said above isn't entirely accurate. The virus that the researchers created wasn't actually a coronavirus. It was a pseudovirus based on HIV, built to express the functional spike proteins of interest. This was obviously much safer than creating a highly infective coronavirus, which I doubt anyone could have done back then anyway.


Wow. Quite an eerie find. Where did you come across it?


Prof. SHI Zhengli elected a fellow of the American Academy of Microbiology [1]. Seems to be a widely respected scientist in this specific field.

[1] http://english.whiov.cas.cn/ne/201903/t20190308_206697.html


[flagged]


It is extremely frightening, and it's important that we continue to fight for freedom of expression on all platforms.

Unfortunately so many have an innate bias against the US that any criticism of it is considered fair game regardless of how conspiratorial or absurd.


> Meanwhile, you can freely conspiracy theorize all you like about the machinations of the CIA, NSA, etc., regardless of how little evidence you have

That's not true at all. Rather, there's randomness around what gets flagged and/or moderated. People read into that whatever patterns they want to see, or are already convinced are happening.


It seems like you want me to jump to a conclusion, but all I see is a researcher that studied a virus that has potential to jump to human hosts, and that has a natural reservoir in the area that the do their research.


No, they didn't just study it. They engineered a variant of a SARS-like coronavirus specifically to be able to infect humans. The paper's abstract says they did this by "combining a human immunodeficiency virus-based pseudovirus system with cell lines expressing the ACE2 molecules of human, civet, or horseshoe bat."


Yes, that's how one studies these things. The models we have in simulation are far too crude, so scientists are forced to make the real versions. I have a friend who does this with tuberculosis to study how it develops resistance to drugs. Part of it was trying to enhance or block certain genetic pathways, and yes, sometimes it results in tuberculosis strains with much higher adaptivity than normal. This is how we learn what the pathogen uses, and, hopefully, how to stop it, or at least how not to make the situation worse.


Hey man, no argument from me; I appreciate the need to study deadly diseases in the flesh. I just wonder if they could maybe locate their biosafety-level-4 lab somewhere other than in a metropolis of eleven million people. You know, just in case.

For what it's worth, one of the names on that paper is Shi Zhengli, one of the lab’s top researchers who is now defending the lab to the media:

[Shi Zhengli] said on her social media account that she “guaranteed with her own life” that the outbreak had nothing to do with the lab but was a “nemesis for the barbaric habits and lifestyle of some people – like eating wild game including bats.”

https://asiatimes.com/2020/02/coronavirus-lab-leakage-rumors...

EDIT:

As I noted in my other comment, these researchers didn't actually create a new coronavirus, but rather a much simpler and safer virus purpose-built to test some receptor-binding structures.


The ME in MERS stands for Middle East, since outbreaks of it are always from camels. Such places are some of the hottest on the planet. MERS is another form of coronavirus.

It is foolhardy to think spring and summer weather will save us.


It's a related but different virus. Also, the environment in the Middle East tends to be arid.


MERS evolved to spread in camels, SARS evolved to spread in bats. Their preferred climate would obviously be different.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: