[ Technical Information ]
Asian Journal of Atmospheric Environment - Vol. 14, No. 1, pp.47-61
ISSN: 1976-6912 (Print) 2287-1160 (Online)
Print publication date 31 Mar 2020
Received 22 Jul 2019 Revised 02 Oct 2019 Accepted 06 Jan 2020

Trend Characteristics of Atmospheric Particulate Matters in Major Urban Areas of Bangladesh

Md Masud Rana1), * ; Munjurul Hannan Khan1), 2)
1)Clean Air and Sustainable Environment Project, Department of Environment, E-16, Agargaon, Dhaka - 1207
2)Ministry of Environment, Forest and Climate Change, Bangladesh Secretariat, Romna, Dhaka - 1207

Correspondence to: * Tel: +88-01760900787 E-mail: ranamasud2002@yahoo.com

This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

The urban areas of Bangladesh suffer from severe air quality problem especially in dry season (November-April) when the PM concentrations frequently rise to 7-8 times of the WHO guideline value. The Clean Air and Sustainable Environment (CASE) Project of the Department of Environment has deployed countrywide continuous air monitoring systems to regularly monitor the air quality of the urban areas of Bangladesh. In this paper hourly concentrations of PM10 and PM2.5 captured using β-attenuation method from 2013 to April 2018 in six important cities located in different regions of the country were exhaustively analyzed. Statistical analyses, diurnal and seasonal trends, and polar plots of PM concentrations were examined. Long range sources were spotted by Concentration Weighted Trajectory (CWT) method, where the trajectories were calculated using HYSPLIT-4 model. The analyses identified cities in the middle of the country (Dhaka, Narayanganj, Gazipur) as the most polluted ones while the city to the northeast region (Sylhet) was the least polluted. Average PM10 concentrations at Dhaka, Chattogram, Narayanganj, Gazipur, Sylhet and Barisal stations in dry seasons (November-April) were found 238.7±155.4, 190.7±108.5, 303.6±161.4, 227.3±142.7, 151.7±105.0 and 170.7±108.4 μg m-3 respectively whereas those in wet seasons (May-October) were 75.0±51.6, 55.5±40.8, 102.4±84.4, 60.6±48.5, 52.7±38.3, and 54.4±41.6 μg m-3 respectively. Correlative study of diurnal variations in PM concentrations and meteorological parameters revealed that the congenial meteorology aided in developing higher concentrations of both PM10 and PM2.5 during nighttime. Sources located to the northwestern districts (Naogao, Rangpur, Bogura) were traced by the CWT method contributing to the air pollution in other regions of the country. Outside the boundary, sources in Nepal, and Delhi-NCR and Uttar Pradesh regions of India could have contributed to fine particles at the middle of the country.

Keywords:

Particulate matter, Bangladesh, Conditional bivariate probability function, Diurnal variation, Concentration weighted trajectory

1. INTRODUCTION

The south Asian countries are now the major hotspots in the world for increased level of particulate matter (PM) in air (Pant et al., 2018; Khatum et al., 2017; Rana et al., 2016a; Dey et al., 2012; Gurung and Bell, 2012). Burning of biomass and fossil fuel (Tiwari et al., 2014), dusts and industrial emissions fill the atmosphere with great amount of PM (Sharma et al., 2016; Begum et al., 2010; Sharma et al., 2010; Chowdhury et al., 2007), and especially during dry season congenial meteorology assists in developing PM concentrations in the atmosphere (Schnell et al., 2018). The heightened level of PM in air is not only increasing health burden in this region but also decreasing crop production and playing important roles in climate change (Pommier et al., 2018; Tiwari et al., 2014; Ramanathan and Carmichael, 2008; Venkataraman, 2005). Specially, the potential of black carbon in changing climate is now proven (Kopacz et al., 2011; Shindell et al., 2011; Yasunari et al., 2010; Ming et al., 2008). The State of Global Air/2019 published jointly by the United States’ Health Effect Institute and Global Burden of Disease project of the Institute for Health Matrix and Evaluation reveals that air pollution could be blamed for about 4.9 million early deaths worldwide in 2017 (State of Global Air, 2019); China and south Asian region shared about half of this mortality. Lelieveld et al. (2015) also accounted outdoor air pollution for the early deaths of about 92,000 people in Bangladesh in 2010, 55% of which was attributed to the household energy use like cooking, heating, etc.

