key: cord-0865700-8c4zeozm
authors: Latif, Mohd Talib; Dominick, Doreena; Hawari, Nor Syamimi Sufiera Limi; Mohtar, Anis Asma Ahmad; Othman, Murnira
title: The concentration of major air pollutants during the movement control order due to the COVID-19 pandemic in the Klang Valley, Malaysia
date: 2020-12-15
journal: Sustain Cities Soc
DOI: 10.1016/j.scs.2020.102660
sha: f81617abbbcf9b28c101077ac0e8ad2bb76bb4d6
doc_id: 865700
cord_uid: 8c4zeozm

The COVID-19 pandemic forced many governments across the world to implement some form of lockdown to minimalize the spread of the virus. On 18th March 2020, the Malaysian government put into action an enforced movement control order (MCO) to reduce the numbers of infections. This study aims to investigate the concentrations of air pollutants during the MCO in the Klang Valley. The concentrations of air pollutants were recorded by the continuous air quality monitoring system (CAQMS) operated by the Department of Environment. The results showed that there were significant reductions (p < 0.05) of PM(10), PM(2.5), NO(2) and CO during the MCO compared with the same periods in 2019 and 2018. The highest percentage of reduction during the MCO was recorded by NO(2) with a percentage reduction of between −55 % and −72 %. O(3) concentrations at several stations showed an increase due to the reductions of its precursors such as NO. Further investigation using diurnal patterns of air pollutant concentrations both before and during the MCO showed that NO(2) and CO were both reduced significantly during the rush hours, indicating the reduction in motor vehicles on the roads as a consequence of the MCO influenced the levels of these pollutants.

The pandemic due to Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) or COVID-19 is a tragedy for the human race in this decade. In January 2020, the World Health Organization (WHO) declared the COVID-19 outbreak to be a Public Health Emergency of International Concern as the number of cases of infection was continuing to rise. The number of cases reached approximately 65 million with an estimated 1,507,018 deaths worldwide by the early of December 2020 (WHO, 2020) . The COVID-19 pandemic started in Wuhan, China in late 2019 then spread rapidly across the globe, affecting almost all countries in the world Dong et al., 2020) . The countries with the most infections include the USA, India, Brazil, Mexico, Spain, Russia, Germany, France, Italy, the United Kingdom and China (Gozali, 2020; WHO, 2020) . Despite meteorological factors, air pollution might be one of the environmental drivers that exacerbated the risk of death due to COVID-19 as it may be possible for the virus to be transmitted via aerosols, adsorbed onto particulate matter or be airborne (Comber et al., 2020; Prather et al., 2020; van Doremalen et al., 2020) . Sanità di Toppi et al. (2020) reported that COVID-19 virus was detected in aerosols up to three hours after aerosolization.

As a way of trying to slow down the morbidity and mortality rates of COVID-19, most of the severely affected countries enacted lockdown, where measures such as practising social distancing, the prohibition of public mass gatherings, the cancellation of festivals or events and the restriction of travel within or outside of the country in question were enforced Zambrano-Monserrate et al., 2020) . Factories, industrial sectors and all other forms of economic activity were also forced to cease functioning while restaurants, markets, bars, schools and universities have been closed. Globally, however, there has also been a realization that lockdown may give rise to some positive impacts, particularly on atmospheric air quality Wyche et al., 2021) . Based on monitoring undertaken by satellite technology, the implementation of lockdown across the globe has noticeably reduced severe air pollution rates (Ghahremanloo et al., 2021) . This phenomenon can be clearly seen in some of the world's major cities as reported by Kanniah et al. (2020) , Joshi et al. (2020) , Saadat et al. (2020) , Sharma et al. (2020) and Ramachandran et al. (2020) .

