key: cord-0984106-pafschbu
authors: Othman, Murnira; Latif, Mohd Talib; Hamid, Haris Hafizal Abd; Uning, Royston; Khumsaeng, Thipsukon; Phairuang, Worradorn; Daud, Zawawi; Idris, Juferi; Sofwan, Nurzawani Md; Lung, Shih-Chun Candice
title: Spatial–temporal variability and heath impact of particulate matter during a 2019–2020 biomass burning event in Southeast Asia
date: 2022-05-10
journal: Sci Rep
DOI: 10.1038/s41598-022-11409-z
sha: f0982cab7c606293d3004320c21b430350904252
doc_id: 984106
cord_uid: pafschbu

To understand the characteristics of particulate matter (PM) in the Southeast Asia region, the spatial–temporal concentrations of PM(10), PM(2.5) and PM(1) in Malaysia (Putrajaya, Bukit Fraser and Kota Samarahan) and Thailand (Chiang Mai) were determined using the AS-LUNG V.2 Outdoor sensor. The period of measurement was over a year from 2019 to 2020. The highest concentrations of all sizes of PM in Putrajaya, Bukit Fraser and Kota Samarahan were observed in September 2019 while the highest PM(10), PM(2.5) and PM(1) concentrations in Chiang Mai were observed between March and early April 2020 with 24 h average concentrations during haze days in ranges 83.7–216 µg m(−3), 78.3–209 µg m(−3) and 57.2–140 µg m(−3), respectively. The average PM(2.5)/PM(10) ratio during haze days was 0.93 ± 0.05, which was higher than the average for normal days (0.89 ± 0.13) for all sites, indicating higher PM(2.5) concentrations during haze days compared to normal days. An analysis of particle deposition in the human respiratory tract showed a higher total deposition fraction value during haze days than on non-haze days. The result from this study indicated that Malaysia and Thailand are highly affected by biomass burning activity during the dry seasons and the Southwest monsoon.

Spatial-temporal characteristics of PM 10 , PM 2.5 and PM 1 . The 24 h daily average trends of PM 10 , PM 2.5 and PM 1 during the measurement campaign in the year 2019-2020 are shown in Fig. 1 and a descriptive summary is given in Table S1 . Overall, the ranges of PM 10 Detailed monthly concentrations with hourly averages for PM 2.5 are shown in Fig. 2 , and for PM 10 and PM 1 in Figs. S1 and S2, respectively. It can be seen that in September 2019, the concentrations of PM 2.5 started to increase from 10.00 pm until early morning (7.00 am) for Putrajaya, and the same is true for Bukit Fraser which recorded the highest average concentrations starting from 10.00 pm until midnight. Since Bukit Fraser is located near a mountain range, it is susceptible to being influenced by cold air masses during the night. According to Li et al. 29 , mountains area are more influenced by nighttime drainage flow, which resulted in the accumulation of cold air masses in the surface layer, which facilitated accumulation at night. For Kota Samarahan, the highest concentrations were determined in the morning around 8.00 am-10.00 am in September 2019. Chiang Mai had the highest PM 2.5 concentrations in March and April 2020 with the highest concentrations around 9.00 am in March and around 6.00 am (early morning) in April. A stable boundary layer depth and mixing layer height combined with low wind speeds results in high PM concentrations in the morning in Chiang Mai 30, 31 . The stability of mixing layer height, thereby resulting in the suppression of mixing layer evolution, and further increases PM concentrations near the ground surface in the morning 31 . The diurnal monthly variation of PM 2.5 and PM 10 reached up to 160 µg m −3 while for PM 1 the diurnal monthly variations reached up to 100 µg m −3 . Furthermore, the concentrations in June, July and August had the lowest PM 2.5 values indicating the best air quality during these months for Chiang Mai. To examine the relationships between PM 10 , PM 2.5 and PM 1 and the other parameters (temperature, relative humidity and CO 2 ), a Pearson correlation matrix based on the different seasons was produced for monitoring Air mass trajectory clustering. Figure 4 shows the backward trajectories and cluster analysis of the air masses for Putrajaya, Bukit Fraser and Kota Samarahan in September 2019 and for Chiang Mai in April 2020. All of the trajectories were analysed for the air masses with the highest concentrations of PM 10 , PM 2.5 and PM 1 during the monitoring duration, which happened during the Southwest monsoon. Clustering backward trajectory analysis shows that strong winds blew from the south of the Asian region in September 2019 (Southwest monsoon), bringing together smoke pollutants from the coastal area of Sumatra-about 60% of the air mass to Putrajaya and 59% of the air mass to Bukit Fraser. For Kota Samarahan, the dominant air mass (49%) was from the southeast, i.e. from southern Kalimantan where high numbers of hotspots were observed in the south coastal area of Kalimantan. Another 29% of the clustered air mass was identified as being from the east of Kota Samarahan and central Borneo Island. Cluster trajectory air masses determined for Chiang Mai were mainly associated with westerly winds from the coastal area that blew across the high number of biomass burning areas in Southern Myanmar. Other source contributions of air mass were from east of Chiang Mai that blew from southern Laos, also with a high number of hotspots. Another trajectory was mapped to see the contribution of S7) where 62% of the air mass originated from the northwest of Chiang Mai, specifically from southern Myanmar.