The country of Bangladesh has been experiencing continuous growth in urbanization as a result of ongoing industrial expansion. Urban population estimated at the last national consensus in 2011 was found about 28% of the total population, about 3.5% high from that found in the previous consensus in 2001 (BBS, 2011). The economic activities of the country have accordingly been changing from agriculture sector (contribution to GDP falls by 1.9% in last four years) to industrial sector (contribution to the GDP increases by 3.29% in last four years). Source apportionment of the air pollution in the cities is also changing gradually (Begum et al., 2013, 2011, 2008). Vehicle sector contributed much (~50%) of the fine particles in Dhaka city in the 90’s; but its share fell sharply as a result of the government’s decisions of importing unleaded petroleum from 1999, complete expulsion of 2-stroke baby taxis in 2003, and introducing cleaner fuel compressed natural gas (CNG) in the vehicle sector. The latest apportionment result shown by CASE (2014) found brick manufacturing sector now contributing more than 50% of the fine particles in Dhaka, the capital city of the country. Not only Dhaka other divisional cities also suffer greatly from the emissions from brick kilns (CASE, 2014). Other than brick kilns, wood burning was found as a big source of fine particulates in Rajshahi, Khulna and Chattogram cities, and sea salt/zinc sources with contribution of about 30.5% were found another strong source of fine particulates in Chattogram (CASE, 2014).

Although the entire country plunges into devastating air pollution during dry season, rare works have been done on monitoring and characterizing the air quality outside the Dhaka city. This study undertook massive initiatives of continuously capturing particulate matter (PM) concentrations in the air of 06 important cities located in different regions (center, south, southeast and northeast) of the country from 2013 to April 2018. The main objectives of this study were to characterize the temporal and spatial variations in atmospheric particulate matter concentrations in the country, and also to predict source directions responsible for high PM pollution. Seasonal, diurnal and directional trends and statistical parameters of PM10 and PM2.5 concentrations were analyzed. Possible long range source-regions for the high PM concentrations in the cities were also investigated. This sort of countrywide observations and analyses of air quality were not done before in Bangladesh and hence carry a great deal of importance for the air quality management in the country.

2. MATERIALS AND METHODS

2. 1 Monitoring Sites

Continuous PM monitoring in air has been done in six important cities located at different regions of the country - Dhaka, Narayanganj and Gazipur in the middle, Barisal to the south, Chattogram to the southeast, and Sylhet to the northeast part of the country (Fig. 1). Table 1 provides information of the sampling sites and possible sources of air pollution in the respective cities.

Locations of air monitoring sites with population sizes of the respective town/city.

Descriptions of the cities, sampling sites, and possible sources of air pollution.

2. 2 PM Monitoring, Quality Control and Analyses

Continuous measurement of PM concentrations at the sites was done using beta attenuation method by the PM monitors (Model BAM-1020) made by the Met One Instrument Inc, USA. In the monitor, Carbon-14 was used as the beta radiation source and Geiger Mueller (GM) counter as the detector for the beta particle counts. Individual monitors were used for measuring PM10 and PM2.5 concentrations continuously. The PM monitors were calibrated every week using reference gauges with precisely-defined surface density. Relative humidity of the sample air was controlled to less than 35% using a Smart Inlet Heater system. Not only PM concentrations meteorology parameters (wind speed, wind direction, temperature, relative humidity, etc) were also recorded. Hourly PM concentration data was scrutinized to remove the invalid ones which were identified in several ways, (i) the ratio of PM2.5 to PM10 must not be greater than 1.0, (ii) the ratio should not be unusually high or low compared to its neighboring ones; such ratio indicates error in instrument signal or unusual source encroachment, (iii) the data should not be fixed on a certain value for more than 2 consecutive hours. Year-wise rate of valid data at the stations are provided in Table 2.

Year-wise rate (in percent) of valid data generated at the stations.

2. 3 Conditional Bivariate Probability Function (CBPF)

Conditional Probability Function (CPF) shows the probability of a receptor to acquire from a particular wind direction sector a species greater than some specified concentration value (usually 90 percentile). CPF is primarily used to find the directions having high probability of associating high concentrations of a pollutant at a site. However, the CPF can be usefully applied to the bivariate polar plots. In this case, the CPF along with a third variable (most of the cases wind speed, or any other meteorology variable) is plotted with the bivariate polar plots. The CPF in bivariate polar system which is termed as the conditional bivariate probability function (CBPF) can be defined as equation 1 (Uria-Tellaetxe and Carslaw, 2014):

 ${CBPF}_{∆\theta ,∆u}=\frac{{m}_{∆\theta ,∆u}\left(C\ge x\right)}{{n}_{∆\theta ,∆u}}$ (Eq. 1)

Where, mΔθu is the number of samples in the wind sector Δθ with wind speed interval Δu having concentration C greater than a threshold value x, nΔθu is the total number of samples in that wind direction-speed interval. CBPF can also be calculated for an interval of concentration rather than only values greater than some threshold. In that case, CBPF is equated as equation 2,

 ${CBPF}_{∆\theta ,∆u}=\frac{{m}_{∆\theta ,∆u}\left(y\ge C\ge x\right)}{{n}_{∆\theta ,∆u}}$ (Eq. 2)

Where, mΔθu is the number of samples in the wind sector Δθ with wind speed interval Δu having concentration C between the value of x and y, nΔθu is the total number of samples in that wind direction-speed interval (Uria-Tellaetxe and Carslaw, 2014).