Lockdown reduced the number of vehicles on the roads, especially in major urban areas. Consequently, nitrogen dioxide (NO 2 ) gas, which mainly originates from the combustion of the fuel used for vehicles, has been dramatically reduced due to the reduction in vehicle exhaust emissions (Baldasano, 2020; Cameletti, 2020; Dutheil et al., 2020; Kerimray et al., 2020; Krecl et al., 2020; Lal et al., 2020; Lian et al., 2020; Paital et al., 2020; Pei et al.2020; Sarfraz et al., 2020; Tobías et al., 2020; Wang & Su, 2020) . Wuhan, China is considered the epicentre of the coronavirus crisis and was the first city to impose lockdown measures, starting from 23rd January 2020 (Chong et al., 2020; Leung et al., 2020; Wu et al., 2020) . According to Cole et al. (2020) , the lockdown in China has drastically reduced air pollutant concentrations, particularly NO 2 and particulate matter with an aerodynamic diameter smaller than 2.5 micrometres (PM 2.5 ) by about -63 % (from 38 μg m − 3 to 24 μg m − 3 ) and 35 % (from 62 μg m − 3 to 22 μg m − 3 ), respectively. Another study by Bao and Zhang (2020) reported that the reduction in air pollution in northern Chinese cities was significantly linked with the movement restrictions, as results showed a decreasing trend in air pollutants including NO 2 (24.67 %), particulate matter with an aerodynamic diameter smaller than 10 micrometres or PM 10 (13.66 %), SO 2 (6.76 %), PM 2.5 (5.93 %) and CO (4.58 %). In India, Mahato et al. (2020) revealed that the air quality based on National Air Quality Index in the megacity of Delhi, India, improved by about 40 %-50 % in just four days after lockdown was introduced on 24th March 2020. The concentrations of PM 10 and PM 2.5 have shown the maximum reductions (>50 %) in comparison to the pre-lockdown phase. Studies in other parts of the world such as in Sao Paolo, Brazil by Nakada and Urban (2020) showed there was a drastic reduction of NO (up to − 77.3 %), NO 2 (up to − 54.3 %) and . The only pollutant that showed an increase (around 30 %) in the study was ozone (O 3 ), due to the decrease in nitrogen monoxide (NO) levels during the lockdown period.

In Malaysia, the government implemented the Movement Control Order (MCO) or partial lockdown on 18th March 2020 in response to the COVID-19 crisis. By the early of December 2020, there had been 70,236 confirmed positive cases and 376 deaths. Statistics have shown that the Klang Valley recorded the highest number of confirmed cases, representing more than 40 % of the total cases in Malaysia (Department of Statistics Malaysia, 2020). An earlier study by Kanniah et al. (2020) over Southeast Asia and Malaysia using aerosol optical depth (AOD) observations from the Himawari-8 satellite, column density from Aura-OMI and ground-level continuous air pollutant measurements showed significant reductions of air pollutants such as PM 10 , PM 2.5 , NO 2 , SO 2 and CO during the MCO. A study by Ash'aari et al. (2020) shows that 63 % of Malaysian Department of Environment stations showed significant reductions for PM 2.5 and CO, while all stations showed significant reductions in NO 2 concentrations during the MCO in Malaysia.

The Klang Valley represents a large fraction of the anthropogenic sources that have caused the deterioration of air quality over recent years. However, the urban air quality in these "red zones" may show some significant and positive changes as a result of the Malaysian government's MCO. The MCO might not only break the COVID-19 chain but also drastically reduce environmental pollution due to the reduction of activities and emissions from sources which cause air pollution. The MCO may temporarily result in positive feedback from the environment, and experts should identify the best way to sustain any positive changes in air quality over the long term. Therefore, this study aims to evaluate the changes in air pollutant concentrations in the Klang Valley, the most populated area in Malaysia, before and during the MCO, using data recorded continuously by the Malaysian Department of Environment. A comparison was also made between data recorded during the MCO and data recorded during the same periods of time in 2019 and 2018. The potential meteorological influences were also investigated for comparative purposes between the three different years studied (2018-2020).

The Klang Valley was selected to study the impact of MCO as it is situated in the most populated area in Malaysia. Kuala Lumpur, the capital of Malaysia; Putrajaya, the administrative area of the Malaysian government; and part of Selangor state are located in the Klang Valley region. Currently, the population of the Klang Valley is around eight million and the number of vehicles moving around the Kuala Lumpur area on a normal day prior to the MCO was approximately 45,441 vehicles per day (Ministry of Transport, 2019). During the MCO period, the government was only allowing workers listed as working in the essential sectors to go to work and travel between states in Malaysia. This included any movement within the Klang Valley. All industrial activity was stopped, except industries in essential sectors, which were allowed to continue to operate.