Deposition of particle in respiratory tract. Table 2 lists the deposition fraction visualisation in the human lung and deposition fraction values for the head, tracheobronchial (TB), and pulmonary regions. On average, the haze scenario had a higher maximum deposition value for all particle sizes than the non-haze scenario, where clear deposition of particle visualisation could be detected for PM 2.5 when comparing both haze and non-haze scenarios. The individual total deposition fraction values (sum of the head, TB, and pulmonary) was higher for the haze scenario than on non-haze scenario for PM 10 , PM 2.5 , and PM 1 , suggesting higher particle deposition in the human lung on haze days than on non-haze days. The deposition fraction value was higher for the head compared to other regions, with values of 0.9387 (haze) and 0.9171 (non-haze) for PM 10 The increment of PM 2.5 from a normal day to haze day is the highest compared to PM 10 and PM 1 . This can be due to PM 2.5 being the most dominant pollutant produced during biomass burning, also indicated by the high ratio value of PM 2.5 to PM 10 (PM 2.5 /PM 10 ). The higher ratio values for other monitoring site indicate the influence of the transboundary effect and biomass burning during haze days and additional local contributions, which is consistent with previous studies [36] [37] [38] . On normal days, the ratio values of PM 2.5 /PM 10 in central Peninsular Malaysia were recorded as average values of 0.81 ± 0.80 by Othman et al. 6 which is similar to the result of this study, while studies by Rupakheti et al. 39 and Yin et al. 40 had lower PM 2.5 /PM 10 in northwestern and southwestern China which is influenced by the coarse PM fraction from sand and dust storms. Lower PM 1 /PM 2.5 during haze Table 2 . Illustrations of particle deposition in human lungs; and deposition fraction values in the head, tracheobronchial (TB) and pulmonary regions during haze and non-haze scenarios. www.nature.com/scientificreports/ days compared to normal days indicates lower concentrations of PM 1 compared to PM 2.5 during haze days and that PM 1 is not dominantly generated from biomass burning. Thus, other sources of PM 1 can be suggested to be industrial emissions, motor vehicle emissions and coal combustion which are also linked to rapid economic development 41, 42 . Moreover, as reported by Lee et al. 43 , the best indicator for vehicular emissions can be found to be PM 1 for areas with high emissions from motor vehicle compared to PM 2.5 . Higher concentrations of all sizes of PM during the dry season for Chiang Mai compared to other monitoring sites could be due to an increase in particle pollution during this season. A study by Pengchai et al. 44 found high concentrations of PM 10 in the dry season in the Northern part of Thailand. Moreover, PM concentrations were suggested to be high during the dry season and then decreased in the wet season which is related to the wash-out effect, where highest precipitation was observed in August-September 4 . In the case of the Southwest monsoon in Malaysia, normally there are still some rainfall events, especially in the beginning (May and June) of the monsoon, which may affect the PM concentrations. The rainfall pattern in the central region of Peninsular Malaysia was 765 mm year −1 with 35% of the rainfall during the whole the Southwest monsoon, a decreasing trend of monthly rainfall throughout the monsoon 45 and a deficit of rainfall allowing the accumulation of hotspots starting from June 46 . The lowest temperature value during the dry season was also observed by Pongpiachan and Paowa 47 which shows the dry season does not particularly relate to high temperature. Moreover, a stronger correlation between all sizes of PM for Putrajaya and Kota Samarahan compared to Bukit Fraser indicated the effects of urban anthropogenic sources. The relationship between PM 10 and PM 2.5 was stronger in the urban area, suggesting that both particle sizes are influenced by anthropogenic activities, but different sites and locations also play significant roles in the correlation between PM 10 and PM 2.5 48 . A high number of hotspots in coastal areas of Sumatra were suggested to be peatland fires that are usually associated with low rainfall during the Southwest monsoon and which also impact other areas such as Singapore and Southern Thailand. This result is consistent with previous studies 6, 7, 46 . Haze episodes in Southeast Asia, especially Malaysia, are governed by a general wind direction and patterns from the south 49 . It can be said that high numbers of hotspots around Chiang Mai including in Myanmar and Laos contribute to high PM concentrations in this centre of Southeast Asia. As suggested by Pimonsree and Vongruang 38 , major emissions of PM were spotted in March when the contribution of biomass burning was found to be approximately 85% and 89% for PM 10 and PM 2.5 respectively in the centre of Southeast Asia region with strong PM concentration gradients from the biomass burning source within 50 km.