2. 4 Meteorology of the Cities

The meteorology of Bangladesh varies greatly in different seasons. Based on wind and rain pattern the seasons could be primarily divided into two, (a) dry season (November to April) and (b) wet season (May to October). For air quality study, Rana et al. (2016a) divides the dry season as (i) winter (November-January), and (ii) summer (February to April). Based on the data from the Department of Meteorology, Rana et al. (2016a) showed that the atmosphere as a whole in the country remains very dry during the month of February to April and very wet from June to August. The wintertime is characterized with low temperature, wind speed and mixing height while the summertime is typified with high temperature, dryness and mixing height, and clear sky (Rana et al., 2016a). The meteorology also differs regionally; the northeast region of the country experiences more rainfall (annually >4000 mm) while the northwest region gets the less (annually <2300 mm) (Shahid, 2010). The climate of the northwest region is extreme (colder in winter and hotter in summer and wet season) while that of the northeast, south and southeast regions is moderate.

Four stations (Gazipur, Chattogram, Sylhel and Barisal) out of the six under this study have onsite meteorology monitors; the wind roses of these stations are shown in Fig. 2. Meteorological parameters were measured every minute, and an hourly average of a parameter was deducted from an arithmetic mean of the values in that hour. Measuring height of the parameters at the stations was about 15 meter.

Wind-roses at different stations in wet and dry season.

The winter and wet seasonal winds are opposite; the wind in general flows from the west and northwest directions during wintertime and from the south and southeast directions during summertime and wet season (Rana et al., 2016a). This study found most of the regions except the northeast (Sylhet) and southeast (Chattogram) parts of the country followed this wind pattern (Fig. 2). Sylhet experienced major wind from the northeast direction in dry season and from the east, northeast, and southeast directions in wet season (Fig. 2). Dry seasonal wind in Chattogram was observed from the north and northeast directions, and the wet seasonal wind from the south and southeast directions.

3. RESULTS AND DISCUSSION

3. 1 Overview of PM Concentrations

PM concentrations all over the country was found to follow a unique pattern of seasonal variations - high concentrations in dry season (November-April) and low in wet season (May-October) (Fig. 3). The contributions of PM2.5 to PM10 were comparatively higher in wintertime (December-January) and lower in summertime (February-April). The contributions in wintertime in Dhaka, Chattogram, Narayanganj, Gazipur, Sylhet and Barisal cities were respectively 67.0, 64.0, 60.0, 70.0, 58.0, and 76.0% whereas the contributions in summertime were respectively 56.0, 53.0, 46.0, 58.0, 53.0 and 64.0%. PM2.5 contributions to PM10 in wet season in the cities were 52.0, 55.0, 37.0, 58.0, 42.0 and 72.0% respectively. The driving forces for higher PM2.5 contribution to PM10 in winter season could be the meteorology, open burning and the distant sources. Scare rainfall, low wind speed and low mixing height (Rana et al., 2016a) in winter season aid in build up particulate concentrations in the atmosphere of this region. Compared to coarse particles, fine particles have high residence time in the atmosphere, and are expected to gain much over the coarse particles in dry atmosphere. Lots of seasonal sources especially brick manufacturing kilns, open stubble and garbage burning, etc. become active in this season and contribute greatly to particulate pollution. The meteorology of winter season in this region also assists in long range transportation of fine particulate matters. Azkar et al. (2012) investigated influences of long range transport and brick kiln emissions on air quality over Bangladesh by using air quality simulations. Fine particles (PM2.5) have higher health impact potential, and are related with complex respiratory and cardiovascular problems and may even create cancer in human being (Feng et al., 2016). Thus, the air pollution during winter season in Bangladesh could be considered more damaging; extra care and management is needed to control the air pollution in this season. In contrast to the winter season, the wet season experiences fresher air for the frequent rainfall, high mixing height, etc.