Major air pollutant (PM 10 , PM 2.5 , CO, NO 2 , SO 2 and O 3 ) concentrations and meteorological (temperature, wind speed, wind direction, relative humidity and solar radiation) data during the MCO (between 18th March and 22nd April 2020) were recorded on an hourly basis by the Pakar Scieno TW Sdn Bhd on behalf of the Malaysian Department of Environment. Measurements were taken from six continuous air quality monitoring stations (CAQMS) in the Klang Valley. Mobility data (Google LLC, 2020) were used to represent the traffic volume and industrial activities at the Klang Valley. The monitoring stations used were Batu Muda, Petaling Jaya, Cheras, Shah Alam, Klang and Putrajaya (Fig. 1) . Batu Muda and Petaling Jaya stations are surrounded by busy major roads and are affected by emissions from heavy traffic. Cheras and Shah Alam are located in residential areas and near to highways and road traffic. Klang is located near to a port and industrial areas as well as a coal-fired power plant, which means that Klang is also affected by emissions from heavy traffic. Putrajaya is an administrative area of the Malaysian government. This station is influenced by motor vehicle emissions, especially early in the morning and in the late afternoon due to the movement of government staff arriving for work and then leaving work and returning home. All these monitoring stations are located within school compounds and are affected by urban activities within the Klang Valley. In this study, hourly measurements were used and the air pollutant concentrations recorded at all six stations using similar instruments were also analysed for the time period from 18th March to 22nd April for the past two years (2018 and 2019) for a comparison with the year 2020. Meanwhile, temporal patterns of air pollutant concentrations recorded at the same stations were analysed for the period of 1st January to 22nd April 2020 (before and during the MCO). The changes in mobility in the Klang Valley (represented by Kuala Lumpur and the Selangor state) were analysed using daily measurements (Google LLC, 2020) .

The hourly concentrations of PM 10 and PM 2.5 were measured using a Thermo Scientific tapered element oscillating microbalance (TEOM) 1405-DF (USA) monitor, while the hourly concentrations of CO and O 3 were measured using a Thermo Scientific Model 48i (USA) CO analyser and Thermo Scientific Model 49i (USA) O 3 analyser, respectively. The hourly SO 2 concentrations were measured using a Thermo Scientific Model 43i (USA) SO 2 analyser, while the NO 2 concentrations were measured using a Thermo Scientific Model 42i (USA) NO 2 analyser. Relative humidity and temperature were recorded using a Climatronic AIO 2 Weather Sensor (Climatronic Corporation, USA) (PSTW, 2018a). Due to the constraints of obtaining precipitation data from the Malaysian Meteorological Department (MMD), satellite data from the National Aeronautics and Space Administration (NASA) were used to study the influence of precipitation on air quality in the Klang Valley. The Global Precipitation Measurement (GPM) data with a spatial scale of 0.1 • and temporal scale of 30 min were downloaded and analysed for precipitation in the Klang Valley during the MCO and similar time frames in 2018 and 2019. The study by Mahmud et al. (2017) on the effectiveness of GPM satellite precipitation over Malaysia showed that in Peninsular Malaysia, the GPM data is able to depict spatial rainfall patterns varied by homogeneous rainfall with Spearman's correlation coefficient (r) ranging from 0.591 to 0.891.

All air quality data from CAQMS went through quality assurance and quality control (QA/QC) procedures before submission to the Malaysian Department of Environment. Instruments for the detection of gases were manually calibrated once a fortnight. Flow verification for PM 10 and PM 2.5 measurements using TEOM was conducted once a month, as stipulated in the standard operating procedures for all these instruments. The data removed during the QC checks was predominantly due to insufficient data (in turn due to negative values) and as a result of non-adherence to auto calibration targets. The second level QC checks were observed mainly due to outliers. Some of these observations were verified based on observed sources while others were rejected due to instrument failure (PSTW, 2018a, 2018b). 

Descriptive and statistical analyses in this study were carried out using R statistical analysis version 3.6.3 (R Development Core Team, 2011). The main package used in R statistical analysis was the "Openair" Version 2.6 (Carslaw & Ropkins, 2012) . Daily, diurnal and yearly plots of air pollutant datasets (2018-2020) at six study locations were assessed to examine temporal patterns. In addition to R statistical analysis, the Statistical Package for the Social Sciences (SPSS) software version 21 (IBM, USA) was used for one-way analysis of variance (ANOVA) and t-test. The data for ANOVA and t-test were normalised using the Inversed Distribution Function. The determination of normal distributions were then conducted by Kolmogorov-Smirnov and Shapiro-Wilk tests using SPSS. Variables with significance values that are greater than the alpha value (p > 0.05) were removed from the data set. In order to compare the spatiotemporal variation of air pollutants during the MCO and similar time frames in 2018 and 2019, Inverse Distance Weighting (IDW) was applied in the analysis as the IDW method is a practical and suitable method for spatial air quality modelling when the observation points are lower in number (Halim 

The bivariate polar plots were plotted to show the variation of air pollutant concentrations with wind speed and wind direction for polar coordinates. In a bivariate polar plot, wind speed is plotted as the distance from the origin and wind direction as the angle from the direction of the origin. The different colours in the polar plot are the average concentrations of the pollutant variable (Carslaw & Ropkins, 2012; Dotse et al., 2016; Uria-Tellaetxe & Carslaw, 2014) . In this study, the bivariate polar plots for PM 2.5 , PM 10 , O 3 , NO 2 , SO 2 and CO recorded in Petaling Jaya were presented to show the influence of wind speed and wind direction on the pollutant concentrations which then enabled the potential emission sources to be pin pointed during the MCO compared with 2019 and 2018. Petaling Jaya station was chosen because this station recorded the highest concentration of most air pollutants during the MCO.

