Research shows there are more COVID-19 cases as temperature and humidity fall. A recent study points to more severe cases in cold and dry weather. Do these findings suggest COVID-19 is seasonal? Experts disagree.

A couple walks in the fall to accompany an article about weather and COVID-19.Share on Pinterest
Some research suggests COVID-19 may be more prominent in cooler temperatures.

Coronavirus data

All data and statistics are based on publicly available data at the time of publication. Some information may be out of date. Visit our coronavirus hub for the most recent information on COVID-19.

Was this helpful?

Why are these findings so controversial, and why has the United States seen most cases during its hot and humid summer? In this special feature, we investigate which weather conditions are most associated with COVID-19 cases.

We look at what factors might confound these studies and make them hard to interpret. And we describe how one international study tries to get around these problems.

There are good reasons to expect a respiratory virus to show seasonal variation. Infections from influenza and respiratory syncytial virus are more common during winter in temperate areas of the world.

“But the fact is that respiratory viruses are generally seasonal, probably as viruses that transmit on water droplets do so less well if the droplet dries up faster, and temperature, humidity, and UV may be part of the lull in transmission we are now seeing. The flip side, alas, is that the opposite will be true in the autumn and beyond.”

– Prof Ian Jones, Professor of Virology, University of Reading, United Kingdom

Studies of the first SARS-CoV in 2003 suggest weather might be important for coronavirus spread. While this virus did not circulate long enough to establish any potential seasonal pattern, daily weather was associated with the number of cases. In Hong Kong, new cases were 18 times higher in lower temperatures — under 24.6°C, 76°F — than more elevated temperatures.

The epidemic died out during a warm, dry July in 2003, but tight public health control measures were also in place. A recent review of the seasonality of respiratory infections describes how cold, dry winter weather makes us more susceptible to viruses in general.

In these conditions, the mucous lining in our noses dries up, which in turn impairs the function of cilia, the tiny hairs that line the nasal passage. These beat less often, meaning they may fail to clear viruses from the nose. The review concludes that a relative humidity of 40–60% might be ideal for respiratory health.

Americans spend 87% of their time indoors, so how does the outside weather affect them so much? When cold, dry air meets warm air from indoors, it reduces the air’s humidity inside by up to 20%. During winter, indoor humidity levels are 10–40%, compared to 40–60% in fall and spring. The lower humidity aids the spread of virus aerosols and could make the virus more stable.

Coronavirus resources

For more advice on COVID-19 prevention and treatment, visit our coronavirus hub.

Was this helpful?

Laboratory and observational studies of cases of COVID-19 patients show an impact of humidity on the SARS-CoV-2 virus.

A laboratory-generated aerosol of SARS-CoV-2 was stable at a relative humidity of 53% at room temperature, 23°C, 73°F. The virus had not degenerated much even after 16 hours and was more robust than MERS and SARS-CoV. This helps explain its higher levels of airborne infectivity.

Laboratory studies do not necessarily predict how a virus will behave in the real world. However, a study of 17 cities in China with more than 50 cases of COVID-19 found a link between rises in humidity and reductions in COVID-19 cases.

The team measured humidity as absolute humidity, or the total amount of water in the air. For every gram per cubic meter (1 g/m3) increase in absolute humidity, there was a 67% reduction in COVID-19 cases after a lag of 14 days between the humidity increase and the number of cases.

Experts report similar associations between the number of cases and humidity in Australia, Spain, and for both cases and deaths in the Middle East.

The way temperature and humidity interact provides different weather patterns, which are determined by latitude.

A comparison of climate data looked at eight cities with high levels of COVID-19 spread:

  • Wuhan, China
  • Tokyo, Japan
  • Daegu, South Korea
  • Qom, Iran
  • Milan, Italy
  • Paris, France
  • Seattle, U.S.
  • Madrid, Spain

These cities were compared with 42 other cities worldwide with a low COVID-19 spread. All of the first eight cities lay in a narrow band between 30°N and 50°N latitudes.

Between January and March 2020, the affected cities had low mean temperatures of 5–11°C, 41–52°F, and low absolute humidity of 4–7 g/m3. The authors conclude these findings are:

“consistent with the behavior of a seasonal respiratory virus.”

Studies of influenza show tropical areas where rainfall drives humidity have a higher transmission in humid-rainy conditions.

American researchers established a threshold of 18–21°C (64–70°F) and specific humidity below 11–12 g/kg, approximately equivalent to 13–14 g/m3, for increased winter transmission. Tropical countries with temperature and humidity levels above these had higher influenza transmission when rainfall was high, defined as greater than 150 mm per month.

Brazilian researchers looked at rainfall worldwide, and confirm COVID-19 cases also increase with greater precipitation. For each average inch per day of rain, there was an increase of 56 COVID-19 cases per day. No association was found between rainfall and COVID-19 deaths.

Links between COVID-19 cases and temperature are less certain. Studies from China have both found and not found an association with environmental temperature.

Similarly, researchers report no effect of temperature and COVID-19 transmission or deaths in Australia, Spain, and Iran.

However, higher temperatures are associated with a lower number of cases in Turkey, Mexico, Brazil, and the U.S., but it appears there is a threshold. Higher temperatures do not cause a further decline in COVID-19 transmission, which could account for some of the disparities.

This is consistent with laboratory studies that show the SARS-CoV-2 virus is highly stable outside the body at 39.2°F (4°C) but increasingly unstable at temperatures above 98.6°F (37°C).