Time series graph of 24-h average PM10 and PM2.5 Concentrations at different cities of Bangladesh; horizontal lines are national limit values of PM10 (upper) and PM2.5 concentrations (lower).

The cities located in the middle of the country (Dhaka, Narayanganj, Gazipur) were found more polluted compared to other regions of the country, and that Narayanganj was found as the most polluted city in the country. Dhaka and Gazipur experienced almost the similar pollution level, although the Dhaka station is located at congested urban site and is influenced by heavy traffic, and the Gazipur site is sub-urban background with rare nearby sources. The Narayanganj station although received the highest level of both PM10 and PM2.5 concentrations (Table 3), it had comparatively lower PM2.5 to PM10 ratio-indicating massive sources of coarse particles near the station. PM concentrations captured at Gazipur site especially during dry season were too high (Table 3) for a sub-urban background station and comparatively higher PM2.5 to PM10 ratio indicates effects of industrial activities and open cooking. People of Gazipur are used to burn logs, straws and leaves for cooking, and both Gazipur and Narayanganj cities are highly industrialized. Number of battery manufacturing industries, steel mills, dying factories, garment manufacturing units, pharmaceuticals, cement industries, etc. operate within the boundary of those two districts. PM2.5 concentrations in Barisal city were found greater than that of Chattogram city which is bigger, busier, and more industrialized than Barisal city. Contributions of PM2.5 to PM10 in wet season were found exceptionally high (0.72) in Barisal city. Barisal has a big river port and the diesel driven water vessels could be a possible source of higher PM pollution in the city. Among the cities, Sylhet was found the least polluted in respect of both PM fractions and fine particle contributions to PM.

Overview of the daily PM concentrations in different seasons in the cities of Bangladesh. Daily concentrations were calculated with minimum 80% data availability.

The statistics provided in Table 3 were calculated from the daily averaged PM10 and PM2.5 concentrations captured at the stations from 2013 to April 2018-each data thus representing individual day. Most of the days (~90%) of the winter season, especially during the month of December and January, the concentrations of both fractions of PM in the cities (except Sylhet) were greater than the national standards; in case of Sylhet, it is about 75% days of December and January when PM levels were noncompliant in respect to the Bangladesh National Standards (150 μg m-3 for daily PM10 and 65 μg m-3 for daily PM2.5 concentrations). Compared to winter seasons, the summer seasons were found less polluted (Table 3); however, coarse particles were found to dominate over the fine particles in summer in contrast to the winter when fine particles dominated over the coarse particles (Table 3). On an average, the concentrations of PM10 and PM2.5 in summer in Dhaka were respectively 27.7 and 41.0% lower than those in winter season; in other cities the reductions in PM10 and PM2.5 concentrations from winter to summer were respectively as follows, Narayanganj 30.7 and 41.7%, Gazipur 14.5 and 32.2%, Chattogram 22.7 and 36.3%, Sylhet 12.5 and 20.5% and Barisal 33.6 and 43.4%. In all of the cities, decreases in PM2.5 concentrations from winter to summer were greater than the reductions in PM10 concentrations from winter to summer.

The wet seasonal PM10 and PM2.5 concentrations in the cities were very compliant with the NAAQS but still violated the WHO guideline values respectively in Dhaka by 75 & 75% days, Narayanganj by 75 & 50% days, Gazipur by 50 & 50% days, Chattogram by 50 & 50% days, Sylhet by 50 & 25% days, and Barisal by 50 & 50% days of the wet season (Table 3).

3. 2 Directional Influences on the PM Concentrations at the Stations

CPF polar plots of PM2.5 concentrations at low (5-30) and high (70-98) percentile intervals were examined at the stations particularly which had on-site meteorology data.

The CPF polar plots of PM2.5 concentrations (Fig. 4) show that all of the stations were experiencing high level of PM2.5 concentrations at calm weather during the winter season (Fig. 4a, g, m, s)-indicating most of the PM2.5 being generated from the local sources. The Gazipur station during all of the seasons was affected severely with high level of PM2.5 from the sources lying to the northwest and southwest directions (Fig. 4a, c, e); sources to the southeast directions were also crucial for high PM2.5 concentrations especially during summer season (Fig. 4c). High PM2.5 concentrations during all of the seasons at the Chattogram station were found associated with the eastern and northern winds (Fig. 4g, i, k); the impacts on this station from the south, northwest and southwest directions were comparatively weak (Fig. 4h, j, l). Barisal station was experiencing high PM2.5 pollution from the west in winter season (Fig. 4m) and from the north during summer and wet seasons (Fig. 4o, q). The southern and southeastern sources were not contributing highly in any season in Barisal station (Fig. 4p, r). Sylhet station also observed high level of PM2.5 from the northwest direction during winter season, and from the west during summer and wet seasons (Fig. 4u, w).