Hybrid Single-Particle Lagrangian Integrated Trajectory model (HYSPLIT version 4.9) was employed to calculate 72 h backward trajectories using a 6 h temporal resolution of the model to determine the origin of the air mass arriving at the Petaling Jaya station. The results of the backward trajectory for this station during the MCO were compared with the same periods in 2019 and 2018. HYSPLIT, which is used for backward trajectory analysis, was developed by the National Oceanic and Atmospheric Administration (NOAA)'s Air Resource Laboratory (ARL). The height level selected for model calculation was 500 m to ensure that the trajectories started in the atmospheric boundary layer (Eva & Lambin, 1998) . The meteorological drivers used to compute the trajectories from 2018 to 2020 were obtained from the National Center for Environmental Prediction (NCEP) archive, which is maintained by ARL. In order to have an overview of the arrival of air masses to Petaling Jaya station from the 18th March to the 22nd April in 2018, 2019 and 2020, a clustering method in the HYSPLIT model was used to group the individual trajectories obtained based on curvature and length characteristics (Brankov et al., 1998) .

The average concentrations of hourly air pollutants with the addition of daily maximum O 3 concentration recorded during the MCO and the respective time period in the years 2019 and 2018 are presented in Table 1 . During the MCO, the average concentrations of PM 10 from all stations in the Klang Valley were between 21.4 and 30.6 μg m − 3 . The concentrations of PM 2.5 were recorded as being between 16.1 and 22.8 μg m − 3 . For both PM 10 and PM 2.5 , Cheras recorded the lowest average concentrations, while Klang recorded the highest. In terms of gas concentrations during the MCO, NO 2 , SO 2 and CO were between 3.3 and 8.7 ppb, 0.6 and 1.1 ppb and 0.5 and 0.9 ppm, respectively. The average and maximum values of O 3 ranged from 17.2 to 29.3 ppb and 55.8 to 85.5 ppb, respectively. O 3 was clearly recorded as being at the lowest concentration at Petaling Jaya and NO 2 was recorded as having the highest concentration at Klang. One-way ANOVA analysis on hourly data showed there were significant differences (p < 0.05) between air pollutants (PM 10 , CO, NO 2 and SO 2 ) measured at all six stations during the MCO and the same time periods in 2019 and 2018.

A comparison of the hourly concentrations of air pollutants during the MCO and the same periods in 2019 and 2018 are presented in Fig. 2 (a), (b) and Table 2 . Fig. 2(a) is for stations located within and bordering the Kuala Lumpur administrative area (Batu Muda, Cheras and Petaling Jaya) and Fig. 2(b) is for stations located outside Kuala Lumpur (Shah Table 1 Comparison between the average concentration of hourly data of major air pollutants recorded during MCO and at the same time in 2018 and 2019. The NO 2 reduction ranged from − 55 % to − 72 % between 2020 compared to 2019 and 2018, where Cheras recorded the highest percentage reduction. The second highest reduction was for CO with a range of between − 13 % and − 53 %. The reductions in PM 10 and PM 2.5 ranged between − 10 % and − 46 %, and − 7 % and − 45 %, respectively. The results for PM 10 and PM 2.5 in this study confirmed the previous findings by Kanniah et al. (2020) using AOD observations from Himawari-8 satellite as indicators for particulate matter which indicated a reduction of around 40-60 % of AOD within the urban environment. Lower regional transboundary emissions influence the satellite observations more compared to local ground-level measurements and may account for the differences in the percentages of particulate reduction using both methods. (Table 2) . One-way ANOVA analysis showed there were significant differences (p < 0.05) in daily O 3 maximum concentrations measured during the MCO and at the same periods in 2019 and 2018 at Shah Alam and Cheras stations.

