Investigating connections between COVID-19 pandemic, air pollution and community interventions for Pakistan employing geoinformation technologies

Abstract

Several major cities that witnessed heavy air pollution by particulate matter (PM_2.5) concentration and nitrogen dioxide (NO₂) have contributed to high rate of infection and severity of the coronavirus disease (COVID-19) pandemic. Owing to the negative impact of COVID-19 on health and economy, it is imperative to predict the pandemic trend of the COVID-19 outbreak. Pakistan is one of the mostly affected countries by recent COVID-19 pandemic in terms of COVID-cases and economic crises. Like other several Asian countries to combat the virus impacts, Pakistan implemented non-pharmacological interventions (NPI), such as national lockdowns. The current study investigates the effect of major interventions across three out of four provinces of Pakistan for the period from the start of the COVID-19 in March 22, 2020 until June 30, 2020, when lockdowns were started to be eased. High-resolution data on NO₂ was recorded from Sentinel-5's Precursor spacecraft with TROPOspheric Monitoring Instrument (Sentinel-5P TROPOMI). Similarly, PM_2.5 data were collected from sampling sties to investigate possible correlation among these pollutants and COVID-19. In addition, growth and susceptible-infected-recovered (SIR) models utilizing time-series data of COVID-19 from February 26 to December 31, 2020, with- and without NPI that encompass the predicted number of infected cases, peak time, impact on the healthcare system and mortality in Pakistan. Maximum mean PM_2.5 concentration of 108 μgm^-3 was recorded for Lahore with the range from 51 to 215 μgm^-3, during strict lockdown (L), condition. This is three times higher than Pak-EPA and US-EPA and four times for WHO guidelines, followed by Peshawar (97.2 and 58 ± 130), Islamabad (83 and 158 ± 58), and Karachi (78 and 50 ± 140). The majority of sampling sites in Lahore showed NO₂ levels higher than 8.75E-5 (mol/m²) in 2020 compared to 2019 during "L" period. The susceptible-infected-recovered (SIR) model depicted a strong correlation (r) between the predicted and reported cases for Punjab (r = 0.79), Sindh (r = 0.91), Khyber Pakhtunkhwa (KPK) (r = 94) and Islamabad (r = 0.85). Findings showed that major NPI and lockdowns especially have had a large effect on minimizing transmission. Continued community intervention should be undertaken to keep transmission of SARS-CoV-2 under control in cities where higher incidence of COVID-19 cases until the vaccine is available. This study provides a methodological framework that if adopted can assist epidemiologist and policy makers to be well-prepared in advance in cities where PM_2.5 concentration and NO₂ levels are already high in order to minimize the potential risk of further spread of COVID-19 cases.

Keywords: COVID-19; Community interventions; Geoinformation; NO(2); PM(2.5); Pakistan; SIR model; Sentinel-5.

Fig. 1

Locations of PM_2.5 concentrations data sites at Lahore, Karachi, Peshawar and Islamabad US consulates offices and COVID-19 cases distributions in Pakistan February 26 to June 30, 2020 (sources http://covid.gov.pk/stats/pakistan; Pakistan air quality monitoring US-EPA https://aqicn.org/).

Fig. 2

Schematic diagram of susceptibility infected recovered (SIR) model.

Fig. 3

Air quality index during March 15 to June 30, 2020 in Lahore, Karachi, Islamabad and Peshawar.

Fig. 4

Comparison of Tropospheric column of NO₂ (Sentinel 5P) over Lahore, Karachi, Peshawar and Islamabad during community interventions in 2020 versus March 22 to May 30, 2019 (http://www.tropomi.eu/data-products/nitrogen-dioxide).

Fig. 5

Exponential growth model in Punjab, Sindh, Khyber Pakhtunkhwa (KPK) and Islamabad. during February 26 to June 30, 2020.

Fig. 6

SIR model (without NPI) health care system predictions and curve evaluation over Punjab, Sindh, KPK and Islamabad.

Fig. 7

SIR model (with NPI), health care system predictions and curve evaluation over Punjab, Sindh, KPK and Islamabad.