key: cord-0936649-2umjqfv2
authors: Páez-Osuna, Federico; Valencia-Castañeda, Gladys; Rebolledo, Uriel Arreguin
title: The link between COVID-19 mortality and PM(2.5) emissions in rural and medium-size municipalities considering population density, dust events, and wind speed
date: 2021-07-22
journal: Chemosphere
DOI: 10.1016/j.chemosphere.2021.131634
sha: 65228fbf6909a09680514e72545e0cc38d24f1a6
doc_id: 936649
cord_uid: 2umjqfv2

One contemporary issue is how environmental pollution and climate can affect the dissemination and severity of COVID-19 in humans. We documented the first case of association between particulate matter ≤2.5 μm (PM(2.5)) and COVID-19 mortality rates that involved rural and medium-sized municipalities in northwestern Mexico, where direct air quality monitoring is absent. Alternatively, anthropogenic PM(2.5) emissions were used to estimate the PM(2.5) exposure in each municipality using two scenarios: 1) considering the fraction derived from combustion of vehicle fuel; and 2) the one derived from modeled anthropogenic sources. This study provides insights to better understand and face future pandemics by examining the relation between PM(2.5) pollution and COVID-19 mortality considering the population density and the wind speed. The main findings are: (i) municipalities with high PM(2.5) emissions and high population density have a higher COVID-19 mortality rate; (ii) the exceptionally high COVID-19 mortality rates of the rural municipalities could be associated to dust events, which are common in these regions where soils without vegetation are dominant; and (iii) the influence of wind speed on COVID-19 mortality rate was evidenced only in municipalities with <100 inhabitants per km(2). These results confirm the suggestion that high levels of air pollutants associated with high population density and an elevated frequency of dust events may promote an extended prevalence and severity of viral particles in the polluted air of urban, suburban, and rural communities. This supports an additional means of dissemination of the coronavirus SARS-CoV-2, in addition to the direct human-to-human transmission.

The coronavirus disease 2019 (COVID-19) pandemic caused by the novel coronavirus 52 SARS-CoV-2 is generating a high number of deaths (~3.94 million people on June 2021, 53 WHO, 2021), with negative effects on the public health and economic systems (Coccia 54 2020a; 2021d). The main lessons learned from the COVID-19 pandemic are the importance 55 of healthcare expenditures to reduce fatality rates (Coccia, 2021) , the timely application of 56 containment policies, and understanding the influence of climatic factors and air pollution on 57 the spread and severity of COVID-19. Coccia (2021c) proposes that gross domestic product 58 per capita, healthcare spending, and air pollution of nations are critical factors associated 59 with the fatality rate of COVID-19. In addition, Coccia (2021a) found that wind speeds and 60 air pollution may support the dissemination of COVID-19 leading to a higher incidence of 61 infected individuals and deaths; cities with high wind speed have less cases of COVID-19, 62 compared to those with little wind speed and frequent high levels of air pollution. For 63 example, Coccia (2021d) found that nearly 81% of deaths from the first wave of COVID-19 64 pandemic in Italy occurred in industrialized regions with high levels of contaminated air. 65

Atmospheric stability (i.e., low wind speed) reduces the dispersion of gases, and 66 particulate matter (PM) can act as a carrier of the SARS-CoV-2 in the air to sustain the 67 dissemination of COVID-19 in the environment, generating problems in the public health 68 (Coccia 2021e ). The concentration of atmospheric pollutants can also include viral agents 69 et al., 2021). The authors observed that the concentration of PM2.5 increased by 221% after 155 wildfire events. Posteriorly, the number of cases and deaths due to COVID-19 increased by 156 57% and 148%, respectively. These results suggest that the California wildfire produced an 157 increase in ambient levels of toxic pollutants, which were temporally associated with an 158 increase in the occurrence and mortality of COVID-19. Therefore, long-term exposure is not 159 necessarily required to promote the conditions of greater onset and mortality by From such a situation, it is important to examine if this association occurs in different 161 scenarios, such as rural communities, including small and medium municipalities. The main 162 aim of this work is to examine the relationships between the COVID-19 mortality rate and 163 PM2.5, meteorological, and demographic factors in the municipalities of Sinaloa, Mexico. The 164 hypothesis is that COVID-19 severely affects (mortality) communities that have a greater 165 exposure to high levels of PM2.5, a high population density, and low wind speed, including 166 rural towns and small and medium-sized cities. The study case described in this work was carried out in Sinaloa, Northwest Mexico (~3 171 million people). This state includes 18 municipalities ( Fig. 1) which consist of rural 172 communities, towns, and small and medium-sized cities, where a variable incidence and 173 mortality of COVID-19 has been observed. The first case of COVID-19 in Mexico was 174 detected in February 2020 in Culiacan (Sinaloa) by a Mexican citizen that arrived from Italy. 175

