A proposal for isotherm world maps to forecast the seasonal evolution of the SARS-CoV-2 pandemic

This paper investigates whether the Severe Acute Respiratory Syndrome CoronaVirus 2 (SARSCoV-2) pandemic – also known as COronaVIrus Disease 19 (COVID-19) – could have been favored by specific weather conditions. It was found that the 2020 winter weather in the region of Wuhan (Hubei, Central China) – where the virus first broke out in December and spread widely from January to February 2020 – was strikingly similar to that of the Northern Italian provinces of Milan, Brescia and Bergamo, where the pandemic has been very severe from February to March. The similarity suggests that this pandemic worsens under weather temperatures between 4°C and 11°C. Based on this result, specific isotherm world maps were generated to locate, month by month, the world regions that share similar temperature ranges. From January to March, this isotherm zone extended mostly from Central China toward Iran, Turkey, West-Mediterranean Europe (Italy, Spain and France) up to the United State of America, coinciding with the geographic regions most affected by the pandemic from January to March. It is predicted that next spring, as the weather gets warm, the pandemic will likely worsen in northern regions (United Kingdom, Germany, East Europe, Russia and North America) while the situation will likely improve in the southern regions (Italy and Spain). However, in autumn, the pandemic could come back and affect the same regions again. The Tropical Zone and the entire Southern Hemisphere, but in restricted southern regions, could avoid a strong pandemic because of the sufficiently warm weather during the entire year. Google-Earth-Pro interactive-maps are provided as supplements.


Introduction
The Severe Acute Respiratory Syndrome CoronaVirus 2 (SARS-CoV-2) pneumonia -also known as COronaVIrus Disease 19 (COVID-19) -allegedly broke out in a wet market of the city of Wuhan, the capital city of the Hubei Province, in Central China (30.60°N -114.05°E). The first case of hospital admission was reported on December 12, 2019 [1]. Since January 2020 the pandemic rapidly spread throughout the whole province of Hubei, in other regions of China and, further, began to spread all over the world [2].
The high fatality rates of this pandemic requires the development of epidemic control strategies such as lock-down of the infected region and others [3], which, however, in time can negatively affect the economy of a society. Thus, it is a priority to forecast how the COVID-19 pandemic could geographically propagate for optimizing these strategies. The situation is severe and is worsening. As of the date -04/01/2020 -there are more than 860  with more than 8000 cases each. 7 See Figure 4. 8 Although the propagation of a pandemic and its health severity may have several causes, Figure 3 shows that the spread of the recent pandemic could also have some geographical preferences. In fact, it appears to have spread quite fast in moderately cold places where the daily average temperature may have been roughly between 0°C and 15°C. On the other hand, up to April 1, countries with warm climates (e.g., India, Central America, South America, Southern Asia, Africa and Oceania), as well as very cold countries (e.g. Canada, Russia, North-Est Europe) appear to have been less affected. It is also noticeable that even when the number of cases is relatively large relative to the population, the number of deaths is usually lower in warm countries, as shown in Figure 3C. Thus, it is reasonable to ask whether geographical regions within a specific climatic/weather zone could be more vulnerable to this epidemic. Indeed, a number of studies have established that influenza virus transmission and virulence depend also on meteorological conditions such as temperature, relative humidity and wind speed, and also that, in the Northern Hemisphere, influenza is more widely spread during the winter seasons [4,5,6].
In fact, influenza rate is seasonal [7] and, usually, the infection nearly disappears when weather gets warm. A similar behavior has been observed for other SARS coronaviruses [8,9,10,11] that belong to the same Coronaviridae family of SARS-CoV-2 [12]. The finding has always been that these viruses spread 7 Source: http://www.salute.gov.it/imgs/C_17_notizie_4370_1_file.pdf, accessed on 04/01/2020. 8 COVID-19 Italy situation: http://opendatadpc.maps.arcgis.com/apps/opsdashboard/index.html, accessed on 04/01/2020. mostly within given ranges of meteorological conditions. For example, Yuan et al. [13] determined that the SARS virus identified in November 2002 in Guangdong Province, China, present a peak spread at a mean temperature of 16.9°C (95% CI, 10.7°C to 23.1°C), with a mean relative humidity of 52.2% (95% CI, 33.0% to 71.4%) and wind speed of 2.8 m/s (95% CI, 2.0 to 3.6 m/s). Thus, also for the COVID-19 it has been proposed that a dry, cool environment could be the most favorable state for the spreading of the virus [14].
The seasonal, climate and weather-condition influence on the spread or slowdown of a pandemic induced by aerially transmitted viruses can have several biological, physical, solar-light mechanisms that involve both the virus survival and transmission in the air and the susceptibility of the host immune system [15]. Some of these results came from laboratory-experimental studies on how viral etiology and host susceptibility vary under different environmental conditions, while other findings are from epidemiological studies relating large-scale patterns to various climate signals and atmospheric conditions.
In this paper I take the latter approach. I compare the weather conditions from January to March 2020 between the region of Wuhan and the Italian provinces of Milan, Brescia and Bergamo. The striking similarity between the weather conditions of the two regions leads us to reason that there is an optimal weather condition that could favor the spreading of the COVID-19 epidemic. Based on this reasoning, I propose a set of optimized monthly isotherm world maps from January to December 2020 to forecast the course of the pandemic time-line evolution by identifying the geographical regions that are likely to experience similar weather conditions as in Wuhan and Milan during the high peak of the COVID-19 infection.