In terms of health effects, it can be suggested that all sizes of PM were highly deposited in the head when inhaling of air. This is due to the facts that 90% of air is inhaled by humans through their noses, which are located in the head 50 . There was also a higher deposition of particles during haze compared to non-haze days especially for coarse particle while fine particles (PM 2.5 and PM 1 ), higher deposition occurred during non-haze but still total deposition fraction value were higher during haze episode. Long et al. 51 observed a significantly higher deposition fraction in the head compared to TB and pulmonary for particle sizes ranging from 0.43 µm to much larger sizes, as well as a significantly lower deposition fraction during haze days compared to non-haze days. It has been suggested that smaller particles enter and accumulate in the innermost reaches of the human respiratory system, where prolonged exposure to small particles such as ultrafine and nanoparticles may have adverse human effects 52 . Additionally, carcinogenic metals bound to PM 2.5 such as arsenic (As), cadmium (Cd), Cobalt (Co), chromium (Cr) and nickel (Ni) were found to be higher in human lung fluid during haze days compared to non-haze days, with Cr having the highest cancer risk value, followed by As 53 . Thus, it is clearly shown that haze episodes have a detrimental effect on human health. It is recommended that outdoor activities be limited during high pollution days.

This study used monitoring data of PM 10 , PM 2.5 and PM 1 concentrations from four monitoring sites in Malaysia and Thailand revealing that all sizes of PM had similar trends in concentrations during the monitoring duration. The highest daily mean concentrations were observed for Kota Samarahan (Malaysia) in September, which was identified as a haze event, with average daily average concentrations of 196 ± 59.5 µg m −3 for PM 10 , 185 ± 57.2 µg m −3 for PM 2.5 and 108 ± 24.1 µg m −3 for PM 1 . For Chiang Mai (Thailand), the high concentration of all size of PM was recorded in end of March and peak concentration in early of April 2020. An increment of PM 2.5 concentration was observed during haze days where the PM 2.5 /PM 10 ratio value was close to 1, indicating that PM 2.5 was significantly contributed to the haze episode. Air mass trajectories coupled with hotspots data clearly show that contributions of all sizes of PM in Putrajaya, Bukit Fraser and Kota Samarahan were from biomass burning, particularly in Sumatra and Kalimantan. The westerly air masses which coincided with high numbers of hotspots related to biomass burning activity in the northern Southeast Asia region was suggested to be the source of all sizes of PM in Chiang Mai during the dry season. Total deposition of particles in the human respiratory tract for outdoor exposure was observed to be higher during haze compared to non-haze, indicating that human health is severely impacted during haze episodes. Further studies investigating the human health impacts of high concentrations of PM need to be undertaken to look at the combined overall impact of PM in the Southeast Asia region.