Since then, the pandemic has fluctuated in intensity within the country, and the highest 176 number of deaths were recorded in January and February 2021. In particular, this study 177 analyzes the accumulated mortality rate that included the first and second wave of the 178 COVID-19 pandemic in Mexico from February 2020 to April 2021. Mexico had exhibited two Descriptive statistics is performed classifying municipalities of Sinaloa in groups, 206 considering: 207 -Wind speed, based on the arithmetic mean of the sample given by 9.3 km/h (Coccia, 208 2021a), which includes (i) municipalities with high wind speed (>9.3 km/h) indicating 209 a wind force from light to moderate breeze according to the Beaufort wind scale 210 (average wind force of light breeze means that wind is felt on face, leaves rustle, 211 vanes begin to move, whereas a wind force of moderate breeze generates the wind 212 effect of dust, leaves and loose paper lifted, and the movement of small tree 213 branches); and (ii) municipalities with low wind speed (<9.3 km/h) indicating a wind 214 force from calm air to light breeze according to the Beaufort wind scale (Table 2A) . 215 -Population density, which is low in Sinaloa, but was grouped in two categories: (i) 216 municipalities with a low population density between 100 and 200 inhabitants/km 2 ; 217 and (ii) municipalities with very low population density <70 inhabitants/km 2 (Table  218 1A). 219 -Air pollution based on PM2.5 emissions estimated from anthropogenic sources that 220 includes: (i) municipalities with moderate or low PM2.5 emissions (100-350 t/year); 221 and (ii) municipalities with very low PM2.5 emissions (<80 t/year) (Table 2A) . 222

Air quality monitoring stations are absent in most municipalities, except for Culiacan, 224

Los Mochis, and Mazatlán; however, these stations operate irregularly. Therefore, an 225 alternative was used to estimate the exposure to air contaminants. The estimated exposure 226 to PM2.5 in each municipality was inferred from their emissions following two scenarios: A) 227 considering only PM2.5 emitted from vehicles per municipality, calculating these 228 microparticles from the combustion of gasoline and diesel; and B) considering PM2. (Table 2A) . Therefore, the total anthropogenic emissions of PM2.5 per municipality in t year -281

The study analyzes trends with a simple regression model based on a linear semi-log 285 model: 286

[3] 287 y = accumulated mortality rate (number of deaths per 10 5 people); e = PM2.5 emission 288

(t year -1 ); b is a constant; and m is the slope of regression. 289

Ordinary least squares method is applied (Coccia, 2021b) The ordinary least squares method is applied to quantify the unknown parameters of 303 linear models using SigmaPlot 11.0 (Systat Software, Inc.) (Coccia, 2020b) . 304

