Upgrade climate and COVID-19
According to specialized literature, air temperature (Ta), relative humidity (RH), and absolute humidity (AH) are the main climatic and meteorological parameters to address the influence of climate on the proliferation, transmission, incidence, survival, and variations of the SARS-CoV-2 (Ahmadi et al., 2020; Bukhari and Jameel, 2020; Casanova et al., 2010; Caspi et al., 2020; Liu et al., 2020; Ma et al., 2020; Wang et al., 2020; Xie and Zhu, 2020). The emphasis of these studies has been placed on the establishment of thresholds, or climatic optimum, extracted from statistical associations between meteorological parameters measured in meteorological stations (outdoor environment) or controlled in laboratories (indoor environment) and confirmed cases of COVID-19 (and other types of Coronavirus) (Table 2).
Table 2
Climate and COVID-19 – Literature review until April 15, 2020.
References | Type article | Country | Tested environment | Range | Target Virus | Outcome |
Temperature | Relative humidity | Absolute humidity |
Tan et al., 2005 | Published | China, Taiyuan | Outdoor environment | 16–28 °C | -- | -- | SARS | The optimum environmental may encourage virus growth. There is a higher possibility for SARS to reoccur in spring than that in autumn and winter. |
Casanova et al., 2010 | Published | China | Laboratory | 20 than at 4 °C | 20–50 - 80% | -- | SARS-CoV | For temperature, the viruses were inactivated more rapidly on surfaces and all humidity levels. |
Van Doremalen et al., 2013 | Published | USA | Laboratory | 20 °C | 40% | -- | MERS-CoV | MERS-CoV was more stable at low temperature and low humidity conditions. |
Xie & Zhu, 2020 | Published | China | Outdoor environment | < 3.0 x > 3.0ºC | -- | -- | SARS-COV-2 | The mean temperature has a positive linear relationship with the number of COVID-19 cases with a threshold of 3 °C. |
Chan et al., 2011 | Published | China | Laboratory | 22–25.0 ºC | 40–50% | -- | SARS-COV | Low temperature and humidity may facilitate COVID-19 transmission. |
Bukhari et al., 2020 | Preprint | USA | Outdoor environment | 3–17 ºC | - | 3–9 g/m3 | SARS-COV-2 | COVID-19 would not spread in warm humid regions. |
Sajadi et al., 2020 | Preprint | USA | Outdoor environment | 5–11 ºC | 3–6 g/kg | 4–7 g/m3 | SARS-COV-2 | Established significant community spread in cities and regions along a narrow east-west distribution roughly along the 30-50o N’ corridor. |
Wang et al., 2020 | Preprint | Iran | Outdoor environment | 6.7–12.4 °C | 35–50% | - | SARS-COV-2 | The arrival of summer and rainy season can effectively reduce the transmission of COVID-19. |
Luo et al., 2020 | Preprint | China | Outdoor environment | -- | -- | < 9 g/m3 | SARS-COV-2 | Yielded positive relationship with local exponential growth of COVID-19. |
Ma et al., 2020 | Published | China | Outdoor environment | 1.8–18.7 °C | 59–97% | 4.27 − 11.63 g/m3 | SARS-COV-2 | Absolute humidity is negatively associated with daily death counts of COVID-19, but positively associated with temperature. |
Araujo and Naimi, 2020 | Preprint | China | Outdoor environment | -4.01- 15.58ºC | 40–50% | - | SARS-COV-2 | Temperate warm and cold climates are more favorable to spread of the virus, whereas arid and tropical climates are less favorable. |
Auler et al., 2020 | Published | Brazil | Outdoor environment | 27.5ºC | 80% | | SARS-COV-2 | High mean temperatures and intermediate relative humidity influenced the COVID-19 transmission rate. |
Prata et al., 2020 | Published | Brazil | Outdoor environment | < 25.0 ºC | | | SARS-COV-2 | There is no evidence supporting that case counts of COVID-19 could decline when the weather becomes warmer, in temperatures is above 25.8 °C. |
An increase in Ta to reduction in the spread of COVID-19 has been reported by medical studies (Araujo and Naimi, 2020; Bukhari and Jameel, 2020; Oliveiros et al., 2020; Wang et al., 2020). Timing of these studies matches with winter that dominated the greatest development of pandemic in the Northern Hemisphere (Table 2), in which suggests the temperate climates (hot and cold) are more favorable to the spread of COVID-19 than arid and tropical climates (Araujo and Naimi, 2020). However, some studies showed that high Ta could not decrease cases of COVID-19 (Bukhari and Jameel, 2020; Xie and Zhu, 2020). It was demonstrated that the gradual increase of 1.0 °C in the average Ta (> 3.0 °C) had no significant effect on the number of daily COVID-19 cases in 122 Chinese cities (Xie and Zhu, 2020). For non-tropical countries (30ºN and above), it was identified an increase in cases of COVID-19 with Ta higher than 18.0ºC, although most COVID-19 tests (90%) were between 3.0 and 17.0ºC (Bukhari and Jameel, 2020).
