key: cord-1023652-sfp63o3j
authors: Thomas, Jobin; P.J., Jainet; Sudheer, K. P.
title: Ambient air quality of a less industrialized region of India (Kerala) during the COVID-19 lockdown
date: 2020-11-28
journal: nan
DOI: 10.1016/j.ancene.2020.100270
sha: 542e277b5d9dc2a2f2825a80be47d3167d76e1bd
doc_id: 1023652
cord_uid: sfp63o3j

This study assesses the effect of lockdown, due to the coronavirus disease (COVID-19) pandemic, on the concentration of different air pollutants and overall air quality of a less industrialized region (Kerala) of India. We analysed data from four ambient air quality stations over three years (January to May, 2018-2020) with pairwise comparisons, trend analysis, etc. Results indicated unprecedented reduction in the concentration of the air pollutants: nitrogen dioxide, NO2 (-48%), oxides of nitrogen, NOx (-53% to -90%), carbon monoxide, CO (-24% to -67%) and the particulate matter (-24% to -47% for particulate matter with a diameter of less than 2.5 μm, PM2.5; -17% to -20% for particulate matter with a diameter of less than 10 μm, PM10), as well as the reduction of the National Air Quality Index (NAQI). These reductions indicate an overall improvement of the ambient air quality due to restrictions on transportation, construction, and the industrial sectors during lockdown, even in an area considered less industrial. Despite the general decreasing trend of the concentration of various air pollutants from January to May, suggesting seasonal influences, the trend was intensified in the year 2020 due to the added effect of the lockdown measures. Comparison of results with those from larger and more industrialized cities suggests that the effects of lockdown are more variable, and focused on the levels of gaseous pollutants. Findings from this study demonstrate the far-reaching effects and implications of the COVID-19 lockdown on ambient air quality, even on less industrialized and less urbanized regions.

The rapid spread of the COVID-19 pandemic caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) induced inenarrable social and economic impacts across the globe.

The COVID-19 was first detected in Wuhan (China) and the disease had affected more than 15 million people in 216 countries until 26 th July 2020 (https://covid19.who.int). The first COVID-19 case in India was reported on 30 th January 2020 in Kerala State, followed by a few more in the first week of February 2020, with escalating cases since the second week of March 2020. As of 26 July 2020, the Government of India had reported more than 1.4 million confirmed cases and 32,771 deaths (https: //www.mygov.in/covid-19) . Kerala was one of the most affected states of India during the initial stages of the COVID-19 outbreak in the country.

In response to the COVID-19 outbreak, various countries implemented diverse nonpharmaceutical interventions (e.g., personal protection and hygiene, physical distancing, environmental and travel-related measures, etc.) to slow down and to reduce the mortality rates associated with the COVID-19, with the ultimate objective to reach and to maintain the state of low-level or no transmission. In addition to these efforts, many countries imposed large-scale public health and social measures, including restriction of private and public transportation, suspension of educational/commercial/business/religious activities, and geographical area quarantine (often collectively referred to as lockdown) to curtail the spread of the highly transmissible virus within the society (World Health Organisation, 2020). The preventive lockdown was first implemented in Wuhan on 23 January 2020 and subsequently extended for entire China for at least three weeks (Le et al., 2020) . Later, several countries across the globe (e.g., Belgium, Italy, Spain, Germany, UK, South Africa, Argentina, Colombia, USA, Israel, New Zealand) implemented the lockdown for varying periods.

Kerala (India) implemented the state-wide lockdown on 24 March 2020, as numbers of COVID-19 cases drastically increased in the State (Table S1 ; Fig. 1 ). The Government of India also J o u r n a l P r e -p r o o f enforced the nation-wide lockdown from 25 March 2020 onwards. Later, the nation-wide lockdown was extended further till the end of May 2020, with varying exemptions in different parts of the country. The lockdown implemented in Kerala was effective against the transmission of the virus, as the COVID-19 cases (registered during March, April and the first week of May) were almost completely recovered by 9 May 2020 (Fig. 1) . The rising number of cases since 10 May 2020, however, was attributed to the controlled re-entry of the natives of Kerala stranded in other parts of the country and the globe. As of 26 July 2020, the State Government of Kerala had reported 19,025 confirmed cases and 61 deaths (https://dashboard.kerala.gov.in).

