key: cord-1006436-gpsefxz0 authors: Ismail, Iqbal M.I.; Rashid, Muhammad Imtiaz; Ali, Nadeem; Saeed Altaf, Bothinah Abdullah; Munir, Muhammad title: Temperature, humidity and outdoor air quality indicators influence COVID-19 spread rate and mortality in major cities of Saudi Arabia date: 2021-09-23 journal: Environ Res DOI: 10.1016/j.envres.2021.112071 sha: 588ddcf04e91e828398dc896681646015accf04a doc_id: 1006436 cord_uid: gpsefxz0 There is an increasing evidence that meteorological (temperature, humidity, dew) and air quality indicators (PM2.5, PM10, NO2, SO2, CO) are affecting the COVID-19 transmission rate and the number of deaths in many countries around the globe. However, there are contradictory results due to limited observations of these parameters and absence of conclusive evidence on such relationships in cold or hot arid tropical and subtropical desert climate of Gulf region. This is the first study exploring the relationships of the meteorological (temperature, humidity, dew) and air quality indicators (PM10, CO, and SO2) with daily COVID-19 infections and death cases for a period of six months (1st March to August 31, 2020) in six selected cities of the Kingdom of Saudi Arabia by using generalized additive model. The Akaike information criterion (AIC) was used to assess factors affecting the infections rate and deaths through the selection of best model whereas overfitting of multivariate model was avoided by usingcross-validation. Spearman correlation indicated that exponentially weighted moving average (EWMA) temperature and humidity (R > 0.5, P < 0.0001) are the main variables affecting the daily COVID-19 infections and deaths. EWMA temperature and humidity showed non linear relationships with the number of COVID-19 infections and deaths (DF > 1, P < 0.0001). Daily COVID-19 infections showed a positive relationship at temperature between 23 and 34.5 °C and humidity ranging from 30 to 60%; a negative relationship was found below and/or above these ranges. Similarly, the number of deaths had a positive relationship at temperature ˃28.7 °C and with humidity ˂40%, showing higher number of deaths above this temperature and below this humidity rate. All air quality indicators had linear relationships with the number of COVID-19 infections and deaths (P < 0.0001). Hence, variation in temperature, humidity and air pollution indicators could be important factors influencing the COVID-19 spread and mortality. Under the current scenario with rising temperature and humidity, the number of cases is increasing, hence it justifies an active government policy to lessen COVID-19 infection rate. subtropical desert climate of Gulf region. This is the first study exploring the 38 relationships of the meteorological (temperature, humidity, dew) and air quality From the reports, it appeared that the seafood market was responsible for spreading the 66 majority of the early cases in China, but laterally the disease spread around the globe 67 through its transmission by human-to-human contact (Zhou et al., 2020) . Therefore, its 68 "community spread" nature made this a high frequency spreading virus in humans within 69 short time (Kraemer et al., 2020) . Because of its high spreading nature, the number of in humid air and therefore influenced the daily infections and mortality rate associated 89 with COVID-19 pneumonia (Biktasheva, 2020) . Consequently, it is important to 90 understand how environmental variables such as air humidity and temperature can play a 91 role in the spread, transmission and preventing COVID-19 infection. Such information 92 will not only be helpful for policy maker to make guidelines about the viral spread but 93 also beneficial for the public. Seasonal cyclicity is a ubiquitous feature of any acute infectious disease (Martinez, 95 2018). Such phenomenon is generally found in most viral respiratory diseases (Martinez, 96 2018). The common example is influenza which displayed a strong seasonal cyclicity and 97 spread during winter in the temperate regions (Tamerius et al., 2013) . Therefore, a strong 98 robust association was observed between environmental parameters (temperature and 99 humidity) and influenza incidence in these regions (Lowen and Steel, 2014) . Similarly, 100 the activity and transmission efficacy of SARS-CoV-2 in people were also strongly 101 J o u r n a l P r e -p r o o f influenced by meteorological factors such as temperature and humidity (Lin et al., 2020) . 102 In mainland China, the COVID-19 transmission was negatively associated with 103 temperature and humidity (Qi et al., 2020 and least humid months with maximum humidity reached to 70% and minimum to 30% 228 in Abha. The other selected cities were located in the subtype "BWh" hot arid tropical 229 and subtropical desert climate. In these cities, the temperature ranged -1.1 to 50 °C, 230 during the year, with the minimum temperature observed in January and maximum in 231 July. The average minimum temperature ranged from 9.6 to 23.3, and the maximum 232 temperature varies from 30.1 to 36.1 C in these cities. The mean humidity of these cities 233 ranged between 10 to 95%. Where DF shows degrees of freedom, f indicates frequencies number used in the model, 254 and denotes the log likelihood which is calculated in the model at converged estimates. 