The results from this study indicated that the concentrations of NO 2 , CO, PM 10 and PM 2.5 were significantly reduced during the MCO. The reduction in the number of vehicles caused by the reduction of mobility due to the MCO in the Klang Valley (representing the state of Selangor and Kuala Lumpur) as shown in Supplementary 2 (Google LLC, 2020) has clearly reduced the concentrations of NO 2 and CO. The location of sampling stations near roads with heavy traffic led to the reduction of these two gases when there were low numbers of vehicles on the road. Motor vehicles, as well as industrial, construction and combustion activities, also contribute to the amount of particulate matter such as PM 10 and PM 2.5 in ambient air. The lower numbers of motor vehicles and a limited number of industrial, construction and combustion activities would be expected to reduce the concentrations of PM 10 and PM 2.5 in ambient air. These results support similar previous studies on the levels of NO 2 , CO, PM 10 and PM 2.5 in ambient air recorded during the lockdown in Malaysia from neighbouring countries including Singapore (Li & Tartarini, 2020) and Thailand (Stratoulias & Nuthammachot, 2020) and other parts of the world as suggested by Mahato et al. (2020) , Menut et al. (2020) and Tobías et al. (2020) . The concentrations of SO 2 and O 3 did not follow the patterns of the concentrations of NO 2 , CO, PM 10 and PM 2.5 . The average concentration of SO 2 recorded in the Klang Valley ranged between 0.7 and 3.9 ppb, which can be considered low compared with other gases (Table 1) . A small change of SO 2 emissions in this area can contribute to a significant change in the SO 2 percentage. Even though there are signs of a reduction during the MCO compared with the same period of time in 2019 and 2018, the reduction was almost similar compared with the reduction between years without an MCO, i.e. between 2019 and 2018. The concentration of SO 2 during the MCO is expected to have originated from motor vehicles and other industrial activities as well as natural sources of SO 2 . The variations in emissions, as well as wind direction and other meteorological factors, contribute to the amount of SO 2 in the ambient air in the Klang Valley. There are two power plants located near to the studied stations which are a coal-fired power plant (Klang station) and a gas turbine power plant (Putrajaya station). The power plants could be one of the reasons for the increase in SO 2 at the Klang station (28 %) during the MCO compared with the same period in 2019 and in Putrajaya during MCO compared with the same time period in 2018 (16 %). This result needs further detailed investigation which would involve detailed emission data from anthropogenic and natural sources, as well as the influence of local and regional meteorological factors.

Surface O 3 has been known as a secondary air pollutant that is influenced by its precursors such as NO x and volatile organic compounds (VOCs) Sillman, 1999) . The reduction in O 3 during the MCO compared with the same period in 2019 was significant (p < 0.05) at the suburban stations outside the Kuala Lumpur city centre, which usually recorded high concentrations of O 3 on normal days, namely Cheras and Shah Alam (Ahamad et al., 2014; Banan et al., 2013; Latif et al., 2012) . Other stations which are influenced by heavy traffic and industrial activities such as Batu Muda and Klang showed higher O 3 concentrations (at 31 % and 62 %, respectively) based on comparisons during the MCO and the same period in 2019. A similar increase in Klang (17 %) was also observed during the MCO and the same period in 2018 ( Table 2) .

The increasing and decreasing phenomena of O 3 at the sampling stations was to be expected due to the influence of O 3 precursors in the atmosphere. The limited emissions of NO x , as represented by the NO 2 concentrations, reduced the concentrations of NO. The low rate of titration of NO to O 3 in these areas will automatically increase the O 3 concentration in ambient air as shown by other similar studies during the lockdown in tropical regions (Cazorla et al., 2020; Dhaka et al., 2020; Li & Tartarini, 2020; Stratoulias & Nuthammachot, 2020) and other parts of the world (Collivignarelli et al., 2020; Menut et al., 2020; Sicard et al., 2020; Siciliano et al., 2020; Tobías et al., 2020; Zoran et al., 2020) . This phenomenon usually occurs in busy areas influenced by the emissions of NO from motor vehicles. The process of formation of O 3 is quite complicated and influenced by NO x and other oxidants from other Table 2 Percentage of the average concentration differences between hourly air quality data recorded during precursors (RO x = OH + OH 2 + RO 2 ) from VOCs and CO in the atmosphere. The formation and destruction of O 3 are determined by the formation of O( 3 P) and interaction between O 3 and NO as illustrated in Eqs.

(2)-(6) (Wang et al., 2017) : 

The reduction of the O 3 maximum around midday recorded at almost all stations may be due to the limited amounts of O 3 precursors available to form the O 3 maximum around midday.