Temperature comparison: Wuhan versus Milan, Brescia and Bergamo
The COVID-19 epidemic curve in Hubei and the rest of China until the 3rd of March, is shown in Figure 5 and its patterns are extensively commented in Macintyre [16]. It is observed that the epidemic peaked in the first week of February and the number of new infections rapidly decreased in March. Figure 6 shows the recorded mean daily temperature of Wuhan from 01/01/20020 to 03/31/2020 (black curve). 9 The figure also shows the mean seasonal Max-Min temperature range curve from Wuhan Airport, which is 23 kilometers from Wuhan (yellow area). 10 The temperature data show that in Wuhan the daily temperature roughly ranges between 1°C and 8°C in January, between 3°C and 11°C in February and 7°C and 15°C in March. During the period mostly affected by the CONVID-19 infection (January and February), the temperature in the region roughly varied between 0°C and 12°C. The seasonal average relative humidity in Wuhan was on average around 70% and the average wind speed was around 10 km/h, which are the typical humidity and high pressure conditions that characterizes Continental China during winter, when only the weak winter monsoons blow.   daily temperature records of the Italian cities of Milan, 11 Bergamo 12 and Brescia 13 during the same months. It shows an excellent correlation between the temperature records. This temperature correlation occurred mostly for the months of January and February when Wuhan experienced a mean temperature equal to 4.1 ± 3°C and 8.4 ± 3.6°C, respectively. The Italian cities experienced a mean temperature equal to: 3.8 ± 2.0°C and 8.1 ± 2.1°C, respectively, in Milan; 4.3 ± 1.6°C and 7.7 ± 1.7°C, respectively, in Bergamo; and 3.2 ± 2.0°C and 7.2 ± 1.6°C, respectively, in Brescia.
However, Figure 8 also highlights that the three Italian cities experienced a relatively cold March with a mean temperature equal to 9.5 ± 2.8°C in Milan, 8.8 ± 2.9°C in Bergamo and 8.7 ± 2.4°C in Brescia.
In March, the daily mean temperature spanned between 4°C and 15°C. This temperature range fits well with the one measured in Wuhan in February where the mean temperature was equal to 8.4 ± 3.6°C and the daily mean temperature spanned between 3°C and 18°C. In March, Wuhan was significantly warmer: its mean temperature was equal to 12.9 ± 4.1°C and the daily mean temperature spanned between 6°C and 19°C.
The above results are summarized in Table 1 together with other meteorological indices such as monthly means of the relative humidity, wind speed and atmospheric pressure, 14  to be comparable in the two locations. It is to be noted that the observed low relative humidity (from 61% to 85%), low speed wind (from 6 km/h to 11 km/h), high atmospheric pressure (from 1016 mbar to 1026 mbar) induce atmospheric stability facilitating the spreading of virus.

