key: cord-0742784-ll1kc7sn
authors: Kerimray, Aiymgul; Baimatova, Nassiba; Ibragimova, Olga P.; Bukenov, Bauyrzhan; Kenessov, Bulat; Plotitsyn, Pavel; Karaca, Ferhat
title: Assessing air quality changes in large cities during COVID-19 lockdowns: The impacts of traffic-free urban conditions in Almaty, Kazakhstan
date: 2020-05-04
journal: Sci Total Environ
DOI: 10.1016/j.scitotenv.2020.139179
sha: 10096cf5f6a86de9362a5085c2c9c9d1e65c007b
doc_id: 742784
cord_uid: ll1kc7sn

Abstract Number of cities worlwide experienced air quality improvements during COVID-19 lockdowns; however, such changes may have been different in places with major contributions from nontraffic related sources. In Almaty, a city-scale quarantine came into force on March 19, 2020, which was a week after the first COVID-19 case was registered in Kazakhstan. This study aims to analyze the effect of the lockdown from March 19 to April 14, 2020 (27 days), on the concentrations of air pollutants in Almaty. Daily concentrations of PM2.5, NO2, SO2, CO, O3, and BTEX were compared between the periods before and during the lockdown. During the lockdown, the PM2.5 concentration was reduced by 21% with spatial variations of 6–34% compared to the average on the same days in 2018–2019, and still, it exceeded WHO daily limit values for 18 days. There were also substantial reductions in CO and NO2 concentrations by 49% and 35%, respectively, but an increase in O3 levels by 15% compared to the prior 17 days before the lockdown. The concentrations of benzene and toluene were 2–3 times higher than those during in the same seasons of 2015–2019. The temporal reductions may not be directly attributed to the lockdown due to favorable meteorological variations during the period, but the spatial effects of the quarantine on the pollution levels are evidenced. The results demonstrate the impact of traffic on the complex nature of air pollution in Almaty, which is substantially contributed by various nontraffic related sources, mainly coal-fired combined heat and power plants and household heating systems, as well as possible small irregular sources such as garbage burning and bathhouses.

J o u r n a l P r e -p r o o f 3 quarantine) were identified using satellite data from NASA and the European Space Agency (ESA) (Earth Observatory, 2020) . (Tobías et al., 2020) also depicted substantial air quality improvements after two weeks of lockdown in Barcelona (Spain) . The results support the idea that air pollution could be substantially improved in cities where transport was a major source.

However, the air quality improvements during COVID-19 lockdowns may not clearly favor improving the air quality in areas with a more complex mix of sources, where transport emissions have minor impacts compared to emissions from other sources (e.g., coal combustion for power and heating).

Concerning the levels of BTEX, Almaty is among the most polluted cities in the world (Carlsen et al., 2018) . In terms of priority pollutants, it is one of the most polluted cities of Kazakhstan (Kerimray et al., 2019) , and there were 21 days in 2018 on which the PM 2.5 concentrations exceeded 250 µg m -3 at least at one station (Kerimray et al., 2020) . The wintertime concentrations of major atmospheric air pollutants were several times higher than those during summertime, which could be explained by intensive coal combustion at power plants and in households for heating. Two coal-fired combined heat and power plants named J o u r n a l P r e -p r o o f 4 scientists largely criticized it. In February 2020, more than 20000 citizens signed an online petition urging officials to acknowledge the coal-fired power plants as the main emitters in Almaty (Vlast, 2020) . The inappropriate identification of the inventory is caused not only by the lack of capacity and outdated methodologies but also by the scarcity of data and nontransparent energy statistics (Kerimray et al., 2017) . Since the data on fuel consumption and emissions are not publicly available, producing independent inventories of pollutants is a complicated task.

Source apportionment with chemical analysis of PM particles is needed; however, due to the scarcity of funding for expensive laboratory equipment and the lack of capacity, it has not been conducted so far.

In this study, changes in the air quality before and during the period of COVID-19 lockdown in Almaty were quantified. The possible effects of traffic emissions were discussed.

Daily concentrations of PM 2.5 , NO 2 , SO 2 , CO, and O 3 were compared between the periods before (e.g., during the preceding three weeks or the same days in earlier years) and during the lockdown. Benzene, toluene, ethylbenzene, and o-xylene (BTEX) concentrations were also measured during three days in the middle of the lockdown and compared with the concentrations observed during the same periods of previous years (2015) (2016) (2017) (2018) (2019) . This study aims to assess the impacts of COVID-19 lockdown conditions (traffic-free) on the air quality of Almaty, which is one of the most polluted large cities in the world.