255 We performed descriptive analysis on the data of daily COVID-19 infections and deaths, Where is the number of confirmed infected/death cases. β 1 (EWMA temp), β 1 (EWMA This software procedure is based on two DF for spline functions: humidity and dew, and air quality variable data (Table 3) . SO 2 ) as confounding variables (Table 3 ). Fig. 1a shows the estimated smoothing spline humidity level of 45% (Fig. 1b) . The linear part of the relationship showed that for each 1 365 ˚C rise in temperature there was 13% increase in the number of COVID-19 infections 366 (P<0.0001; Table 3 ). In contradiction to temperature, each 1% rise in air humidity, 367 decreased the number of COVID-19 patients by 5% (Table 3) . The relationship of EWMA temperature and humidity with number of COVID-19 deaths 369 were different than what was observed for number of infected cases. Fig. 2a where the linear relationship coefficient showed an increase in number of deaths by 376 33.7% with each 1% increase in humidity (Table 3 ). Fig. 2b shows the estimated spline 377 J o u r n a l P r e -p r o o f function which indicated nonlinear relationship between EWMA humidity and death rate 378 with the highest influence when the EWMA humidity below 40%. The interaction of EWMA temperature and humidity on the number of COVID-19 380 infections and death cases is a three-dimensional relationship presented in Fig. 3 (Fig. 3a) . On the other hand, humidity below 30% resulted in higher nonlinear effects on 387 the number of deaths at all temperatures under investigation and this parameter above 388 42% resulted in low effect on the number of deaths at all temperatures (Fig. 3b) . The critical part of these relationships is the Tests for Smoothing Components (Table 5) . Our strong motivation for this study was the discussion that geographical scope and surfaces (Biryukov et al., 2020) . In contradiction to their study, we observed higher 461 number of infections at this temperature range (Fig. 1a) , however the humidity above 462 62% resulted in lower number of infections (Figs. 1b) at temperature range of 14-35 ˚C 463 (Fig. 3a) . The difference in our results with the above reported studies could be explained Therefore, as observed in previous studies (Biktasheva, 2020; Ma et al., 2020) , in present 520 study the COVID-19 mortalities may related to lower humidity (Fig. 1b) . The air pollution also significantly affecting the COVID-19 related mortalities in (Table 1, 3) . There are number of limitations of the present study that could not be ignored. Table 5 . Tests for smoothing components of major environmental variables (temperature, humidity and their interaction) effect on number of COVID-19 infected cases and deaths. The effective degrees of freedom (EDF) is a summary statistic of GAM and it reflects the degree of non-linearity of a curve. a) An EDF equal to 1 is equivalent to a linear relationship. b) EDF > 1 and ≤ 2 is a medium non-linear relationship, and (c) an EDF > 2 indicates a highly non-linear relationship. The EDF for test examines the existence of a contribution for each smoothing component. Investigation of effective climatology parameters on COVID-19 591 outbreak in Iran. Science of The Total Environment Presentation of a developed sub-epidemic model for estimation of the 593 COVID-19 pandemic and assessment of travel-related risks in Iran. Environmental 594 Science and Pollution Research Modelling the impact of testing, contact tracing and household quarantine 596 on second waves of COVID-19 Exploring the linkage between PM2.5 levels and COVID-19 spread and 598 its implications for socio-economic circles Trends of temperature extremes in Saudi Arabia Climate factors and incidence of Middle East respiratory 602 syndrome coronavirus Climate change, humidity, and mortality in the United States Correlation between climate indicators and COVID-19 pandemic in 606 New York, USA. Science of The Total Environment Role of a habitat's air humidity in Covid-19 mortality. Science of The 608 Total Environment Increasing temperature and relative humidity accelerates inactivation of 610 SARS-CoV-2 on surfaces Spread of SARS-CoV-2 through Latin America and the 612 Caribbean region: A look from its economic conditions, climate and air pollution 613 indicators Effects of air temperature on climate-sensitive mortality and morbidity 615 outcomes in the elderly; a systematic review and meta-analysis of epidemiological 616 evidence The effects of temperature and relative humidity on the viability of the 618 SARS coronavirus Can atmospheric pollution be considered a co-factor in extremely high 620 level of SARS-CoV-2 lethality in Northern Italy? Environmental pollution Death from respiratory diseases and temperature in Meteorological patterns, technical validation, and chemical comparison 624 of atmospheric dust depositions and bulk sand in the Arabian Gulf region. Environmental 625 Pollution Human mobility restrictions and the spread of the novel coronavirus (2019-ncov) 627 in china Effect of air temperature and universal thermal climate 629 index on respiratory diseases mortality