CBPFs of PM2.5 concentrations in two percentile ranges (70-98 and 5-30) shown for different seasons in the cities. GZ=Gazipur, CTG=Chattogram, BAR=Barisal and SYL=Sylhet. The color bar shows the CPF probability of PM2.5 for the concentration range given in the parentheses below the figure.

3. 3 Diurnal Variations of PM Concentrations

The diurnal variations of PM concentrations (both fractions) in Dhaka, Narayanganj, Gazipur and Barisal were mostly the same in winter and summer seasons (Fig. 5)-the trends were bimodal in pattern, having peaks at 9:00 am and 9:00 pm. After the concentration peak at 9:00 am, PM concentrations (both fractions) in all of those cities steeply went down throughout the day; the concentrations started to rise from 6:00 pm till the peak at 9:00 pm. The concentrations plummeted throughout the night and started to rise again from 6:00 am next morning (Fig. 5). The wet seasonal diurnal trends of PM in those cities also experienced two pikes at 9:00 am and 9:00 pm, but the daytime variations of PM in this season were little different from the winter and summer seasons - tiny upward trend in the afternoon of wet season was noticed, and that the daytime trends of PM concentrations were little flat compared to those in winter and summer (Fig. 5). Diurnal variations of PM in the city of Chattogram and Sylhet were found little different from other cities discussed above, especially the PM peak in the morning in these two cities were less pronounced.

Diurnal trends in PM10 and PM2.5 concentrations in different seasons in the cities.

Although the cities of Chattogram, Sylhet and Barisal are located about 300 km away from the middle part of the country (Dhaka, Narayanganj, Gazipur) and have different type of responsible sources of PM, the diurnal variations of the PM concentrations in all of the cities especially after 6:00 pm were the same (Fig. 5). This is perhaps for the similar trend characteristics in the meteorology parameters in the cities (Fig. 6). The country of Bangladesh (except Sylhet and Chottogram regions) has a very flat terrain; climatic changes throughout the country are almost similar. Correlative trends of normalized PM2.5 concentrations and meteorological parameters are shown in Fig. 6.

Relationships among normalized level of diurnal variations of PM2.5 and meteorology parameters; ws=wind speed, T=temperature, RH=relative humidity, SR=solar radiation. gz=Gazipur, ctg=Chattogram, syl=Sylhet, bar=Barisal.

Meteorology has profound impact on the air quality in a region (Schnell et al., 2018; Tiwari et al., 2014). Collective impacts of the local emissions, meteorology parameters, and boundary layer dynamics determine the trends in the PM concentrations in a region (Dumka et al., 2015, 2013). Stull (1988) suggests that the fumigation effect and the evolution of the boundary layer height (BLH) just after the sunrise favor building up PM concentrations, which when combined with the rush hour emissions from especially motor vehicles result in the morning peaks. As Chattogram and Sylhet stations were far from the busy roads, and any other big local sources, the morning peaks of PM concentrations at those stations could not develop much (Fig. 5). However, nighttime hike in PM concentrations at 9:00 pm was very common in all of the seasons in the cities (Fig. 5).

Although the source activities (vehicles, road dust, constructions, industries, etc.) in the cities run in large scale at daytime, the PM concentrations were found to increase at nighttime (Figs. 5 & 6) and the average nighttime PM concentrations in all of the cities were found greater than the average daytime concentrations (Table 4). Fig. 6 reveals that the daytime meteorology of the cities was characterized with comparatively greater solar radiation, wind speed and temperature, and lower relative humidity. Nighttime PM10 and PM2.5 concentrations at Dhaka stations in winter season were on average 33.5 and 26.4% greater than the daytime concentrations respectively. Similarly, the nighttime PM10 and PM2.5 concentrations were greater than the daytime concentrations in other cities in winter respectively by 26.2 and 22.0% in Narayanganj, 63.8 and 54.6% in Gazipur, 45.6 and 45.5% in Chattogram, 85.4 and 98.7% in Sylhet, and 38.2 and 35.2% in Barisal. On the other hand, the increase in the nighttime PM10 and PM2.5 concentrations compared to daytime concentrations in summertime in the cities were 23.2 and 16.7% in Dhaka, 22.8 and 23.7% in Narayanganj, 46.8 and 38.8% in Gazipur, 13.6 and 21.5% in Chattogram, 52.3 and 57.6% in Sylhet, and 28.6 and 24.2% in Barisal. Thus, the increase in nighttime PM10 concentrations compared to that in PM2.5 concentrations during both winter and summer seasons were found higher in Dhaka, Gazipur and Barisal. Narayanganj experienced mostly similar increases in both fractions of PM concentrations at nighttime; however, Chattogram and Sylhel experienced higher increase in PM2.5 concentrations compared to that of PM10 concentrations at nighttime during both winter and summer seasons. Coarse particles (PM2.5-10) have comparatively shorter residence time in the atmosphere compared to fine particles (Seinfeld and Pandis, 1998); the fine particles are usually expected to dominate at nighttime when major source activities are ceased off. The increase in PM10 concentrations compared to PM2.5 concentrations at nighttime indicates presence of some sort of source activities at nighttime in Dhaka, Gazipur and Barisal cities. Chattogram and Sylhet stations are located far from major local sources and thereby experienced relatively higher PM2.5 concentrations at nighttime.