Time series of daily concentration of major air pollutants before the MCO (1st January 2020 to 17th March 2020) and during the MCO (18th March 2020 to 22nd April 2020) are presented in Fig. 3(a) and (b) . The independent sample t-test (p < 0.05) results illustrated that all the six air pollutants measured at Cheras, Petaling Jaya and Shah Alam stations showed a significant difference during and before the MCO. In general, there were indications of PM 10 , PM 2.5 , NO 2 and CO reductions recorded at all stations during the MCO. The daily concentrations of NO 2 and CO were clearly significantly reduced (p < 0.05) after the MCO was implemented. The daily average of O 3 (Fig. 3(a) and (b) ) and daily Fig. 4 . Diurnal pattern of major air pollutants before MCO (1st January 2020 to 17th March 2020) and during MCO (18th March 2020 to 22nd April 2020). maximum O 3 (Supplementary 3) showed similar patterns before and during the MCO. The results once again indicated that NO 2 and CO were the two major air pollutants which were clearly influenced by the reduction in motor vehicle emissions and other anthropogenic emissions during the MCO period. The same results were observed in other locations where the effect of lockdown was a reduction of NO 2 (Baldasano, 2020; Cameletti, 2020; Kerimray et al., 2020; Lian et al., 2020; Pei et al., 2020) and CO (Ghahremanloo et al., 2021; Kerimray et al., 2020) . Other pollutants, such as PM and SO 2, are expected to originate from other sources such as soil dust, sea breezes, construction, coal-fired power plants and industrial activities, aside from the use of motor vehicles (Azhari et al., 2018; Keywood et al., 2003; Mohtar et al., 2018; Sulong et al., 2017) . Therefore, PM and SO 2 concentrations were not necessarily reduced by the lower numbers of motor vehicles during the MCO. The lower concentrations of NO 2 as an O 3 precursor does not significantly follow the reduction of O 3 concentrations. The lower concentrations of NO as an O 3 titrant may once again increase the concentration of O 3 recorded at stations in the city centre and busy areas.

The diurnal patterns of major air pollutants are presented in Fig. 4 . Overall, the concentrations of CO, NO 2 , PM 10 , PM 2.5 and SO 2 before the MCO were recorded at the highest concentrations between 7.00 a.m. and 11.00 a.m., while the lowest concentrations were observed around midday and increased again after 5.00 p.m. The concentrations of these pollutants were recorded as being higher at night and then dropping in the early morning. The MCO clearly reduced the concentrations of all these pollutants, particularly during the rush hours, at all stations. The changes in O 3 were observed to be similar for all locations studied both before and during the MCO, with a maximum concentration at around 3.00 p.m. that then rapidly decreased in the late afternoon. The maximum O 3 concentrations around midday at several stations were increased or almost similar during the MCO compared to the period of time before it. Stations located in the city centre and busy areas such as Petaling Jaya and Klang were found to show increasing concentrations of O 3 around midday during the MCO compared with the other stations. Lower concentration O 3 precursors such as NO x were once again expected to contribute to the lower titration process of O 3 in the city centre and busy areas.

The high concentration of air pollutants during the rush hour in the morning shows the influence of motor vehicle emissions (Azmi et al., 2010) . The diurnal pattern of NO 2 and CO during the MCO showed the flattened peak of their concentrations during the rush hour in the morning and late afternoon. This was followed by the lower concentration of these two gases at night. The patterns of lower concentrations during the rush hour and at night were hardly noticeable for PM 10 , PM 2.5 and SO 2 . The concentrations of O 3 were recorded at the highest level around midday due to the intensity of UV radiation that can separate O radicals from O 3 precursors, such NO 2 and VOCs. The low concentrations of NO 2 also mean lower concentrations of NO, which is usually directly emitted by motor vehicles and behaves as the main titrator for O 3 , especially in urban and suburban areas. A similar result was obtained by Tobías et al. (2020) where O 3 was observed to have increased during lockdown in Barcelona, Spain, suggesting that contributing factors were a decrease in NO x and NO concentrations resulting in an increase of temperatures from February to April which in turn led to an increased O 3 concentration.

The time period chosen in this study was during the inter-monsoon season in Malaysia which occurs from March until May. During this period, the wind is weak and comes from a different direction than usual and thunderstorm events frequently occur in the evening (Malaysian Meteorological Department, 2020) . The Klang Valley is located on the western coast of the Malaysian Peninsula and has maximum rainfall cycles during the inter-monsoon period (Jamaluddin et al., 2018) . Strong local convection and precipitation are the other characteristics of the inter-monsoon season in the Klang Valley region (Ooi et al., 2017) . Apart from traffic emissions and industrial emissions, meteorological factors are also one of the key subjects to be examined to understand the distribution and concentration of air pollution (Wang et al., 2006) .