The calculation involved in the equation [4] using the explanatory variable of population 305 density and population density, is also performed with the categorization of municipalities 306 according to wind speed (higher or lower than 9.3 km/h). Similarly, the calculation of the 307 equation [3] using the explanatory variable of anthropogenic PM2.5 emissions, is also performed with the categorization of municipalities according to moderate or low (100-350 309 t/y) or very low (<80 t/y) PM2.5 emissions. Table 1 shows that municipalities in regions with low wind speed exhibit a higher number 333 of days <9.3 km/h (300.2 days) compared to municipalities with high wind speed (146.2 days). In contrast, the population density of the first is lower compared with the second; this 335 combination of factors partially explains the comparable incidence mean of the COVID-19 336 mortality rate between municipalities of high and low wind speed. Table 1 shows also that 337 municipalities with high wind speed (and lower number of days <9.3 km/h) have a greater 338 population density and a higher number of older population than municipalities with low wind 339 speed. These results are compatible with the suggestion given by Coccia (2020; 2021a) , 340 which indicates that high intensity and higher prevalence of wind speed improves the 341 dispersion of particulate matters and polluted gases, and consequently, it can mitigate the 342 dissemination of COVID-19. 343 Table 2 presents descriptive statistics considering municipalities with low or very low 344 annual emissions of anthropogenic PM2.5. Municipalities with low wind speed (average <9.3 345 km/h) also show a lower population density and most people are >60 years old; also, they 346 have low mortality rates, which is explained mainly by the very low load (31.4 t/year) of 347 anthropogenic emission of PM2.5. Contrarily, those municipalities that receive a higher load 348 of anthropogenic emissions of PM2.5 (mean, 213.7 t/year) with a more elevated population 349 density, but lower number of days with <9.3 km/h and higher wind speed, as well as an older 350 population, exhibit the highest mortality rate by COVID-19. Los Mochis, Culiacan, and 351

Mazatlan evidence higher PM emissions, and there is evidence that people are exposed to 352 PM2.5 levels that exceed the guidelines of the WHO (2016) (annual mean 10 µg/m 3 ) (Ramos-353 Álvarez, 2020). Table 2 indicates that the COVID-19 mortality rates are moderately higher 354 in those municipalities with high PM2.5 emissions compared to those with low emissions; 355 however, the difference between these municipalities is relatively low. The population 356 Fig. 2A (supplementary material) shows the regression lines with the 409 COVID-19 mortality rate increases with population density. Also, it is evident that the change 413 of mortality rate on the population density is comparable for both municipalities of low and 414 high wind speed. When the Log (Population density) is > 2.0, i.e., > 100 inhabitants per km 2 , 415 the mortality rate is more elevated for the municipalities with high wind speed than those 416 with low wind speed. This indicates that the population density has a greater influence when 417 this is > 100. In contrast, for municipalities < 100 inhabitants per km 2 , mortality is higher in 418 those with low wind speed. It is important to highlight that this behavior is observed where 419 the differences of wind speed (between high and low) are < 5 km/h. 420 The positive association found (Figs. 1A, 2, 3, and 4) indicates that a higher PM2.5 466 emission is associated with higher cumulative COVID-19 mortality, a behavior similar to that 467 Clearly, this is a critical point that requires consideration in future studies. 502

Although this study is limited to rural communities and small and medium-sized cities, 503 our results suggest that the COVID-19 mortality rate is significantly associated with PM2.5 504 and population density in the municipalities examined in the present study. Wind speed 505 exhibited an evident effect only in very low population densities (<100 per km 2 ). These NO2 and PM2.5 levels (Paital and Agrawal, 2021) . 520

This study is limited in several points, including the duration of the study period. association. Third, the heterogeneity of air pollution within each municipality is ignored. 532

Fourth, it is assumed that the people of each municipality reside effectively in such location, 533

although it is well known that some people in Sinaloa eventually reside in the main cities 534 and in rural towns. In this exercise, individuals who were considered were theoretically 535 exposed to PM2.5 in the long term or during the study period, originating from the 536 anthropogenic sources based on residence in their municipality. Fifth and last, only PM2.5 537 emissions were used in this exercise, and not PM2.5 exposure, which may imply significant 538 failures for the reasons discussed in the methodology. Apart from these limitations, our work 539 reveals important part of the COVID-19 situation in rural communities and suburban areas 540 regarding the link between COVID-19 and air pollution.