Implementation of the lockdown restrictions has resulted in the stagnation of economic activities, such as construction, industrial projects, transportation, etc., leading to rising unemployment, decrease in income generation, and reduced consumer activity. The growth rate of the Gross Domestic Product (GDP) of India in the fourth quarter of the fiscal year 2020 dropped to 3.1% compared to the previous quarters (4.1-5.2%), mainly due to the effect of the COVID-19 on the country's economy (http://mospi.gov.in).

Despite the infliction on the economic sector, the lockdown measures have unintentionally bring forth benefits for the environment in improved air quality. During the lockdown period, across the globe, emissions from the transportation and industrial sectors were remarkably lower leading to a significant reduction in the levels of environmental pollution (Arora et al., 2020; Muhammad et al., 2020; Paital, 2020) . Numerous researchers (e.g., Collivignarelli et al., 2020; Dantas et al., 2020; He et al., 2020; Kanniah et al., 2020; Chin et al., 2020) have analysed data on environmental pollution (mostly air pollution) of different parts of the globe and have reported improvement in the ambient air quality levels during the COVID-19 lockdown. For India, researchers have similarly quantitied the impacts of the nation-wide lockdown on the ambient air quality, using observed data from the ambient air quality monitoring stations Sharma et al., 2020) as well as satellite-derived/reanalysis products (Gautam, 2020) or a combination of both (Lokhandwala & Gautam, 2020; Selvam et al., 2020) .

J o u r n a l P r e -p r o o f climate. Kerala receives a mean annual rainfall of about 2,800 mm, of which, 70% occurs as the Indian summer monsoon rainfall (Thomas & Prasannakumar, 2016 In general, the levels of air pollutants (except particulate matter) in Kerala are within the permissible limit prescribed by the national ambient air quality standards (http://kerenvis.nic.in).

The concentration of various air pollutants, however, especially particulate matter, in Ernakulam and Kozhikode districts was reported nearing the alarming level in the recent years (Government of Kerala, 2020a). Tobollik et al. (2015) observed a considerable health burden for the population living in urban Kerala mainly due to particulate matter with a diameter of less than 2.5 μm and 10 μm (PM2.5 and PM10) concentrations, albeit local air quality standards being met, and reported that a decrease of 10% in particulate matter concentrations may save 15,904 (95% confidence interval: 11,090-19,806) life years. The industries and vehicular traffic are the major drivers for the deterioration of air quality of the State. The industrial sector of Kerala is not well-developed compared to other states of India, which is evinced by the substantially lower figures of the per-capita domestic product and per-capita manufacturing value added in the State (Thomas, 2005) . Among the different states of India, Kerala ranks 12 th position in terms of J o u r n a l P r e -p r o o f using the processed air quality data. The NAQI uses NO2, NH3, SO2, CO, O3, PM2.5, PM10 and lead (Pb) for computing the index. The index requires the concentration of a minimum of three pollutants, including at least PM2.5 or PM10. We converted the concentration of each air pollutant (Cp) to a normalized number (i.e., sub-index, Ip) using segmented linear functions (Eq. 1). 

The NAQI contains six classes: good (0-50), satisfactory (51-100), moderate (101-200), poor (201-300), very poor (301-400) and severe (401-500).

The analysis utilized daily data (24 hours) from 1 January to 31 May of 2018, 2019 and 2020.

We considered the period between 1 January and 23 March the pre-lockdown period, and the duration from 24 March to 17 May as the lockdown period. Following this classification, the data between 1 January 2020 and 23 March 2020 represent the pre-lockdown period (PLD) 2020, while the data from 24 March 2020 to 17 May 2020 are termed as the lockdown (LD) 2020.