In order to investigate the influence of meteorological factors on the concentrations of air pollutants during the MCO, parameters such as precipitation, relative humidity, temperature and solar radiation were compared with the precipitation, relative humidity, temperature and solar radiation data recorded during the same time periods in 2019 and 2018. The results in Supplementary 4, Supplementary 5, Supplementary 6 and Supplementary 7 show the readings of precipitation, relative humidity, temperature and solar radiation during the MCO (c) and the same periods in 2019 (b) and 2018 (a) where the readings from all stations used in the study are averaged out. In general, precipitation, relative humidity, temperature and solar radiation during the MCO ranged between 0.22 and 76.37 mm day -1 , 68.36 and 87.35 %, 27.0 and 29.9 • C, 37.26 and 188.79 W m -2 , respectively The range of meteorological factors represents the conditions between the end of the northeast monsoon and the inter-seasonal monsoon in Peninsular Malaysia. There were no significant differences (p > 0.05) between the meteorological parameters recorded during the MCO compared with the same parameters recorded in 2019 and 2018. The influence of precipitation, relative humidity, temperature and solar radiation during the MCO compared with the similar data recorded in the previous years can therefore be considered as minimal.

The bivariate polar plot for Petaling Jaya station (Fig. 5) Further investigation of the Petaling Jaya station, using the backward trajectories analysis by HYSPLIT (Fig. 6) , showed that all of the air masses originated from the northeast of the Malaysian Peninsula. The backward trajectories during the MCO in 2020 had almost the same patterns as the backward trajectories recorded in 2019 and 2018. The wind coming from the South China Sea may be influenced by natural and anthropogenic sources after it has passed through mainland areas on the eastern coast and central region of the Malaysian Peninsula. 

This study investigated the concentrations of major air pollutants recorded during the MCO in the Klang Valley, Malaysia. The results indicated that there were significant reductions (p < 0.05) in the levels of major air pollutants such as PM 10 , PM 2.5 , NO 2 and CO during the MCO compared with the same periods in 2019 and 2018. The highest reduction was recorded for NO 2 , of between − 55 % and − 72 %. There were fluctuations of O 3 and SO 2 concentrations during the MCO compared with the concentrations from the same period in 2019 and 2018, based on the location of the stations. The lower concentrations of NO during the MCO influenced the increase in O 3 concentration due to the limitation of titration processes.

The diurnal pattern of air pollutants showed that air pollutants such as NO 2 and CO were clearly influenced by motor vehicle emissions. The maximum concentrations of O 3 at around midday showed an increase at several stations influenced by motor vehicles when compared with the concentrations before the MCO. This study shows that the concentrations of air pollutants in the Klang Valley were reduced during the MCO in line with the lower numbers of motor vehicles on the roads. Clean ambient air can ultimately be achieved through the reduction of emissions from motor vehicles and other anthropogenic activities which originate from local sources. Limitations of people's mobility in urban environments such as in the Klang Valley will help the government to reduce the concentrations of major air pollutants significantly. Several initiatives to limit the number of motor vehicles such as by encouraging people to use public transport and sharing cars for their mobility are really important to reduce air pollution. Government and private sectors also need to take the initiative to encourage workers to work from home to reduce emissions in urban areas. Emissions control through working from home can be conducted to reduce local and regional air pollutants during air quality episodes such as haze, which occurs in Southeast Asia in the Klang Valley almost every year. Fundamentally, more detailed studies need to be conducted on how to reduce surface O 3 in ambient air within urban and sub-urban area in the Klang Valley with emphasise on its precursors' chemical interactions.

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.

Variation of surface ozone exceedance around Klang Valley

Spatiotemporal variations and contributing factors of air pollutant concentration in Malaysia during movement control order due to the pandemic COVID-19

Road traffic as an air pollutant contributor within an industrial park environment

Trend and status of air quality at three different monitoring stations in the Klang Valley

COVID-19 lockdown effects on air quality by NO2 in the cities of Barcelona and Madrid (Spain). The Science of the Total Environment

Characteristics of surface ozone concentrations at stations with different backgrounds in the Malaysian Peninsula

Does lockdown reduce air pollution? Evidence from 44 cities in northern China

A trajectory-clustering-correlation methodology for examining the long-range transport of air pollutants

The effect of corona virus lockdown on air pollution: Evidence from the City of Brescia in Lombardia Region (Italy)

Openair -an R package for air quality data analysis

What the COVID-19 lockdown revealed about photochemistry and ozone production in Quito

Monitoring disease transmissibility of 2019 novel coronavirus disease in Zhejiang

The impact of the Wuhan Covid-19 lockdown on air pollution and health: A machine learning and augmented synthetic control approach. Discussion Papers 20-09

Lockdown for CoViD-2019 in Milan: What are the effects on air quality? The Science of the Total Environment

Airborne transmission of SARS-CoV-2 via aerosols

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

Living environment matters: Unravelling the spatial clustering of COVID-19 hotspots in Kolkata megacity

-19 current situation in Malaysia

diminution and haze events over Delhi during the COVID-19 lockdown period: An interplay between the baseline pollution and meteorology