The initial studies showed inclusion on temperature influences on COVID-19 spread in Brazil. Auler et al., (2020) reported that high (27.5 ºC) temperature influenced the COVID-19 transmission rate in six capital Brazilian cities. Prata et al., (2020) suggested that there is no evidence supporting that case counts of COVID-19 could decline when the weather becomes warmer, in temperatures above 25.8 °C. They showed a negative linear relationship between temperatures and daily cumulative confirmed cases of COVID-19 in the range from 16.8 °C to 27.4 °C in 27 state capital cities of Brazil.
The literature review identified both positive and negative associations between RH and number of COVID-19 cases in recent studies, which indicates that RH does not shape the spared of SARS-CoV-2. The most obvious explanation is the RH content is inversely proportional to Ta, which puts the variation of the number of COVID-19 cases to Ta variation. The increased AH, by contrast, could impact the COVID-19 case curve because it facilitates the interaction between the virus and water droplets and aerosols (in the air and on surfaces), intensifying environmental transmission of SARS-CoV-2. Even so, the recent studies do not indicate that trend (Auler et al., 2020; Bukhari and Jameel, 2020; Luo et al., 2020; Ma et al., 2020). Even though, 90% of COVID-19 cases from countries located at latitudes above 30ºN had been reported in places with AH below 9 g/m3 (Bukhari and Jameel, 2020), the daily reduction in COVID-19 mortality cases was associated with high levels of AH (8–11 g/m3) in Wuhan, China (pandemic onset), after controlling for effects of air pollution and other conditions, as reported by Ma et. al., (2020).
This review reveals that Ta, RH and AH alone do not able to explain the variation of number COVID-19 cases and predict its behavior to different climatic zones because the thresholds of these parameters that could result in the establishment of an optimal climate for transmission and contagion of SARS-CoV-2 are undefined. Outcomes from COVID 19 – related studies are carried out in outdoor environment and indoor environments (laboratories). This latter is limited to controlled contamination environments and generally do not include in the analytical model all climatic variables (solar radiation, nebulosity, precipitation and air pollutants), which have an important role in the configuration of outdoor environments. Hence, the inclusion of other climatic variables, in addition to temperature and humidity, should guide future ecological models on the relationship between climate and COVID-19.
Descriptive analysis
The correlation coefficient values from multicollinearity test showed most cities have a strong correlation between climatic and number of COVID-19 cases for CC2 and CC14 periods, as illustrated in Table 2. Each city exhibited a particular behavior in term of climatic variable and COVID-19 cases. The São Luís, Fortaleza, Manaus, and Recife, localized around º0 lat with dominated equatorial climate, showed significantly positive and negative association with number of COVID-19 cases (r > 0,70) for Tmax, Tmin, RHmax, RHmin, and WS. On the other hand, DPmax, DPmin, Tmax, Tmin and SR presented a greater correlation with COVID-19 cases in the central (Brasília) southeast (Rio de Janeiro and São Paulo) cities. Overall, these results suggest that the climatic parameters had greater association with variation of exponential curve of COVID-19 cases were Tmax, Tmin, DPmax followed by SR and WS. Further studies are needed to address the integrated climatic variable analysis, e.g., the types of weather and COVID-19 cases. It was highlighted the time lag of CC14 revealed a suitable way to evaluate the correlation between number of COVID-19 cases and climatic parameters, as shown in Table 3.
Table 3
Description of multicollinearity’s correlation coefficient values and climatic parameters in six Brazilian cities.
| FORTALEZA | MANAUS | RECIFE | BRASILIA | RIO DE JANEIRO | SAO PAULO |
CC1 | CC7 | CC14 | CC1 | CC7 | CC14 | CC1 | CC7 | CC14 | CC1 | CC7 | CC14 | CC1 | CC7 | CC14 | CC1 | CC7 | CC14 |
Tmax | -0,587 | -0,665 | -0,599 | -0,232 | -0,613 | -0,726 | -0,594 | -0,785 | -0,867 | -0,502 | -0,518 | -0,584 | -0,268 | -0,526 | -0,781 | -0,268 | -0,380 | -0,635 |
Tmin | -0,584 | -0,678 | -0,609 | -0,226 | -0,625 | -0,679 | -0,609 | -0,789 | -0,868 | -0,539 | -0,514 | -0,552 | -0,306 | -0,577 | -0,811 | -0,301 | -0,423 | -0,674 |
RHmax | 0,389 | 0,622 | 0,612 | 0,210 | 0,588 | 0,582 | 0,405 | 0,704 | 0,819 | -0,009 | 0,046 | 0,105 | -0,194 | -0,214 | -0,465 | -0,533 | -0,554 | -0,609 |