In this study, daily PM 2.5 concentration levels were obtained from the "Airkaz" public air J o u r n a l P r e -p r o o f 5 periods. None of the selected stations was located close to CHP-2 since the station close to CHP-2 did not record full data for March. PM 2.5 concentrations were compared between the lockdown period during March 19 to April 14, 2020, and the same period in previous years. Additionally, the air quality was compared within 2020 between the periods before lockdown (February 21 -March 18) and during lockdown (March 19 -April 14).

Monitoring of benzene, toluene, ethylbenzene, and o-xylene (BTEX) was conducted every spring at 8 AM and 8 PM at six different locations (https://goo.gl/maps/6UPRmjJoYpwEg2D56) during the period from the end of March to the beginning of April from 2015 to 2020. The sampling and analysis methods developed by (Baimatova et al., 2016) and (Ibragimova et al., 2019) were followed. The lockdown BTEX sampling was conducted on three days in the middle of the lockdown.

Daily NO 2 , O 3 , SO 2 , and CO concentration values for the period of March 2 -April 14, 2020, from one station (located in the city center) were obtained from the "Skymax Technologies" company. NO 2 , O 3 , SO 2 , and CO concentration values were not available for the previous years.

The wind speed, wind direction, temperature, relative humidity, precipitation were obtained from the http://rp5.kz website (Weather Schedule, 2020), which collects data from the National Oceanic and Atmospheric Administration (USA) from the station located at 43.15°N, 76.57°E at an elevation of 848 m above sea level.

The cokriging method utilized in the ArcGIS ® Geostatistical Analyst tool (https://desktop.arcgis.com/ru/arcmap/) was used to map PM 2.5 and benzene distributions across Table 1 . The period between February 21 to April 14, 2020 was characterized by a substantial difference (23.3 °C) between the minimum daily temperature (-6 °C) and the maximum daily temperature (17.3 °C). The average temperature before lockdown was 5.5 °C, while it was 8.7 °C during lockdown. Additionally, there were less frequent rains before lockdown period (9 days out of 27) compared to the lockdown period (16 days out of 27). These results show that the meteorological conditions were in favor of air pollution reductions during the lockdown period compared to the preceding days.

On the other hand, the meteorological conditions during the lockdown were almost similar to those of the same periods in the previous years of 2018 and 2019 (Table 1 ; Fig. S1 , Supplementary file). The numbers of rainy days were 15, 16, and 16 days, and the average temperatures were 11.2, 11.6, and 8.7 °C in 2018, 2019, and 2020, respectively. The lockdown period was slightly colder compared to previous years, as there were six days during the lockdown period when the daily average temperature was below 5 °C, while such temperature falls were observed only on one day in 2018 and two days in 2019. These results show that the lockdown period had slightly unfavorable meteorological conditions for air pollution compared to the earlier years. 

J o u r n a l P r e -p r o o f 7 3.1. Impact of the lockdown on the PM 2.5 concentration

The study under analysis (February to April) is a transitional period characterized by rising temperatures and subsequent declining coal combustion by private houses (heating purposes) and

CHPs. For example, the monthly coal combustion at CHP-2 shows significantly varying levels throughout the year (seasonality), with twice lower values in June compared to January and 8- . This record needs further investigation to better understand whether the effect is from policies and measures will last over the years. One possible (alternative) explanation could be the slightly higher temperatures during the period of February 21 -March 18 in 2020 (5.5 °C) compared to previous years (4.6-4.7 °C).

In this study, to exclude the "temperature effect" and "precipitation effect" and to explore only the effect of the lockdown, the concentrations during the same period (March 19 -April 14) of 2018, 2019, and 2020 were compared. The PM 2.5 concentrations (averaged for all stations) during the lockdown period were 38, 40, and 31 μg m -3 in 2018, 2019 and 2020, respectively, indicating a reduction of the PM 2.5 concentration by 18% and 23% in 2020 (during lockdown) compared to the same periods in 2018 and 2019 (before the lockdown year). Fig. 1 shows that the trend of the daily PM 2.5 was fluctuating during the period of March-April in all three years, with no clear downward/upward trend between days or between years.