Similarly, we considered the daily average of the air quality data of 2018 and 2019 between 1 January and 23 March as the PLDmean, and we treated the daily average of the data between 24

March and 17 May as the LDmean. We computed descriptive statistics of the different air pollutants for the PLD 2020, LD 2020, PLDmean and LDmean for pairwise comparisons. Since the air quality data rarely follow the normal distribution, we applied the Mann-Whitney test for analysis of statistical significances between groups, as the Mann-Whitney test is applicable for differences in medians as well as in the shape and spread of the distributions.

We tested the daily time series of each pollutant for the entire period (i.e., 1 January to 31 May) of all years (i.e., 2018, 2019 and 2020) for presence of monotonic trends using the Mann-Kendall test (Kendall, 1975; Mann, 1945) . Since the Mann-Kendall test assumes that the data are J o u r n a l P r e -p r o o f independent and identically distributed, we tested the data series for serial correlation. We applied the Mann-Kendall test to the original time series, when a significant autocorrelation lacks at the 5% level, and to the pre-whitened series when a significant serial correlation occurred (von Storch, 1999) .

We compared changes in the ambient air quality data during the LD 2020 in Kerala State with the changes in the levels of various air pollutants recorded in the urbanized and industrialized regions of India as well as across the globe during the lockdown period. This comparison evaluated differences in the magnitude of changes between the urbanized/industrialized and less urbanized/industrialized regions.

The various air pollutants measured at the different ambient air quality monitoring stations of Kerala show notable temporal variability during January 2020 to May 2020 (Fig. 3) . The wider temporal variations are noted in PM2.5 and PM10 at K1, NOx and NH3 at K2, NO, NO2, NOx and NH3 at K3 and CO at K4. In general, all the air quality monitoring stations of Kerala witnessed a gradual reduction in most of the air pollutants during the lockdown window.

(PLD 2020 vs. LD 2020) NAQI of all the stations shows significant differences between the PLD 2020 and LD 2020 (p ≤ 0.001) (Fig. 4 ). On average, the NAQI of K2 and K4 during the LD 2020 shows a reduction by 43%, whereas the NAQI of K1 is decreased by 26% compared to the NAQI of the PLD 2020 (Table 2 ). The reduction in the NAQI implies an improvement in the overall air quality. Following the classification of the NAQI, the PLD 2020 is characterized by moderate to satisfactory air quality in all the stations, wheras the LD 2020 is dominated by good to satisfactory air quality. All the monitoring stations do follow this trend, whereas differences in the NAQI between the periods are less in K3 compared to other stations. Although most of the air pollutants show a significant decrease in their concentration during the LD 2020 (compared to the PLD 2020), O3

at K4 exhibits a significantly higher concentration during the LD 2020 (p ≤ 0.001) (Fig. S1d ).

Comparison of the average concentration of the air pollutants, as well as the average NAQI,

indicates that the air quality of the major cities of Kerala during the LD 2020 is significantly different compared to the PLD 2020.

Although the LD 2020 showed a decrease of the concentration of the air pollutants in Kerala State, the reduction in the concentration (due to lockdown) cannot be determined without estimating the role of short-term fluctuations or the seasonal trends in the concentration of the Figure S2 indicates that the number of days exceeding the national standards for PM2.5 and PM10 also shows a declining pattern from 2018 to 2020 for both the pre-lockdown and lockdown phases but in differing magnitudes. As definite trends are notable within the concentration of the air pollutants during the analysis period, the estimation of the change in the concentration of the pollutants due to the effect of lockdown should consider this trend also.

Since where, ΔPx is the relative change in the concentration of any given air pollutant during the LD 2020 with respect to the PLD 2020, = 2020 and = 2020 .

The ΔPx of most of the air pollutants in all the stations shows a remarkable reduction of the concentration in the LD 2020 with respect to the PLD 2020 (Fig. 5) . The significant reduction in the concentration of the pollutants, such as NO2 (-48%) NOx (-53% to -90%), CO (-24% to -67%) as well as the particulate matter (-24% to -47% for PM2.5, -17% to -20% for PM10) is correlated to the decreased emissions from the transportation and industrial sources during the lockdown period. One may notice that the rate of reduction in the concentration of the air pollutants is non-uniform across the State, implying the role of regional socio-economic, meteorological and anthropogenic factors controlling air quality.