Average PM concentrations (μg m-3) at daytime and nighttime in different seasons in the cities.

3. 4 Long Range PM Sources

In dry weather PM gets longer residence in the atmosphere, especially the fine particles may have weeks of residence time and travel hundreds to thousands of kilometers before being removed from the atmosphere by dry deposition (Seinfeld and Pandis, 1998). The entire south Asian region during the dry season suffers from increased level of PM in air. Industries, soil dust open cooking and stubble burning, etc. altogether emit tons of PM into the atmosphere of south Asian countries. The PM so emitted during dry season disperses around thousands of kilometer over the south Asian countries (Begum et al., 2011; Lawrence and Lelieveld, 2010). Several researches earlier have shown incursions of trans-boundary PM into Dhaka, and the source locations identified by those studies are Nepal and the Indian regions lying on the Himalayan Valley and on the Indo-Gangetic Plain (Rana et al., 2016b; Begum et al., 2010). In all of the previous works, trans-boundary pollution was investigated over Dhaka and its surroundings; however, in this paper, long range PM pollutions (both within and beyond the national boundary) were investigated around Dhaka, Chattogram, Sylhet and Barisal cities.

Ninety six (96)-hour backward isobaric trajectories with every 3-hours interval during the period of 15 December 2015 to 15 January 2016 on Gazipur, Chattogram, Sylhet and Barisal stations were calculated using Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT-4) model (Draxler and Rolph, 2003) on the platform of the analytical software “R” which was used to analyze the trajectories and concentration data. Global Reanalysis Meteorology Data downloaded from the archive of National Oceanic and Atmospheric Administration (NOAA) was used as input to the model. As those trajectories were analyzed with their association with station concentration data, starting points of the trajectories were kept at 10.0 m high from the ground. PM concentrations in the cities usually remain at the highest level during this time period (Fig. 3). Hourly PM2.5 concentrations at the stations were associated with the trajectories arrived at the respective hours at the stations. Concentration weighted trajectory (CWT) method (seibert et al., 1994) was applied to identify possible long range regions responsible for contributing to PM pollution at the stations. CWT is calculated from the residence time of trajectories over a grid cell and the respective PM2.5 concentrations the trajectories were associated with. In this method, for each cell of a domain, mean CWT or logarithmic mean concentration of a pollutant species was calculated according to the Equation 3 as follows:

 $\mathrm{ln}\left({\overline{C}}_{ij}\right)=\frac{1}{{\sum }_{k=1}^{n}{\tau }_{ijk}}{\sum }_{k=1}^{n}\mathrm{ln}\left({C}_{k}\right){\tau }_{ijk}$ (Eq. 3)

where i and j are the indices of grid, k the index of the trajectory, n the total number of trajectories used in analysis, Ck the pollutant concentration (PM2.5 conc. in this case) measured upon arrival of trajectory k, and τijk the residence time of trajectory k in grid cell (i, j). A high value of ${\overline{C}}_{ij}$ means that, air parcels passing over cell (i, j) would, on average, cause high concentrations at the receptor sites.

Gridded trajectory concentrations in Fig. 7 show that the stations were possibly receiving PM2.5 emitted from the long range sources both within and outside the national boundary. Within the boundary, the sources in the northern districts of Naogaon, Bogura and Rangpur were contributing to fine particles in the middle (Fig. 7a), south (Fig. 7d), and southeast (Fig. 7b) parts of the country. The south (Fig. 7d) and southeast (Fig. 7b) regions were also influenced by the sources in the middle of the country (Dhaka, Gazipur, and Narayanganj). Sources in the eastern districts (Narsindhi and Bramhanbaria) and northeastern districts (Sylhet) were also found contributing to PM2.5 concentrations at the south (Fig. 7d) and southeast (Fig. 7b) stations. Sylhet station was found gaining PM2.5 from the sources located in the middle of the country (Fig. 7c) during the time from December 2015 to January 2016. The northern districts possess a large number of rice mills and brick kilns; rice husks are widely used as fuel in those rice mills, and the brick kilns burn coals. Agricultural residue burning and open cooking are also broadly practiced in the northern districts. Emissions from these sources may have contributed to PM2.5 concentrations in other parts of the country (Fig. 7).