A study in Spain found after 5 days of lockdown, the longer the hours of sunshine, the more cases there were of COVID-19. This positive association held true with a lag — between sunshine hours and cases — of 8 and 11 days. There was no link between the hours of sunshine before lockdown and during the first 5 days.

This contradicts findings from influenza research, which suggests lower transmission with longer hours of sunshine. The authors say:

“The positive sign of sunshine may well be another instance of behavioral adaptations, whereby compliance with lockdown orders weakens on sunny days.”

In contrast, there appears to be no effect of solar UV light, as the wavelength required to kill viruses and bacteria is under 280 nanometers (nm).

This type of UV light (UVC) does not reach Earth as it is absorbed in the ozone layer. If it did reach Earth, humans would suffer severe burns to their skins and eyes within minutes.

Some minor effects of UVB light, defined as 280–320 nm, have been proposed to explain the contradictory findings of lower transmission of COVID-19 in cold and dry conditions at a higher altitude. However, other factors, such as higher vitamin D levels within people in these regions, might be more important.

“This virus demonstrates no seasonal pattern as such so far. What it clearly demonstrates is that if you take the pressure off the virus —the virus bounces back. That’s the reality, that’s the fact.”

– Dr. Michael Ryan, WHO press briefing 8/10/2020 @ 20:51 mins

Researchers in Oxford, United Kingdom list reasons why people should not use observational studies on the number of COVID-19 cases and associated weather conditions to establish the seasonality of COVID-19 transmission.

They argue that testing capacity has been a major problem in most countries, which means there are many more cases than are reported.

Therefore, any factor linked to the weather and increased chances of testing could make it seem like the number of cases was due to weather, while increased testing is simply driving the numbers.

For example, other respiratory illnesses are common in winter months and could prompt people to have a test for COVID-19. Milder cases will be identified, which would not have come to light without another virus’s respiratory symptoms.

Furthermore, other conditions, such as cardiovascular diseases, are more common in cold weather. Patients who present at the hospital are more likely to be tested, which leads to further identification of cases. However, these would be related to other conditions linked to the weather and not necessarily COVID-19.

Nevertheless, COVID-19 deaths are less likely to be confounded by testing capacity since those with severe symptoms are expected to attend the hospital independently of the weather. Some studies above report an association between deaths and changes in the weather.

During a pandemic, a new virus will spread rapidly through a population where no one has immunity. The National Academies of Sciences, Engineering, and Medicine state in their consensus report on COVID-19 transmission there have been:

“[Ten] influenza pandemics in the past 250-plus years—two started in the northern hemisphere winter, three in the spring, two in the summer and three in the fall. All had a peak second wave approximately 6 months after the emergence of the virus in the human population, regardless of when the initial introduction occurred.”

Researchers at Princeton University and the National Institutes of Health, Bethesda, have modeled the spread of SARS-CoV-2 in relation to the weather using data on two beta coronaviruses, similar to SARS-CoV-2, which usually cause the common cold.

They found pandemic transmission in the community was likely to be so strong it would negate the minor effects of weather changes, such as higher temperatures and humidity.

The model explains why some countries with weak public health control measures, such as avoiding close contact, closed spaces, and crowds, and where this is not possible, wearing masks, are showing high transmission rates in the hot and humid summer conditions.

“As Rachel [Baker] argues in the paper, there is likely a seasonal impact on transmission, but given the high rate of susceptible people, it was unable to suppress transmission. The current outbreak in the US would likely be worse if we didn’t have the weather on our side, and is likely to get worse going into the fall and winter, assuming everything else stays the same. Once we have enough people who are vaccinated or recovered, we will probably see small, seasonal outbreaks of COVID-19 returning every winter, similar to colds and the flu.”

– Marta Shocket, PhD, Postdoctoral Fellow, UCLA, personal communication 8/12/2020

To overcome the problem of non-weather factors that confuse the picture of seasonality and COVID-19, an international group of researchers has analyzed the severity of COVID-19 instead of the number of cases.

Using data from admissions to six European hospitals and 13 hospitals in the Zhejiang province in China, they found decreases in deaths, the average length of stay, and admission to intensive care units for COVID-19 with each additional day of the pandemic.

This was found in most of the European hospitals, but not the Chinese hospitals. China’s pandemic rise took place entirely during winter, while in Europe, COVID-19 spread throughout the winter and spring months.

Deaths decreased in the European hospitals with each unit of temperature increase but not in the Chinese hospitals. The authors disregarded improvements in treatment during February and July, citing only a small impact from the use of dexamethasone.

They hypothesize the decrease in severity is related to humidity-driven changes in nasal mucous and viral clearance by nasal cilia.

The findings of decreasing severity were corroborated in their U.S. and UK data set of four million citizens self-reporting symptoms associated with COVID-19. Over 37,000 people had a symptom cluster with a close correlation to positive COVID-19 testing. There was a similar decrease in symptom duration across the course of the study.

What this study means

This research is a preprint and observational study. Therefore, it cannot establish causal links, but it does go some way to overcome the potential confounding factors in earlier weather and COVID-19 case studies.

If COVID-19 is seasonal, experts will likely establish this in 2021 or 2022 after the main pandemic waves.

In the meantime, the authors suggest the importance of considering hydration for patients and the public, including the ancient practice of nasal irrigation.

“… providing humidified air to patients in the early stages of the disease may be beneficial. [and] … in the situation of rapidly progressing COVID-19 pandemics it would be essential to actively promote universal humidification of dry air in all public and private heated spaces, as well as active nasal hygiene and hydration.”

For live updates on the latest developments regarding the novel coronavirus and COVID-19, click here.