An interactive web-based dashboard to track COVID-19 in real time

Influence of southeast asian haze episodes on high PM 10 concentrations across Brunei Darussalam

COVID-19 as a factor influencing air pollution? Environmental Pollution

Remote sensing of biomass burning in tropical regions: Sampling issues and multisensor approach

Impact of the COVID-19 outbreak on air pollution levels in East Asia

COVID-19 community mobility report

Singapore's COVID-19 cases highest in Southeast Asia

Spatial assessment of land use impact on air quality in mega urban regions

Investigating the mechanisms of diurnal rainfall variability over Peninsular Malaysia using the non-hydrostatic regional climate model

Reduction of atmospheric effects in satellite images during the COVID-19 induced lockdown

Air quality index prediction using IDW geostatistical technique and OLS-based GIS technique in Kuala Lumpur

COVID-19's impact on the atmospheric environment in the Southeast Asia region

Assessing air quality changes in large cities during COVID-19 lockdowns: The impacts of traffic-free urban conditions in Almaty

Haze in the Klang Valley of Malaysia

Drop in urban air pollution from COVID-19 pandemic: Policy implications for the megacity of São Paulo

Temporary reduction in fine particulate matter due to 'anthropogenic emissions switch-off' during COVID-19 lockdown in Indian cities

The dark cloud with a silver lining: Assessing the impact of the SARS COVID-19 pandemic on the global environment

Variations of surface ozone concentrations across the Klang valley

First-wave COVID-19 transmissibility and severity in China outside Hubei after control measures, and second-wave scenario planning: A modelling impact assessment

Changes in air quality during the COVID-19 lockdown in Singapore and associations with human mobility trends

Effect of ambient air pollution on premature SGA in Changzhou city

Impact of city lockdown on the air quality of COVID-19-hit of Wuhan city

Characterization and sources of volatile organic compounds (VOCs) and their related changes during ozone pollution days in 2016 in Beijing

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

How effective is the new generation of GPM satellite precipitation in characterizing the rainfall variability over Malaysia?

Iklim Malaysia

Impact of lockdown measures to combat COVID-19 on air quality over western Europe

Ministry of Transport

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

COVID-19 pandemic: Impacts on the air quality during the partial lockdown in São Paulo state, Brazil. The Science of the Total Environment

Interaction of urban heating and local winds during the calm intermonsoon seasons in the tropics

Inter nation social lockdown versus medical care against COVID-19, a mild environmental insight with special reference to India

Response of major air pollutants to COVID-19 lockdowns in China

Spatiotemporal variation of air pollutants and the impact of anthropogenic effects on the photochemical buildup of ozone across Delhi-NCR

Standard operational procedure for continuous air quality monitoring (CAQM). operation, scheduled maintenance, troubleshooting. Calibration And Verification

Standard operational procedure for continuous air quality monitoring (CAQM) data validation

R: A language and environment for statiscal computing. Austria: R foundation for statistical computing Vienna

Aerosol-induced atmospheric heating rate decreases over South and East Asia as a result of changing content and composition

Environmental perspective of COVID-19. The Science of the Total Environment

Novel coronavirus: How atmospheric particulate affects our environment and health

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

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. The Science of the Total Environment

The relation between ozone, NOx and hydrocarbons in urban and polluted rural environments

Air quality development during the COVID-19 pandemic over a medium-sized urban area in Thailand

Source apportionment and health risk assessment among specific age groups during haze and non-haze episodes in Kuala Lumpur, Malaysia. The Science of the Total Environment

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

Conditional bivariate probability function for source identification

Aerosol and surface stability of SARS-CoV-2 as sompared with SARS-CoV-1

A preliminary assessment of the impact of COVID-19 on environment -A case study of China

Ozone pollution in China: A review of concentrations, meteorological influences, chemical precursors, and effects. The Science of the Total Environment

The variation of characteristics and formation mechanisms of aerosols in dust, haze, and clear days in Beijing

Coronavirus disease (COVID-2019): Situation report -10

Does Wuhan need to be in lockdown during the Chinese Lunar New Year?

Changes in ambient air quality and atmospheric composition and reactivity in the South East of the UK as a result of the COVID-19 lockdown

Indirect effects of COVID-19 on the environment

Assessing the relationship between surface levels of PM 2.5 and PM 10 particulate matter impact on COVID-19 in

We thank Universiti Kebangsaan Malaysia for Research University Grant (GUP-2018-109) and the Department of Environment, Malaysia, for air quality data. Special thanks to Ms K Alexander and Dr Rose Norman for proofreading this manuscript.

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.scs.2020.102660.