Gridded back trajectory concentrations showing mean PM2.5 concentrations on each grid cell of 1°×1° area, using CWT method.

Sources of Nepal, and Delhi-NCR and Uttar Pradesh regions of India could have contributed to the fine particles at the Gazipur station (Fig. 7a). Delhi-NCR is currently the top ranked polluted city in the world (WHO, 2018); Nepal and Pakistan are also heavily polluted (Gurung and Bell, 2012; Khatum, 2017). The stations could have also gained PM from the eastern state of Tripura and northeastern state of Assam/Meghalaya of India (Fig. 7). Barisal station was also found to have probability of getting PM from Nepal and its neighboring Indian regions (Fig. 7). Wind directions during dry season in Sylhet were different from those in Gazipur and Barisal (Fig. 2). While Gazipur and Barisal experienced air that entered Bangladesh territory through the western or northern borders, Sylhet received air that crossed the border through the north eastern borders. Thus, the Sylhet station was much affected by the sources in Meghalaya and Assam rather than sources in Nepal and its neighboring regions (Fig. 7). Sources of Meghalaya and Assam could also be crucial for PM pollution in other parts of Bangladesh as shown by Fig. 7.

4. CONCLUSIONS

Trend characteristics of PM concentrations in six cities of Bangladesh from 2013 to April 2018 were exhaustively studied. The cities were Dhaka, Gazipur and Narayanganj in the middle of the country, Chattogram to the southeast, Sylhet to the northeast and Barisal to the south of the country. The concentrations were monitored every minute with the continuous PM monitoring system BAM 1020 of the Met One Instrument Inc, USA. Two separate systems of the BAM 1020 were applied for monitoring the PM10 and PM2.5 concentrations. After the study the following observations and findings were noted,

a) Atmospheric PM concentrations in all of the regions of the country were notably influenced by the seasonal variations. Average PM10 concentrations at Dhaka, Chattogram, Narayanganj, Gazipur, Sylhet and Barisal stations in dry seasons (November-April) were 238.7±155.4, 190.7±108.5, 303.6±161.4, 227.3±142.7, 151.7±105.0 and 170.7±108.4 μg m-3 respectively whereas those in wet seasons (May-October) were 75.0±51.6, 55.5±40.8, 102.4±84.4, 60.6±48.5, 52.7±38.3, and 54.4±41.6 μg m-3 respectively. The ratios of PM2.5 concentrations to PM10 concentrations were comparatively higher in wintertime (December-January) and lower in summertime (February-April). The ratios in winter/summer time in Dhaka, Chattogram, Narayanganj, Gazipur, Sylhet and Barisal cities were 0.67/0.56, 0.64/0.53, 0.60/0.46, 0.70/0.58, 0.58/0.53 and 0.76/0.64 respectively. Ratios in wet season were 0.52, 0.55, 0.37, 0.58, 0.42 and 0.72 respectively.

b) The middle area of the country (Dhaka, Narayanganj and Gazipur) was found more polluted compared to other parts of the country, and the northeast region (Sylhet) was found the least polluted. Primary investigations reveal a relation between wind pattern and PM pollution level in the country-the northern, middle and southern part of the country (Dhaka, Narayanganj, Gazipur, Barisal) experienced wind mostly from the west and north-west directions during dry season and were characterized with very high PM pollution as well as higher contribution of fine fraction to the PM concentrations. In contrast, the southeast (Chattogram) and northeast (Sylhet) regions of the country having wind directions other than west and northwest directions in dry season experienced less pollution.

c) The diurnal variations of PM concentrations (both fractions) in Dhaka, Narayanganj, Gazipur and Barisal were mostly the same in winter and summer seasons-the variations were bimodal in pattern, having peaks at 9:00 am and 9:00 pm. Diurnal variations of PM in the city of Chottogram and Sylhet were found little different from other cities, especially the PM peak in the morning in these two cities were less pronounced. The nighttime PM10 and PM2.5 concentrations at Dhaka stations in winter season were on average 33.5 and 26.4% greater than the daytime concentrations respectively. Similarly, the nighttime PM10 and PM2.5 concentrations were greater from the daytime concentrations in other cities in winter respectively by 26.2 and 22.0% in Narayanganj, 63.8 and 54.6% in Gazipur, 45.6 and 45.5% in Chattogram, 85.4 and 98.7% in Sylhet, and 38.2 and 35.2% in Barisal. Daytime meteorology of the cities was characterized with comparatively greater solar radiation, wind speed and temperature, and lower relative humidity. Mixing height thus formed remains very deep at daytime compared to nighttime, giving a general tendency to generate high PM concentration at nighttime.

