Irish Centre for High-End Computing Investigates possible link between Climate and Covid-19
Recent research has suggested that climate, and in particular temperature and humidity, may influence the spread of COVID-19. For example, Wang et al. (2020) [1] reported that “high temperature and high humidity significantly reduce the transmission of COVID-19”. Similarly, Sajadi et al. (2020) [2] found that COVID-19 “has established significant community spread in cities and regions along a narrow east west distribution roughly along the 30-50°N corridor at consistently similar weather patterns consisting of average temperatures of 5-11°C, combined with low specific humidity (3-6 g/kg)”. The EU Copernicus Climate Change Service has used these results to map regions where climate conditions are considered to be more conducive to the spread of COVID-19, alongside regional mortality data [3]. Other studies have indicated that regions with higher levels of ultraviolet light are associated with lower COVID-19 growth rates [4,5].
To date, the transmission of COVID-19 continues to be pronounced in regions with moderate temperature and humidity (e.g., Europe and US). Conversely, hot and/or humid regions with large populations (e.g., Thailand, Philippines, Malaysia, Vietnam, India, Africa, Australia, etc.) have, to date, reported relatively low numbers of COVID-19 cases per population. It is worth noting that a number of these countries have close trading, cultural, tourism and migration ties with China, and yet the virus has not spread to the same extent as in Europe and the US.
It is very clear that the transmission of the virus is multi-faceted, underpinned by strong social, economic and geographic drivers. Linkage with local climate may be tenuous and difficult to establish to robust standards, particularly now with the introduction of social-distancing measures, but nevertheless the weather/climate may have some role in the spread of the virus. This study looks at the evidence.
To attempt to quantify links between COVID-19 and climate, researchers at ICHEC have compared ECMWF ERA5 reanalysis climate data with national COVID-19 deaths (the methods are outlined below). The results indicate that to-date, countries with high levels of mortality have mean temperature in the approximate range of 4 to 12°C (Figure 1a). Figure 1(b) compares national COVID-19 deaths with temperature and specific humidity; the results indicate that countries with high levels of mortality have mean specific humidity ranging (approximately) from 3 to 7 g/kg .AND. mean temperatures in the approximate range of 4 to 12°C. Similarly, Figure 1(c) indicate that countries with high levels of mortality have mean relative humidity ranging (approximately) from 67 to 77% and mean temperature in the approximate range of 4 to 12°C. Figure 2 suggests a weak signal between increasing surface shortwave and UV radiation, and decreasing COVID-19 mortality. Additional work is required to establish if these weak radiation links are simply a by-product of the temperature and humidity results, or are directly related to factors such as increased immunity due to enhanced vitamin D production and/or the possible effects of radiation on the sterilization of COVID-19.
Commenting on the preliminary investigations, Dr Paul Nolan, Climate Science Programme Manager at ICHEC said,
"As scientists we are always reluctant to release results which are preliminary, on this occasion, however, the results may assist in the understanding of the effects of local climate on the spread of COVID-19 and play a small part in informing authorities tasked with implementing national strategies to combat COVID-19, particularly in the event of a second wave of infection during autumn/winter. We recommend further research of these trends nationally and internationally to validate these findings as more data become available".
It should be noted that these results:
- may very well change as the COVID-19 virus progresses globally and more reliable data become available
- have high uncertainty due to numerous external factors including national health care standards, lockdown strategies, tourism (e.g. Italy), trade, politics, population mean age (e.g. Italy), population obedience & cultural norms such as proper facemask usage, national experience/memory from previous epidemics, possible links with other immunisations in the population (e.g., BCG), and so on.
Alastair McKinstry, Environmental Programme Manager at ICHEC said,
“ICHEC will update these results as more data becomes available. This work suggests the need for a more comprehensive examination of the preliminary results indicating a link between local climate and the spread of COVID-19. Understanding this link will help to improve lockdown and confinement planning.The methods will be improved with expert health input and collaborations. Furthermore, large countries with diverse climates will need further analysis. This will be included in further work as sub-country infection data become available”.