Among the different stations, K1 exhibits a comparatively lower percentage of reduction of the concentration of the various air pollutants (Fig. 5a) . The NOx and SO2 concentration at K2, NOx and NH3 concentration in K3, NOx, SO2, and CO in K4 show significant reduction (> 50%) in their concentration during the LD 2020. The concentration of the particulate matter is reduced only in K1 and K2 (Fig. 5a, b) , however, the concentration of the particulate matter in a few stations (K3 and PM2.5 in K4) shows an increase during the LD 2020 (Fig. 5c, d) . The relative change of the NAQI of the different stations suggests that the improvement in the air quality is significant only in K2 (-38%) and K4 (-15%), wheras the change in the NAQI of K1 is insignificant. The relative change in K3 shows an increase (+43%), however, which is probably due to the higher background and residual pollution.

The improvement of the air quality of Kerala State during the LD 2020, though varying in magnitude, results from the combined effect of the lockdown measures (i.e., restrictions in the J o u r n a l P r e -p r o o f transportation sector and cessation of construction and industrial activities) and the seasonal trends. Among the different air pollutants, the most significant and widespread reduction is observed for CO (-24 to -67%), whereas the reduction in the concentration of NOx (-53 to -90%) and SO2 (-58 to -84%) is also apparent in some of the stations (Fig. 5) . Since the air quality in India has been arguably improving consistently since 2018 (Narain et al., 2020) , the improvements in air quality in 2020 could have also been a continuation of the trend. Kerimray et al. (2020) and Zangari et al. (2020) have discussed the significance of this synergism on the improvement of air quality. As a less industrialized and less urbanized States of India, Table S2 shows a comparison of changes in the air quality of Kerala due to the lockdown with various studies worldwide from more urbanized and industrialized areas . to 70%, -8 (i.e., increased by 8%) to 13%, and -1 (i.e., increased by 1%) to 40%, respectively. Contrary to the general reductions in the air pollutant levels across Kerala, a few stations exhibited a substantial increase in the concentration of NH3 (5-100% at K1, K2 and K4), SO2 (36-85% at K1 and K3), O3 (3% at K1 and 224% at K4), PM2.5 (> 50% at K3 and K4) and PM10 (40% at K4) during the LD 2020 (Fig. 5) . The major source of NH3 emissions is agriculture, while minor sources include industrial processes, vehicular emissions and volatilization from soils and oceans (Behera et al., 2013) . Hence, the increase in the concentration of NH3 in Kerala State during the LD 2020 indicates the contributions from the sources other than the industrial and transportation sector. On the other hand, the major urbanized/industrialized areas of India (e.g., Kolkata, Delhi, Chennai) observed a considerable reduction in NH3 concentration due to the shutdown of J o u r n a l P r e -p r o o f transportation sector (Bedi et al., 2020; . One may also note that K3 showed a significant reduction (-67%) during the LD 2020, which could be attributed to the cessation of the industrial activities. The SO2 levels at K1 and K3 showed a remarkable increase during the LD 2020 (Fig. 5a, c) . Although the lockdown exerted cessation of the industrial operations, the relative increase in SO2 levels could be due to the high background/residual pollution and/or due to regional factors (e.g., Kerimray et al., 2020; Wang et al., 2020) . Shehzad et al. (2020) observed a faint trail of NO2 emission in the major maritime routes of the India Ocean, implying the presence of active marine traffic during the period. Hence, the onshore transport of SO2 due to marine traffic may also influence the concentration of SO2 of Kerala State.