Study location. Putrajaya, Bukit Fraser and Kota Samarahan in Malaysia and Chiang Mai in Thailand were the monitoring sites used for this study. Both Chiang Mai and Kota Samarahan are usually associated with high concentrations of PM and haze episodes during the dry season. Chiang Mai, located in northern Thailand, is a basin surrounded by mountain ranges with 30% agricultural areas and is also close to biomass burning sources 3 www.nature.com/scientificreports/ sites, where Putrajaya is impacted by anthropogenic activities from the Kuala Lumpur urban environment and also transport of pollutants from Sumatra, Indonesia. Bukit Fraser, which located in the mountainous terrain of Titiwangsa, is in the centre of Peninsular Malaysia. Measurements of PM in Bukit Fraser will provide insights into the impact of seasons and the long-range transport of pollutants to this high altitude site. Thus spatialtemporal analysis of the different sizes of PM could help in determining the association of PM pollution with different seasons in the Southeast Asia region. Additional information and characteristics of each location are provided in Table S2 and Fig. S8 .

PM data and monitoring. The concentration of PM 10 , PM 2.5 and PM 1 with other parameters such as CO 2 , temperature and relative humidity were monitored over one year period in 2019 and 2020 using the AS-LUNG V.2 outdoor sensing device. The measurement campaigns were different for each site where the duration of the measurements was typically for more than one year, aiming to include all seasons and monsoons. A measurement of CO 2 was performed as an indicator for air pollutant level in the study location, where high level of CO 2 could suggest high air level of air pollution in the surrounding of the study location. The AS-LUNG V.2 was built in with sensors for PM (PMS3003, Plantower, Beijing, China), CO 2 (S8, Senseair AB, Sweden), and temperature/ relative humidity (SHT31, SENSIRION, Staefa ZH, Switzerland) that were installed inside a waterproof housing that was powered by a solar panel. As for backup, eight batteries with 10,000 mAH capacity were also connected to the power outlet to provide sufficient power for the sensor in the event that the solar panel received insufficient power. The weight of the sensor was about 5 kg with the waterproof housing weighing about 1.2 kg, and measuring 60 cm × 50 cm × 50 cm. The sensors were programmed to measure all parameters for every 15 s, and the data was saved to an SD card. AS-LUNG V.2 sensor is small device with no noise; compact outer case and easily set-up for outdoor measurement; and evaluated against research-grade instruments with coefficient of determination (R 2 ) almost 0.895-0.998 that indicated this sensor are qualified for research 54 .

At each monitoring site, the AS-LUNG V.2 sensors were installed at high level locations. In Chiang Mai, the sensor was installed on the rooftop of four-storey building, while in Putrajaya, the sensor was installed on top of an air quality cabin about 5 m above the ground. In Kota Samarahan, the sensor was installed on the second floor of a university building which far away from the parking lot and human interference while in Bukit Fraser, the sensor was installed on a light pole about 2 m from the ground on a site that is a campus site and not open to the public. The seasons in Chiang Mai are described as the wet season (June to September), Transition 1 (October to November), dry season (December to March) and Transition 2 (April to May) 44, 55 . For seasons in Malaysia, Inter monsoon 1 (March to May), Southwest monsoon (end of May to September), Inter monsoon 2 (October to in the middle of November) and Northeast monsoon (November to March) were applied for this study.

Data quality assurance and quality control (QA/QC) were conducted with the data from the Continuous Air Quality Monitoring Station for PM 10 and PM 2.5 for hourly data which indicated that the sensor data is about ± 30% different and the R 2 value > 0.6. Due to the lack of data for PM 1 measured by the Continuous Air Quality Monitoring Station, the data accuracy for PM 1 was derived from Lung et al. 56 who conducted a side-by-side comparison between the AS-LUNG V.2 sensor and the GRIMM instrument to determine the R 2 . The R 2 value for PM 1 appeared to be high, ranging between 0.931 and 0.996. Statistical analysis for the obtained data is performed using R software with the Openair package.