Ray McGrath, meteorology & climate adjunct lecturer, UCD, stated,
“humidity is known to affect the survival of influenza virus in aerosols expelled through coughing. Whether the same applies to the current virus, or indeed whether other meteorological factors play a significant role in transmission, remains to be seen. Analysing the weather conditions prevailing during the current crisis is an important step in addressing this issue”.
Methods:
We compared COVID-19 data [6] with ECMWF ERA5 reanalysis [7] 2m temperature (surface), humidity (lowest pressure level) and downward radiation (surface) data. The following criteria were used:
- For each country, the mean climate fields are calculated from the date of the first confirmed case to six days ago (due to a delay with ERA5 availability).
- Only countries with population greater than one million are considered
- Only countries with cases present for at least 30 days are considered
- Only countries with at least 50 cases are considered
- the y-axis shows #deaths/ population/ (# days since first case)
All figures are valid up to 22/04/2020 (COVID-19 data) and 16/04/2020 (ERA5).
Established links between the influenza virus and climate
It is worth noting that the link between climate and viruses, such as influenza, is well established. For example, epidemiological studies indicate that low levels of specific humidity are associated with greater influenza mortality in the US (e.g., Shaman et al., 2010 [8], 2011 [9]; Noti et al., 2013 [10], Tamerius et al. 2018 [11]). Barreca et al. (2012) [12] showed that the humidity-influenza relation is nonlinear with lower specific humidity levels resulting in “in greater influenza mortality at mean daily specific humidity levels below 6 g/kg” and “incremental changes in humidity do not significantly affect influenza mortality when mean daily specific humidity exceeds a 6 g/kg threshold”. Similar links between UV radiation and influenza are well established, e.g., Sagripanti et al. (2007) [13] found that “inactivation of viruses in the environment by solar UV radiation plays a role in the seasonal occurrence of influenza pandemics”.
References:
[1] Wang, J., Tang, K., Feng, K. and Lv, W., 2020. High temperature and high humidity reduce the transmission of COVID-19. Available at SSRN 3551767. https://papers.ssrn.com/sol3/Papers.cfm?abstract_id=3551767
[2] Sajadi, M.M., Habibzadeh, P., Vintzileos, A., Shokouhi, S., Miralles-Wilhelm, F. and Amoroso, A., 2020. Temperature and latitude analysis to predict potential spread and seasonality for COVID-19. Available at SSRN 3550308. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3550308
[4] Backer, A., 2020. Slower COVID-19 Case and Death Count Growth at Higher Irradiances and Temperatures. Available at SSRN 3567587.
[5] Merow, C. and Urban, M.C., 2020. Seasonality and uncertainty in COVID-19 growth rates. medRxiv.
[6] https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-cases
[7] C3S (Copernicus Climate Change Service), 2017. ERA5: fifth generation of ECMWF atmospheric reanalyses of the global climate. Copernicus Climate Change Service Climate Data Store (CDS). https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5
[8] Shaman J, Pitzer VE, Viboud C, Grenfell BT, Lipsitch M (2010) Absolute humidity and the seasonal onset of influenza in the continental United States. PLoS Biol 8:e1000316.
[9] Shaman J, Goldstein E, Lipsitch M. (2011). Absolute humidity and pandemic versus epidemic influenza. Am J Epidemiol 173(2):127–135. doi:10.1093/aje/kwq347
[10] Noti, J.D., Blachere, F.M., McMillen, C.M., Lindsley, W.G., Kashon, M.L., Slaughter, D.R. and Beezhold, D.H., 2013. High humidity leads to loss of infectious influenza virus from simulated coughs. PloS one, 8(2).
[11] Tamerius, J.D., Shaman, J., Alonso, W.J., Bloom-Feshbach, K., Uejio, C.K., Comrie, A. and Viboud, C., 2013. Environmental predictors of seasonal influenza epidemics across temperate and tropical climates. PLoS pathogens, 9(3)
[12] Barreca, A.I. and Shimshack, J.P., 2012. Absolute humidity, temperature, and influenza mortality: 30 years of county-level evidence from the United States. American journal of epidemiology, 176(suppl_7), pp.S114-S122.
[13] Sagripanti, J.L. and Lytle, C.D., 2007. Inactivation of influenza virus by solar radiation. Photochemistry and photobiology, 83(5), pp.1278-1282.