RHmin | 0,406 | 0,601 | 0,570 | 0,191 | 0,534 | 0,585 | 0,419 | 0,718 | 0,828 | 0,022 | 0,118 | 0,257 | -0,177 | -0,182 | -0,418 | -0,492 | -0,508 | -0,558 |
DPmax | -0,395 | -0,383 | -0,265 | -0,026 | -0,025 | -0,137 | -0,187 | 0,101 | 0,352 | -0,506 | -0,549 | -0,706 | -0,435 | -0,718 | -0,907 | -0,679 | -0,784 | -0,888 |
DPmin | -0,320 | -0,345 | -0,295 | -0,032 | -0,036 | -0,076 | -0,128 | 0,164 | 0,390 | -0,442 | -0,438 | -0,587 | -0,443 | -0,734 | -0,922 | -0,661 | -0,786 | -0,883 |
WS | -0,058 | -0,383 | -0,292 | -0,214 | -0,617 | -0,439 | - | - | - | 0,094 | 0,422 | 0,428 | -0,240 | -0,396 | -0,625 | -0,338 | -0,525 | -0,466 |
SR | 0,430 | 0,711 | 0,659 | -0,181 | -0,530 | -0,792 | 0,063 | -0,082 | -0,039 | -0,111 | -0,165 | -0,170 | -0,337 | -0,877 | -0,935 | -0,120 | -0,423 | -0,708 |
RA | 0,361 | 0,459 | 0,410 | 0,070 | 0,236 | -0,088 | 0,242 | 0,583 | 0,573 | -0,067 | -0,011 | 0,037 | -0,128 | -0,279 | -0,501 | -0,353 | -0,530 | -0,655 |
n/N | -0,263 | -0,299 | -0,258 | 0,103 | 0,119 | -0,174 | -0,266 | -0,352 | -0,273 | 0,136 | 0,208 | 0,241 | 0,156 | -0,199 | -0,323 | 0,365 | 0,512 | 0,535 |
COVID-19 Cases per day | 0,812 | 0,812 | 0,812 | 0,700 | 0,700 | 0,700 | 0,721 | 0,791 | 0,791 | 0,714 | 0,791 | 0,737 | 0,644 | 0,622 | 0,638 | 0,791 | 0,791 | 0,791 |
COVID-19 cases and linkages with insolation
To examine the effects of insolation (L) and direct solar radiation (DSR) associated with Nebulosity (Nb) on the origin of COVID-19 spread, the Figs. 1a-f show daily values of L ratio and number of COVID-19 cases in six Brazilian cities started in January 30, 2020, when WHO decreed that COVID-19 had become an Emergency of Public Health of International Importance. We recorded the extended incubation period up to 20 days before the first notified case to analyze the insolation conditions during the possible initial days of transmission and contagion by the SARS-CoV-2. Most cities presented cumulative daily L ratio of 0.0 during the incubation period of the first case of community spread, indicating that, on average, somewhat more than half of the L ratio of the days leading up to the cases recorded with medium to high Nb (low L and low DSR).
Since mid-March this year, new cases of COVID-19 from community spread have been confirmed in Brazilian cities in almost all regions of the country all in hot and humid regions (Table 4). The first records of autochthonous transmission of COVID-19 in these cities occurred during the summer and beginning of the southern autumn. The maximum monthly average Ta (above 30.0ºC) in the period between January and April 2020 in these cities suggests that high Ta, even in this seasonality, may not limit the survival and transmission of SARS-CoV-2 in tropical environments. The fact that these cities reported the initial phase of pandemic in the middle of summer and in the early autumn encourages the development of new studies in the coming weeks and months, since Ta has not yet proved to be limiting to the spread of the SARS-CoV-2.
Table 4
Brazil - Climate and COVID-19 cases (partial) until 2020/04/12.
City | Latitude | Climate type* | First case Covid-19 | Case registration** | Registration Deaths** | Lethality rate*** |
Manaus | 3ºs | Am | 13-mar-20 | 1053 | 52 | 49.3 |
Fortaleza | 3ºS | Aw | 15-mar-20 | 1747 | 74 | 42.3 |
Recife | 8ºS | Am | 12-mar-20 | 960 | 65 | 67.7 |
Brasília | 15ºS | Aw | 7-mar-20 | 614 | 14 | 22.8 |
Rio de Janeiro | 22ºS | Aw | 5-mar-20 | 2855 | 170 | 59.5 |
São Paulo | 23ºS | Cfa | 26-fev-20 | 6352 | 588 | 92.5 |
BRASIL | 5ºN/34ºS | A – B – C | 26-fev-20 | 22318 | 1230 | 55.1 |
Note: *Climatic groups according to Köppen-Geiger climate classification. ** Data refer to the states, whereas most cases were registered in the capital cities. ***Lethality rate obtained by the ratio between 1000 confirmed cases and the number of deaths. |
These results suggest the transmission and contagion by SARS-CoV-2 seem to have been enhanced under from medium to low DSR. Ahmadi et al., (2020) also found high rate of infection of COVID-19 associated with low SR in five Iranian providences. Sagripanti (2007) reported that the inactivation of viruses in the environment by high solar ultraviolet radiation (UV-C) plays a role in the seasonal occurrence of influenza pandemics. We suggest the need for studies and immediate advances regarding the influence of insolation on the ecology of the vector related to SARS-CoV-2, a fact that may affect public policies and coordinated actions to reduce and control of COVID-19.