Backward air mass trajectory analysis. The Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) with analysis of cluster backward trajectories was used to investigate air mass trajectories for high concentrations of PM at each of the measurement sites from 2019 to 2020. The metrological data input was obtained from the Global Data Assimilation System (GDAS) with a height of 500 m, which was then assessed for the daily run. Following the completion of daily runs, standard clustering was analysed for 36 h to cluster with an input of four number of clusters. The airmass trajectory was then overlaid with Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data on the Aqua and Terra fire hotspot (https:// firms. modaps. eosdis. nasa. gov/ map/#d: 24hrs;@0. 0,0. 0,3z) with confidence values ranging from 0 to 100%.

Human airway particle dosimetry. The deposition fraction of all particle sizes in the human respiratory tract was modelled using Multiple-path Particle Dosimetry software (MPPD, v3.04), which was used to gain a clear understanding on particle deposition in the human respiratory system. Basically, this MPPD was based on single-path and multiple-path methods for tracking air flow and calculating aerosol deposition in the human lung where the single-path method calculates deposition along a typical path per airway generation, whereas the multiple-path method calculates particle deposition along all airways of the lung and provides lobar-specific and airway-specific information 57 . This software was developed by Applied Research Associates, Inc., where it is usually applied to calculate the deposition and clearance of monodisperse and polydisperse aerosols in the respiratory tract. In this study, the input data used to run the software was restricted to human morphology using the Yeh/Schum Symmetric Model, with functional residual capacity (FRC) and Upper Respiratory Tract (URT) volume set to default values of 3300 mL and 50 mL, respectively. For particle properties' input data, the sizes of PM 10, PM 2.5 , and PM 1 were inputted as 10, 25, and 1, respectively, for the count median diameters (CMD). The steps and input data selection for this software were performed as following Manojkumar et al. 58 . In this study, the PM 10 , PM 2.5 , and PM 1 average concentrations across all sites were used as input data for aerosol concentrations to represent the outdoor exposure of the individual adult during both haze and non-haze (normal day) scenarios, without regard for geographical and meteorological conditions.

Emissions of pollutant gases, fine particulate matters and their significant tracers from biomass burning in an open-system combustion chamber

Projection of the near-future PM2.5 in Northern Peninsular Southeast Asia under RCP8

Influence of zero-burning policy and climate phenomena on ambient PM2.5 patterns and PAHs inhalation cancer risk during episodes of smoke haze in Northern Thailand

Atmospheric pollutans in Chiang Mai (Thailand) over a five-year period (2005-2009), their possible sources and relation to air mass movement

The haze problem in Northern Thailand and policies to combat it: A review

Spatial distribution of fine and coarse particulate matter during a southwest monsoon in Peninsular Malaysia

Regional transport, source apportionment and health impact of PM10 bound polycyclic aromatic hydrocarbons in Singapore's atmosphere

Numerical assessment of PM25 and O3 air quality in Continental Southeast Asia: Impacts of potential future climate change

El Niño driven haze over the Southern Malaysian Peninsula and Borneo

Biomass burning spatiotemporal variations over South and Southeast Asia

Southeast Asian smoke haze: Fractionation of particulate-bound elements and associated health risk

Haze, public health and mitigation measures in China: A review of the current evidence for further policy response

Peat-fire-related air pollution in Central Kalimantan

Impacts of biomass burning in peninsular Southeast Asia on PM2.5 concentration and ozone formation in Southern China During Springtime-A case study

Exploring the effects of haze pollution on airborne fungal composition in a cold megacity in Northeast China

Effects of haze pollution on microbial community changes and correlation with chemical components in atmospheric particulate matter

Haze smoke impacts survival and development of butterflies

Effect of heavy haze and aerosol pollution on rice and wheat productions in China

Early-life exposure to three size-fractionated ultrafine and fine atmospheric particulates in Beijing exacerbates asthma development in mature mice

Is PM1 similar to PM2.5? A new insight into the association of PM1 and PM2.5 with children's lung function

Impact of air pollution on the burden of chronic respiratory diseases in China: Time for urgent action

