Climate model know-how to help assess virus spread ‘sweet spot’

“Studies have been published out of China already, looking at the potential spread of the coronavirus that causes COVID-19 and its sensitivity to temperature.” Peter Sousounis, director of climate change research, AIR Worldwide.

· Pandemic presents opportunity to understand climate impact on coronavirus
· Boost for disease modelling as firm leverages climate data expertise
· Temperature and humidity key factors in potential persistence of COVID-19
· Climate impact can be masked by inter-annual El Niños and polar vortex changes

Climate experts say finding and linking the climate change signal to the spread of virus is difficult but studies have found there seems to be a “sweet spot” in terms of temperature and the most efficient transmission.

Climate plays a key role in the potential spread of viruses, yet research that looks at both topics needs greater attention to boost global resilience, something that catastrophe risk modelling firm AIR Worldwide is working to remedy.

Parallels between understanding the impact of climate on extreme weather and the impact of climate on the spread of disease could produce more accurate models, with benefits for insurers and their customers.

Last week, the modelling firm published a projection from its AIR Pandemic Model showing that the number of mild to moderately symptomatic COVID-19 cases globally from April 1 to April 15 could range between one million and seven million. It found that the number of severe cases could range between one million and three million, with deaths ranging between 70,000 and 230,000.

“We estimate that this may represent a moderately conservative projection of cases of COVID-19; the vast majority will be asymptomatic or have mild symptoms,” said Narges Dorratoltaj, principal scientist at AIR Worldwide.

“The difference between the low and high ends of the range is driven by a few factors.”

Peter Sousounis, director of climate change research at the risk modeller, highlights climate, and potentially climate change, as being among these factors.

Sousounis, a meteorologist by education and profession, explained: “With the COVID-19 situation, I took the opportunity to better understand what the potential impact of climate and climate change is on virus transmissions.”

He said a wide variety of studies have looked at other viruses, which have a lot of caveats that “every virus is different” in terms of how it responds to climate or climate change.

“Studies have been published out of China already, looking at the potential spread of the coronavirus that causes COVID-19 and its sensitivity to temperature,” he said.

Winter spike
Sousounis added that these studies have reiterated some of the more common understandings of the seasonal dependence of a typical influenza virus for temperate latitudes (North America, Europe and most of Asia).

“It tends to spike in late winter, which is kind of what COVID-19 is doing now. The expectation is that it will become less transmissible as temperatures rise into spring and summer,” he said.

In tropical areas such as Indonesia, equatorial Africa and South America, researchers are not seeing that effect as yet. Sousounis suggested that this is probably because “the seeds of transmission haven’t been planted”, although in these regions there is no real seasonal dependency, so transmission can occur year-round.

“Other studies I’ve looked at show there seems to be a ‘sweet spot’ in terms of temperature and most efficient transmissivity of the virus,” he said.

People who grew up in the northeastern US, he said, took it for granted that there was a seasonal dependence of influenza “but still no-one really understands the reason for it”.

One hypothesis suggests that in colder months people are indoors and are more likely to transmit the disease “which is ironic because right now we are essentially sheltering indoors to try and do the opposite”, Sousounis said.

Another hypothesis, he explained, is to do with temperature and specific humidity in the atmosphere. It focuses on how rapidly the droplets which transmit the disease from one human to the next through talking, coughing or sneezing, evaporate in the air. This rate of evaporation is important because it determines how far the droplets can transfer before they land.

“That is one of the hypotheses of why the virus is transmitted more efficiently in the winter. Just enough of the droplet evaporates to make it lighter so it stays in the air longer and has a better chance of infecting people. But if it is too cold, too much of the droplet evaporates and there’s basically nothing left,” he said.

COVID-19’s “sweet spot” transmission temperature has been cautiously mentioned in studies from China as anywhere the mean temperature is below 74 degrees F (23.3 degrees C) during the month of June.

“This encompasses a huge area, and is not necessarily something we don’t know,” he said, adding that in colder regions such as Scandinavia or the northern UK the threat of the disease persisting is higher during typical summer months.

“It’s very region-dependent because it is temperature-dependent.”

Impact of climate
“There is a really good analogy between understanding the impact of climate on the spread of disease and understanding the impact of climate on extreme weather,” Sousounis explained.

“With weather there is a lot of inter-annual variability: there are El Niños and other climate signals that occur year to year, how the polar vortex changes, etc, and that can make it difficult to detect the climate change impact.

“The same thing is true with disease. For example, people’s transportation patterns change, and there’s so much else going on that it makes modelling difficult. That said, studies have shown for other transmissions of disease that follow an El Niño year—which provides a much wetter and more humid environment, especially in the tropical regions—the incidence of the disease increases by a factor of five. It’s quite a significant jump.”

Monthly temperature and precipitation are already in use in AIR’s vector disease model and, according to Sousounis, those are probably the most important weather parameters for better understanding and incorporating climate into transmission of disease models.

“The story behind COVID-19 is that a bat infected a pangolin, then the virus jumped from it to a human. The changing climate is certainly going to change the habitats where animals exist with the disease. Air Worldwide is looking at the impact of climate change on the spread of disease.”

This coronavirus—SARS-CoV-2—is clearly sensitive to climate and Sousounis is keen to conduct more research, alongside measures such as asking people to stay at home, to help “flatten the curve” of infection by reducing the rate of transmission.

“It may prolong the time over which the disease is an issue but because it will not overwhelm our healthcare system we’ll be able to treat people more successfully,” he said. “The goal is to reduce the death rate.

“Quantitatively, we’d like to look at how sensitive the virus is to things such as climate. These outbreaks are very rare occurrences, so there is not a lot of data, and the physics of the transmission is not really an exact science. Getting the climate change signal on the impact of disease spread is difficult.”

Cleaning the data so modellers can better see the climate signal will be the challenge.

“In future model releases I think we will be able to legitimately account for climate in the spread of viruses in a lot more detail,” he said.

Previous studies have looked at inter-annual variability of influenza and how it responds to inter-annual variations in temperature. For example, cold winters following warm winters have a much higher tendency for influenza spread.

“The concern for COVID-19 is: if we have a more normal or below normal temperature winter next year, what is that going to mean for the disease?

“It's not all bad news—even if it’s going to be a colder than normal winter, we are going to be so much better prepared, we may even have a vaccine. We hope the impact would be a lot less severe,” he concluded.