The PM 2.5 concentration levels varied across the stations during the lockdown from 27 to 38 μg m -3 . The spatial reductions varied between 6% and 34% during the lockdown period compared to the other years (Fig. 2, Fig. 3 and Fig. S2 , Supplementary File), and this might be attributed to the removal of traffic emissions with their varying contributions to the spatial locations. Almaty is located at an altitude between 600 m and 1300 m due to its proximity to the mountains. A previous study by (Kerimray et al., 2020) depicted that the PM 2.5 concentration was correlated with the elevations of the monitoring stations (R 2 =0.64). In this study, PM 2.5 concentration levels during the 2020 lockdown did not have a correlation with the elevation (R 2 =0.23), distance to CHP-2 (R 2 =0.1), or distance to CHP-3 (R 2 =0.22) (Fig. S2, Supplementary File). A previous study by (Kerimray et al., 2020) used data from 11 monitoring stations for J o u r n a l P r e -p r o o f 9 PM 2.5 , while in this study, data from only 7 stations was used (due to the absence of full datasets for March months in 2018-2020). However, the spatial model results shown in Fig. 3 show that the spatial profiles have similarities-lower levels in the south and higher levels in the northbut their variation ranges are significantly different. The weak correlations with distance to CHP-2 (R 2 =0.10) and CHP-3 (R 2 =0.22) could be due to many contributing factors, including the long distance of the sampling sites from CHP-2, several contributing sources of emissions located at different places, complicated topography, and varying wind directions.

Station 16 is the most polluted place and experienced the most significant reduction in the lockdown period from the average of 57 μg m -3 in 2018-2019 to 38 μg m -3 in 2020 (34% decline). Station 16 is at the lowest elevation above sea level (647 m) compared to the locations of the other stations. This station is located at the border of Almaty and administratively belongs to the Almaty region; however, it was included in this study. Station 16 is only 2.5 km away from the coal-fired CHP-3 and is located near major roads. The impacts of traffic and CHP-3 emissions are evident at this location (Kerimray et al., 2020) . The high levels despite the absence of the traffic contribution (38 μg m -3 ) demonstrate that coal combustion (especially close location to CHP-3) is the primary source impacting the station. Station 5, on the other hand, which is located in the city center with high traffic and a lower elevation (793 m), experienced a J o u r n a l P r e -p r o o f 10 1348 m and is close to the mountains, and it is far away from the densely populated areas with high traffic loads. These results confirm that the city has experienced spatial PM 2.5 reductions during the lockdown period.

The number of days exceeding the daily WHO limit (25 μg m -3 ) was 23, 25, and 18 days in 2018, 2019, and 2020, respectively (for the period of March 19 -April 14) (Fig. 1) . The lockdown in 2020 has resulted in a 25% reduction in the number of days compared to 2018 and 2019. However, even with a traffic-free environment, WHO daily limit values in Almaty were still not met on 18 out of 27 days of the lockdown.

The average concentrations of BTEX analytes from 2015 to 2020 are illustrated in Fig. 4 .

The averages for benzene (101 µg m -3 ) and toluene (67 µg m -3 ) were 3 and 2 times higher, while those for ethylbenzene (1.0 µg m -3 ) and o-xylene (1.6 µg m -3 ) were 4 and 2.7 times lower in 2020 than during the same sampling period in 2015-2019 (Table 3 ). In addition, the average concentration of benzene was 15% higher in January 2020 compared to the lockdown period. Table 3 .

The sampling period during the lockdown in April 2020 was characterized by warmer temperatures ranging from 10.2 to 16.2 °C, while the temperature ranged from -6.2 to 14.5 °C on the sampling days in 2015-2019 ( Fig. S1 and Table S1 One of the reasons for the increased concentrations of benzene and toluene during the sampling days in 2020 could be attributed to the no-precipitation conditions. Since there was no traffic activity during the lockdown, the higher levels of benzene and toluene may indicate that their origins are predominantly nontraffic sources, and the declining levels of ethylbenzene and o-xylene by up to 3 fold could be linked to the traffic-free conditions.

BTEX concentrations were inversely proportional to the elevation (above the sea level) of the sampling sites (Fig. 5) which was similar to the case for PM 2.5 concentrations. At the higher elevations (closer to the mountains), the concentrations of BTEX were lower than those at the lower elevations (Fig. 5) , and this could be explained by the location of the coal-fired power plants and households burning coal at the lower elevations.