Polycyclic aromatic hydrocarbons are associated with increased risk of chronic obstructive pulmonary disease during haze events in China

The interplay of haze characteristics on mortality in the Pearl River Delta of China

Health impacts of the Southeast Asian haze problem-A time-stratified case crossover study of the relationship between ambient air pollution and sudden cardiac deaths in Singapore

Diagnosis of lung cancer increases during the annual southeast Asian haze periods

Pollutional haze as a potential cause of lung cancer

Association of the haze and diabetes in China

demonstration of a new framework for informing fire management strategies to reduce downwind smoke exposure

The impacts of the atmospheric boundary layer on regional haze in North China

The infleunce of meteorology on ambient PM2.5 and PM10 concentration in Chiang Mai

Mixing LayerHeight retrievals from MiniMPLMeasurements in the Chiang MaiValley: Implications for ParticulateMatter Pollution

Comparison of PM2.5 exposure in hazy and non-hazy days in Nanjing

Acute and cumulative effects of haze fine particles on mortality and the seasonal characteristics in

Influence of Southeast Asian Haze episodes on high PM10 concentrations across Brunei Darussalam

Air pollution impacts from COVID-19 pandemic control strategies in Malaysia

Chemical characteristics of PM2.5 during a 2016 winter haze episode in Shijiazhuang

The variation of chemical characteristics of PM2.5 and PM10 and formation causes during two haze pollution events in urban Beijing

Impact of biomass burning and its control on particulate matter over a city in mainland Southeast Asia during a smog episode

Spatio-temporal characteristics of air pollutants over Xinjiang, northwestern China

Gaseous and particulate pollutants in Lhasa, Tibet during 2013-2017: Spatial variability, temporal variations and implications

Estimating spatiotemporal distribution of PM1 concentrations in China with satellite remote sensing, meteorology, and land use information

Seasonal concentration distribution of PM1.0 and PM2.5 and a risk assessment of bound trace metals in Harbin, China: Effect of the species distribution of heavy metals and heat supply

and PM2.5 characteristics in the roadside environment of Hong Kong

Seasonal variation, risk assessment and source estimation of PM10 and PM10-bound PAHs in the ambient air of Chaing Mai and Lamphun

Rainfall characteristics and regionalization in Peninsular Malaysia based on high resolution gridded data set

Riau province Sumatera smoke haze event on regional air pollution

Hospital out-and-in-patients as functions of trace gaseous species and other Meteorological parameters in

Concentrations, correlations and chemical species of PM2.5/PM10 based on published data in China: Potential implications for the revised particulate standard

Determination and analysis of trace metals and surfactant in air particulate matter during biomass burning haze episode in Malaysia

Performance of a new model of the nasal filter when challenged against PM1 aerosols

Characteristics, emission sources and health risk assessment of trace elements in size-segregated aerosols during haze and non-haze periods at Ningbo, China

Characterization and estimation of human airway deposition of sizeresolved particulate-bound trace elements during a recent haze episode in Southeast Asia

Risk assessment of bioaccessible trace elements in smoke haze aerosols versus urban aerosols using simulated lung fluids

A versatile low-cost sensing device for assessing PM2.5 spatiotemporal variation and quantifying source contribution

Influence of biomass burning on the levels of atmospheric polycyclic aromatic hydrocarbons and their nitro derivatives in Chiang Mai

Concurrent assessment of personal, indoor, and outdoor PM2.5 and PM1 levels and source contributions using novel low-cost sensing devices

Applied Research Associates, I. Multiple-path Particle Model, MPPD V3

Application of multiple-path particle dosimetry model for quantifying age specified deposition of particulate matter in human airway

This study is supported by Universiti Kebangsaan Malaysia for research grant GGPM-2020-031. Highly appreciation is given to IRDR ICoE-Taipei and International Science Council Regional Office for Asia and the Pacific 

The authors declare no competing interests.

The online version contains supplementary material available at https:// doi. org/ 10. 1038/ s41598-022-11409-z.Correspondence and requests for materials should be addressed to M.O.Reprints and permissions information is available at www.nature.com/reprints.Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.