In January and February the two locations shared strikingly similar weather conditions but in March
Wuhan got warmer fast while the Italian provinces experienced a cold weather similar to that of Wuhan in February. These facts could explain why the COVID-19 pandemic spread equally fast in both regions, but the Italian regions were more severely affected. In fact, as Figure 7 shows, in Italy the cold weather lasted longer with unusual cold weeks at the beginning and the end of March, favoring the pandemic spread. These considerations suggest that weather temperatures roughly ranging between 4°C and 11°C could be those that mostly favor the propagation of COVID-19 and/or aggravate the susceptibility of people to its secondary pneumonia. For the winter months of January, February and March, the isotherm maps depicted in Figure 8 show a geographical correlation with the COVID-19 pandemic patterns by country and territory shown in Figures 1 and 3.

Monthly isotherm world map analysis
In January, Wuhan gradually turns from the light-green zone to the light-gray zone as this region gets warmer.
In February, when the pandemic affected the region most severely, Wuhan is found in the middle of the light-gray zone. The light-gray zone covers the region spanning from Iran to Italy, Spain and partially covers Southern-East France and North Algeria. In these countries, the epidemic has been observed to spread fast.
In In May, the light-gray zone moves toward latitudes larger than 50°N mostly in the Scandinavian countries, Russia and Canada, as well Argentina, Chile and New Zealand.
In June, the only light-gray regions are those above 60°N latitude, the Tibet, part of Central Argentina and minor regions of Southern Australia and South Africa.
In July, the patterns are similar to those of June, although the Northern Hemisphere continues to get warm.
From August to December (Figures 10 and 11

Discussion and Conclusion
Respiratory virus infection rate is usually seasonal [8,9,10,11]. This applies also to the coronaviridae family of COVID-19 [12]. In general, there are likely several biological, physical and solar-light mechanisms that can seasonally influence the virus survival and transmission in the air, as well as the susceptibility of the host immune system [15]. Usually the weather conditions that facilitate this type of diseases include cold and dry weather, high pressure, low-speed wind and modest rain, as it happened from January to March 2020 both in Wuhan and in Northern Italy. In fact, the primary transmission route from person to person is through contact with respiratory droplets from the infected persons, generated when, for example, they blow their nose, cough or sneeze. Therefore, when relative humidity is low, as it happens in the winter in these regions, potential virus-carrying water-based droplets remain floating longer in the air because they shrink to smaller sizes, whereas high humidity or rain would facilitate their removal from air. In the same way, high atmospheric pressure reduces the wind speed, and virus-carrying droplet density could increase in the urban area. In the Northern Hemisphere, winter also has fewer hours of sun-light and UV exposure that can have a sterilizing effect [18]. In addition, cold weather usually increases the susceptibility of people to virus attacks.
Atmospheric stability related to high atmospheric pressure and low-speed winds also causes higher concentration of air pollutants [19] that could carry viruses. Air pollution increases the risk of respiratory diseases by reducing lung function making people living in particularly polluted areas more susceptible to asthma, respiratory infections, lung cancer and other diseases [20]. However, air quality real-time world-maps provided, for example, by Berkeley Earth 16 suggest that the air in regions such as China, India and Southern-East Asia is significantly more polluted than in the European and Northern America countries. Thus, the relatively low COVID-19 diffusion and mortality rates observed in the warm Asian countries relative to the high mortality levels of the colder Western European and American countries (see Figure 3) indicate that weather, not air pollution is the main cause for the diffusion of the COVID-19 pandemic.
Based on such findings, this work explored the possible link between COVID- 19   should go down as weather temperature goes up, the virus may not disappear completely. The infection rate could simply slow down, as suggested by the evidence that people get infected also in warm weather regions, although in these places the percentage of deaths per capita appears to be lower than in the cold weather regions (Figure 3). Thus, although the optimized isotherm maps proposed in the present work could be useful to optimize the timing of the required COVID-19 epidemic control policies that each country needs to implement, people and governments should be warned against lowering their guard.

Supplement
The online Supplement provides the same twelve isotherm maps shown in Figures 8, 9, 10 and 11 for each month of the year from January to December as Keyhole Markup language Zipped (kmz) files, that is, as Climate Explorer Google-Earth-Pro interactive and zoomable maps. Figure 12 shows an example of the produced maps. Google Earth Pro (used version: 7.3) 17 or equivalent Earth Viewer software is required.

Conflicts of interest: None declared.
Availability of data and material: All data are freely available on line. 17 Web site: https://www.google.com/earth/, accessed on 04/01/2020. Figure 12: Examples of isotherm Google-Earth-Pro interactive-maps provided as online Supplement files.