As a departure from the general behaviour of O3 across Kerala State, K1 and K4 recorded an increase in O3 concentration during the LD 2020. Similar observations were reported by various researchers (e.g., Sicard et al., 2020; Siciliano et al., 2020; Tobías et al., 2020) and is attributed to the decrease of NOx concentration along with the increase in the reactivity of the volatile organic compounds (VOC) mixture, the decrease of NO and corresponding reduction of the O3 consumption (i.e., titration) and the higher rate of insolation and increased temperatures (Tobías et al., 2020) . In general, the bivariate relationship between

NOx and O3 at K4 is negatively correlated (Fig. S3) implying that the increase in O3 levels (224%) at K4 could be associated with the decrease in the NOx concentration (i.e., -68%), while K1 and K3 show hardly any definite relationships. Although the general pattern of PM2.5 and PM10 can presumably reflect the reduction of transportation and industrial activities, K3 and K4 recorded a significant increase of particulate matter during the LD 2020 (Fig. 5c, d) . The anomaly at K3 and K4 could relate to various reasons, such as emission from the sources other than transportation and industrial sectors (e.g., domestic/residential sectors, burning of biomass, log-range transport, etc.) and a higher degree of background/residual pollution, and the contributions from these sources might have offset the reduction of the particulate matter due to the lockdown Otmani et al., 2020; Sicard et al., 2020) . Since the transboundary transport is postulated for the increased levels of air pollutants of Kerala State, the role of long-range transport on the air quality was investigated by analysing the air mass trajectories for the LD 2020 by HYbrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) transport model (Rolph et al., 2017; Stein et al., 2015) . The back-trajectory analysis at K3 and K4 indicates the movement of air mass mostly from the Arabian Sea and the eastern parts of peninsular India (Fig. 6) . Contrastingly, K1 receives air parcels from multiple regions including the Arabian Sea, the Bay of Bengal, the Indian Ocean and the northern parts of Sri Lanka. However, the seasonal effects on the transport of the pollutants are not addressed in this study. Although the effect of lockdown in Kerala State is manifested as the reduction in the concentration of most of the pollutants, the long-range transport of the pollutants may act as a limiting factor for further improvement of air quality.

Results of this study help understand the differences in the response of the ambient air quality to short-term human interventions in less industrialized and less urbanized regions to recognize, monitor, and prioritize potential public health concerns and opportunities for action, which are beneficial to develop appropriate policy measures considering reductions in the concentration of the primary and the secondary pollutants as well as the background/residual pollution levels.

Although the unprecedented restrictions on the transportation, construction and industrial sectors caused serious negative effects on the economy, the lockdown offered an incomparable opportunity to investigate the role of emissions from various sectors controlling the ambient air quality. The improvement of the ambient air quality due to the lockdown measures seems to help develop better ambient air quality management programmes, especially in the J o u r n a l P r e -p r o o f urbanised/industrialised areas but has differing implications in the less-industrialised/lessurbanised regions.

This study addressed the effect of lockdown on the ambient air quality of Kerala (India)

considering the temporal variability and trends in the air quality data. This study has three major limitations. The first limitation is a focus of the analysis on the major pollutants, such as NO, NO2, NOx, NH3, SO2, CO, O3, PM2.5, PM10, but all the stations (except K3) did not have the comprehensive data of the pollutants. The second limitation is the lack of long-term air quality data for the monitoring stations. We have analysed the trends and estimated the averages of the pollutants based on the data of the previous two years (2018 and 2019). If data were available for a quite long period, however, the estimates of the average concentration of the pollutants as well as the seasonal trends would have been more reliable. Thirdly, unavailability of meteorological data at the ambient air quality stations limited the present study to understand the role of local meteorological variables on the ambient air quality. We have analysed the rainfall, minimum and maximum temperature, and relative humidity of January to May (2018 to 2020) recorded by India Meteorological Department at Thiruvananthapuram, however, which indicates very few significant differences in the meteorological conditions between 2020 and the previous years.

In summary, examination of air pollutants in Kerala between the pre-lockdown and lockdown periods, seasonal trends in air pollutants, along with comparison with data from more industrialized/urbanized areas, furnished the following answers to the research questions posed in this study. First, the levels of different air pollutants showed a significant reduction during the lockdown period compared to the period before lockdown. The sizeable reductions in concentrations of the pollutants, including NO2 (-48%) NOx (-53% to -90%), CO (-24% to -67%) as well as the particulate matter (-24% to -47% for PM2.5, -17% to -20% for PM10), correlates with decreased emissions from transportation and industrial sources during the lockdown period.