The BTEX concentrations in 2020 were inversely correlated with the distance to CHP-3, with R 2 =0.87 for benzene and R 2 =0.82 for toluene. The distance-concentration correlations for CHP-2 were weak (R 2 <0.1), which could be due to the large distances of sampling sites from CHP-2 (Fig. S4, Supplementary File) . The correlation of the benzene and toluene concentrations with the distance from CHP-3 was stronger than the correlation with the elevation (Fig. 5) , and this may indicate the dominant contribution of CHP-3 to BTEX pollution in the city. According to the environmental reports of CHP-3 in 2015, coal consumption at CHP-3 was expected to increase in the future due to the rising demand for electricity (Department of Ecology of Almaty region, 2015). There were substantial increases in benzene and toluene during the lockdown period compared to the average during the 2015-2019 years, while some reductions were observed in ethylbenzene and o-xylene concentrations. The variations were significant and ranged between 123% and 227% for benzene and between 36% and 241% for toluene. The highest increases in the concentrations were observed at Station S4, which were 274% (by 119 µg m -3 ) for benzene and 241% (by 86 µg m -3 ) for toluene (Table 3) . Station S4 is located at a low elevation (700 m), close to coal-burning housing developments and at the distances of 12 km from CHP-2 (Fig. S4 , Supplemental materials) and 14 km from CHP-3 (Fig. 5 ). There is also an Almaty bus fleet park located 2.6 km away, and the public bus service was still in operation during the lockdown. The burning of coal at residential houses could have been higher, as people remained in their homes all the time during the 2020 lockdown, and there are plenty of nearby public bathhouses (saunas) that are often heated by burning their garbage or coal. On one of the sampling days, a bonfire was also observed.

Relatively lower concentrations of benzene (76-78 µg m -3 ) and toluene (41-42 µg m -3 ) during the 2020 lockdown were observed at sites S1 (978 m) and S6 (803 m). Sampling site S1 is located in the upper part of Almaty (closer to the mountains), while site S6 is located in a public park at 803 m above sea level. Sampling sites S3 (764 m) and S5 (770 m) are located near significant roads; however, the high levels of BTEX at sites S3 and S5 during the 2020 lockdown indicate the significant contribution from coal combustion.

J o u r n a l P r e -p r o o f 13 that the primary source of BTEX is biomass, biofuel, or coal burning, while T/B > 1 indicates that it mainly originates from vehicle emissions (Liu et al., 2015) .

The varying T/B observed during 2015-2019 indicated the complex nature of BTEX in the ambient air of Almaty (Fig. 7) . In 2015, the obtained T/B ratios were <1 in 18 out of 36 measurements, indicating that sources of BTEX were both vehicle exhaust and coal combustion (Baimatova et al., 2016) . The T/B found in most of the analyzed samples in 2016 (30 from 36 measurements) and 2018 (31 from 35 measurements) were ≥1, suggesting that BTEX mainly originated from transport-related sources. The T/B of the vast majority of collected samples in 2017 (33 from 36 measurements) and 2019 (23 from 35 measurements) were <1, which indicated that BTEX mainly originated from coal burning (Ibragimova et al., 2019) . 

Simple and accurate quantification of BTEX in ambient air by SPME and GC-MS

Determination of aromatic hydrocarbons in urban air of Rome

Use of partial order in environmental pollution studies demonstrated by urban BTEX air pollution in 20 major cities worldwide

Adjustment of the standards for maximum permissible emissions (MPE) in the atmospheric air for the Department of JSC ALES CHP-2". Department of Ecology of Almaty region

Low-cost quantitation of multiple volatile organic compounds in air using solid-phase microextraction 6

Information bulletin on the state of the environment of the Republic of Kazakhstan

Trends and health impacts of major urban air pollutants in Kazakhstan

Analysis of the energy intensity of Kazakhstan: from data compilation to decomposition analysis

Letter to the Public Council of the city of Almaty

Serious BTEX pollution in rural area of the North China Plain during winter season

Ministry of Environmental Protection of Republic of Kazakhstan

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

More than 20 thousand signatures were collected by a petition for the recognition of polluted air as the main problem in Almaty

Spatiotemporal patterns and source implications of aromatic hydrocarbons at six rural sites across China's developed coastal regions

This work was supported by the Postdoctoral Fellowships of AK and NB provided by Al-Farabi Kazakh National University. BTEX sampling and analysis were supported by the Ministry of Education and Science of the Republic of Kazakhstan (Grant No. AP05133158) and "AirVision.kz" Public Fund. The authors are grateful to Skymax Technologies for providing data.

No potential conflict of interest was reported by the authors. 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.