We documented the first case of association between PM2.5 and COVID-19 mortality 543 rates that involved rural communities, as well as small and medium-sized municipalities from 544

Mexico. In addition to major cities, industrial regions, and localities with wildfires, the link 545 between COVID-19 and air pollution can also involve rural and small municipalities, which 546 can in turn be related to conventional sources (vehicles, industrial sources, and burning of 547 The authors declares no known competing financial interests or personal relationships 571 that could have appeared to influence the work reported in this paper. 572

Delgado, E., Cereceda-Balic, F., 2020. Spread of SARS-CoV-2 through Latin America 588 and the Caribbean region: A look from its economic conditions, climate and air pollution 589 indicators, Environ. Res. 191 J o u r n a l P r e -p r o o f 

Investigation of 579 effective climatology parameters on COVID-19 outbreak in Iran

Correlation between environmental pollution indicators and COVID-19 pandemic: a brief 583 study in Californian context

Factors determining the diffusion of COVID-19 and suggested strategy to 591 prevent future accelerated viral infectivity similar to COVID

An index to quantify environmental risk of exposure to future epidemics 594 of the COVID-19 and similar viral agents: Theory and Practice

How (un)sustainable environments are related to the diffusion of COVID-597 19: the relation between coronavirus disease 2019, air pollution, wind resource and 598 energy

The relation between length of lockdown, numbers of infected people and 600 deaths of Covid-19, and economic growth of countries: lessons learned to cope with 601 future pandemics similar to Covid-19 and to constrain the deterioration of economic 602 system

How do low wind speeds and high levels of air pollution support the 605 spread of COVID-19?

The impact of first and second wave of the COVID-19 pandemic in 608 society: comparative analysis to support control measures to cope with negative effects 609 pandemic and epidemics of future vital agents

The effects of atmospheric stability with low wind speed and of air 619 pollution on the accelerated transmission dynamics of COVID-19

COVID-19 Sinaloa 2021. COVID-19 Enfermedad por el Nuevo coronavirus

Characterizing all-626 cause excess mortality patterns during COVID-19 pandemic in Mexico

Influence 629 of population density, temperature, and absolute humidity on spread and decay 630 durations of COVID-19: a comparative study of scenarios in China

Critical review of social, environmental and health risk 634 factors in the Mexican indigenous population and their capacity to respond to the 635 INEGI. 2021a. Vehículos de motor registrados en circulación

COVID-19 and climatic factors: a global analysis

Remote sensing of PM, NO, CO, CO and HC 674 emission factors for on-road gasoline and diesel engine vehicles in Las Vegas

Impacts of the COVID-19 lockdown on air quality and its association with human 678 mortality trends in megapolis Mexico City

Particulate matter concentrations and their association with COVID-19-related mortality

Saharan dust event

Air pollution exposure and 685 COVID-19: a look at mortality in Mexico City using individual-level data. Sci. Total mortality of SARS-COV-2 infection in ten wildfire affected counties in California

Air pollution by NO2 and PM2.5 explains COVID-19 infection 694 severity by overexpression of angiotensin-converting enzyme 2 in respiratory cells: a 695 review

Country-level 698 determinants of the severity of the first global wave of the COVID-19 pandemic: an 699 ecological study

Oxidative potential of ambient PM and related health endpoints 702 over South Asia: a review

Bidirectional 706 association between COVID-19 and the environment: a systematic review

Evaluación del impacto en la salud por material particulado 709 (PM2.5) en las principales ciudades de Sinaloa, México. Tesis Doctoral

Environmental perspective of COVID-19

Impact of meteorological factors on COVID-19 719 pandemic: evidence from top 20 countries with confirmed cases

Evaluación rápida del uso de la energía

COVID-19 and air pollution and meteorology-an intricate relationship: 724 a review

Exposure to different fractions of diesel exhaust PM2.5 induces 728 different levels of pulmonary inflammation and acute phase response

Association of environmental and 731 meteorological factors on the spread of COVID-19 in Victoria, Mexico, and air quality 732 during the lockdown

WHO, 2021. WHO Coronavirus (COVID-19) Dashboard

Commission on healthy diets from sustainable food systems

Air pollution and COVID-752 19 mortality in the United States: strengths and limitations of an ecological regression 753 analysis

Association 755 of particulate matter pollution and case fatality rate of COVID-19 in 49 Chinese cities

Chemical 758 characterization of fine particulate matter in gasoline and diesel vehicle exhaust

Combustion sources of particles: 2. Emission factors and 761 measurement methods

We would like to thank Instituto de Ciencias del Mar y Limnología, UNAM the support to 575 develop this study: institutional project for FPO and GVC. 576