Second, a significant decreasing trend in air pollutant levels of Kerala State from January to May over three years ( J o u r n a l P r e -p r o o f NOx -2021 -2088 -1873 -3.28** -3.43*** -3.01** -1885 -1945 -191 -3 .06** -3.16** -0.31

-454 1713 -328 -0.77 2.81** -0.53 -2007 -333 -2259 -3 .26** -0.54 -3.74***

-2126 -1372 1497 -3.45*** -2.23* 2.41* PM2.5 -1358 -1359 -2439 -2.21* -2.21* -3.92***

-1766 -2045 -1825 -2.90** -3.32*** -2.93** ND-Not tested due to multiple data gaps Confidence levels: 95% (*), 99% (**) and 99.9% (***)

Coronavirus lockdown helped the environment to bounce back

Assessment of Air Quality Changes in the Four Metropolitan Cities of India during COVID-19 Pandemic Lockdown

Ammonia in the atmosphere: A review on emission sources, atmospheric chemistry and deposition on terrestrial bodies

National Air Quality Index. Central Pollution Control Board, Ministry of Environment, Forest and Climate Change

Seasonal variations of atmospheric pollution and air quality in Beijing

Pandemics and the future of human-landscape interactions

Dispersion of atmospheric air pollution in summer and winter season

Lockdown for CoViD-2019 in Milan: What are the effects on air quality?

The impact of COVID-19 partial lockdown on the air quality of the city of Rio de Janeiro, Brazil

The Influence of COVID-19 on Air Quality in India: A Boon or Inutile

Air quality changes during the COVID-19 lockdown over the Yangtze River Delta Region: An insight into the impact of human activity pattern changes on air pollution variation

Indirect impact of COVID-19 on environment: A brief study in Indian context

Short-term exposure to ambient air quality of the most polluted Indian cities due to lockdown amid SARS-CoV-2

Effect of lockdown amid COVID-19 pandemic on air quality of the megacity Delhi

Nonparametric Tests Against Trend

Variation of major air pollutants in different seasonal conditions in an urban environment in Malaysia

COVID-19 pandemic and environmental pollution: A blessing in disguise?

COVID-19 pandemic: Impacts on the air quality during the partial lockdown in São Paulo state

In India, air quality has been improving despite the COVID-19 lockdown

Impact of Covid-19 lockdown on PM10, SO2 and NO2 concentrations in Salé City (Morocco)

Nurture to nature via COVID-19, a self-regenerating environmental strategy of environment in global context. The Science of the Total Environment

Air quality during the COVID-19: PM2.5 analysis in the 50 most polluted capital cities in the world

Real-time Environmental Applications and Display sYstem: READY. Environmental Modelling and Software

The impact of COVID-19 as a necessary evil on air pollution in India during the lockdown

SARS-CoV-2 pandemic lockdown: Effects on air quality in the industrialized Gujarat state of India

Effect of restricted emissions during COVID-19 on air quality in India

Amplified ozone pollution in cities during the COVID-19 lockdown

Increased ozone levels during the COVID-19 lockdown: Analysis for the city of Rio de Janeiro, Brazil

NOAA's HYSPLIT atmospheric transport and dispersion modeling system

Rural and urban differences in air quality

Kerala's industrial backwardness: A case of path dependence in industrialization? World Development

Temporal analysis of rainfall (1871-2012) and drought characteristics over a tropical monsoon-dominated State (Kerala) of India

Changes in air quality during the lockdown in Barcelona (Spain) one month into the SARS-CoV-2 epidemic

Burden of outdoor air pollution in Kerala, India-a first health risk assessment at state level

Misuses of Statistical Analysis in Climate Research

Unexpected rise of ozone in urban 1723

Considerations in adjusting public health and social measures in the context of COVID-19. World Health Organisation

Air quality changes in New York City during the COVID-19 pandemic

The authors thank the Kerala State Pollution Control Board, Government of Kerala (specifically Er. Bindhu Radhakrishnan, Senior Environmental Engineer) for providing the ambient air quality data of the air quality monitoring stations of Kerala. The authors are grateful to the India Meteorological Department for providing the meteorological data.J o u r n a l P r e -p r o o f

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.