in Mashhad, Iran. Archives of Iranian medicine Generalized linear and generalized additive models in studies of species 632 distributions: setting the scene. Ecological modelling Relationship between COVID-19 infection rates and air pollution, 634 geo-meteorological, and social parameters Bronchoconstriction triggered by breathing hot humid air in patients 637 with asthma: role of cholinergic reflex. American journal of respiratory and critical care 638 medicine Clinical features of patients infected with 2019 novel coronavirus in 640 The lancet The effects of regional climatic condition on the spread of COVID-19 642 at global scale Source apportionment and elemental composition of PM2.5 and PM10 644 in Jeddah City, Saudi Arabia. Atmospheric Pollution Research Longer incubation period of coronavirus disease 2019 (COVID-19) in older 646 adults The effect of human mobility and control measures on the COVID-648 19 epidemic in China Low ambient humidity impairs barrier function and innate resistance 650 against influenza infection Ozone augments interleukin-8 production induced by ambient particulate 653 matter Temperature, temperature extremes, and cause-specific respiratory mortality 655 in China: a multi-city time series analysis Urban air pollution may enhance COVID-19 case-fatality and mortality 658 rates in the United States. The Innovation A mechanism-based parameterisation scheme to investigate the association 660 between transmission rate of COVID-19 and meteorological factors on plains in China. 661 Science of the Total Environment Ambient particulate air pollution and daily mortality in 652 cities Impact of meteorological factors on the COVID-19 transmission: A multi-665 city study in China. Science of The Total Environment Aerodynamic analysis of SARS-CoV-2 in two Wuhan hospitals Influenza virus transmission is dependent on relative humidity and 669 temperature Roles of Humidity and Temperature in Shaping Influenza 671 Seasonality Effects of temperature variation and humidity on the death of COVID-19 in 673 Science of The Total Environment Effects of air pollution on the potential transmission and mortality of 675 COVID-19: A preliminary case-study in Tarragona Province The calendar of epidemics: Seasonal cycles of infectious diseases Spatiotemporal variability of COVID-19 pandemic in relation to air 680 pollution, climate and socioeconomic factors in Pakistan Revealing the relationship between solar activity and COVID-19 682 and forecasting of possible future viruses using multi-step autoregression (MSAR) Association of COVID-19 pandemic with meteorological parameters 685 over Singapore. Science of The Total Environment A review of the impact of weather and climate variables to COVID-687 19: In the absence of public health measures high temperatures cannot probably mitigate 688 outbreaks The Dose-Response Association between Nitrogen Dioxide Exposure 690 and Serum Interleukin-6 Concentrations Regional and global contributions of air pollution to risk of death from 693 COVID-19 Temperature significantly changes COVID-19 transmission in (sub) 695 tropical cities of Brazil Association between viral seasonality and meteorological factors COVID-19 transmission in Mainland China is associated with temperature 699 and humidity: A time-series analysis Estimation of incubation period distribution of COVID-19 using disease 701 onset forward time: A novel cross-sectional and forward follow-up study How air quality and COVID-19 transmission change under different 704 lockdown scenarios? A case from Dhaka city, Bangladesh. Science of The Total 705 Environment Nutrient use efficiency (NUE) for sustainable wheat production: a 707 review SARS-Cov-2RNA Found on Particulate Matter of Bergamo in Northern 709 Italy: First Evidence Association between climate variables and global transmission oF 711 SARS-CoV-2. Science of The Total Environment COVID-19 and air pollution and meteorology-an intricate relationship: A 713 review Environmental predictors of seasonal influenza epidemics across 715 temperate and tropical climates Links between air pollution and COVID-19 in England. 717 Environmental Pollution Naming the coronavirus disease (COVID-19) and the virus that causes it Thin plate regression splines Generalized Additive Models Air pollution and COVID-19 mortality in the United States: Strengths and 724 limitations of an ecological regression analysis Exposure to air pollution and COVID-19 mortality in the United States Effects of temperature and humidity on the daily new cases and new deaths 728 of COVID-19 in 166 countries Beware of the second wave of COVID-19. The Lancet Zika virus: A call to action for physicians in the era of climate 731 change The role of weather conditions in COVID-19 transmission: A study of a 733 global panel of 1236 regions Associations between air pollution and COVID-19 epidemic during 735 quarantine period in China A pneumonia outbreak associated with a new coronavirus of probable bat 737 origin Association between short-term exposure to air pollution and COVID-19 739 infection: Evidence from China. Science of The Total Environment Mixed effects models and extensions in ecology with R J o u r n a l P r e -p r o o f