d) The study found that the sources from one region (within the boundary of the country) were contributing to PM in other regions located downwind. For example, some hotspots in the northwestern districts (Naogaon, Bogura, Rangpur) were identified contributing to PM concentrations in the middle and south part of the country. Outside the boundary, sources in Nepal, and Delhi-NCR and Uttar Pradesh regions of India could have contributed to fine particles at the Gazipur station. The stations could have also gained PM from the eastern state of Tripura and northeastern state of Assam and Meghalaya of India.

Acknowledgments

This work has been financially supported from the World Bank Credit to the Government of Bangladesh (Cr. 4581BD and Cr. 5924BD). Some of the works especially the data production was highly benefitted by the members of the air quality monitoring team of the Clean Air and Sustainable Environment Project of the Department of Environment, Dhaka. The Team was comprised of Dr. Swapan Kumar Biswas (Consultant), Ms. Sabera Nasrin (Data Quality Assurance Officer), Mr. Md. Mizanur Rahman (Sr. Technician) and Mr. Mohammad Ashrafuzzaman (Technician).

References

• BBS (2011) Bangladesh Bureau of Statistics: Population and housing census 2011, national volume 3: urban area report. Bangladesh Bereau of Statistics, Statistics and Information Division, Ministry of Planning, Bangladesh. http://www.bbs.gov.bd/, ..
• Begum, B.-A., Biswas, S.-K., Hopke, P.-K. (2008) Assessment of trends and present ambient concentrations of PM2.2 and PM10 in Dhaka, Bangladesh, Air Quality. Atmosphere and Health, 1(3), 125-133.. [https://doi.org/10.1007/s11869-008-0018-7]
• Begum, B.-A., Biswas, S.-K., Markwitz, A., Hopke, P.-K. (2010) Identification of sources of fine and coarse particulate matter in Dhaka, Bangladesh. Aerosol and Air Quality Research, 10(4), 345-353.. [https://doi.org/10.4209/aaqr.2009.12.0082]
• Begum, B.-A., Biswas, S.-K., Pandit, G.-G., Saradhi, I.-V., Waheed, S., Siddique, N., Seneviratne, M.-C.-S., Cohen, D.-D., Markwitz, A., Hopke, P.-K. (2011) Long-range transport of soil dust and smoke pollution in the South Asian region. Atmospheric Pollution Research, 2(2011), 151-157.. [https://doi.org/10.5094/APR.2011.020]
• Begum, B.-A., Hopke, P.-K., Markwitz, A. (2013) Air pollution by fine particulate matter in Bangladesh. Atmospheric Pollution Research, 4(1), 75-86.. [https://doi.org/10.5094/APR.2013.008]
• Begum, B.-A., Hopke, P.-K. (2018) Ambient air quality in Dhaka Bangladesh over two decades: Impacts of policy on air quality. Aerosol and Air Quality Research, 18, 1910-1920.. [https://doi.org/10.4209/aaqr.2017.11.0465]
• Carslaw, D.C. (2018) The openair manual - open-source tools for analyzing air pollution data. Manual for version 2.2-4. University of York, U.K..
• Carslaw, D.-C., Ropkins, K. (2012) Openair-an R package for air quality data analysis. Environmental Modelling Software, 27, 52-61.. [https://doi.org/10.1016/j.envsoft.2011.09.008]
• CASE (2014) Clean Air and Sustainable Environment Project: Identification and Apportionment of Sources from Air Particulate Matter at Urban Environments in Bangladesh. Department of Environment, Dhaka, Bangladesh, http://case.doe.gov.bd/, ..
• Chowdhury, Z., Zheng, M., Schauer, J.-J., Sheesley, R.-J., Salmon, L.-G., Cass, G.-R., Russell, A.-G. (2007) Speciation of ambient fine organic carbon particles and source apportionment of PM2.5 in Indian cities. Journal of Geophysical Research Atmosphere, 112(D15).. [https://doi.org/10.1029/2007JD008386]
• Dey, S., Di Girolamo, L., van Donkelaar, A., Tripathi, S.-N., Gupta, T., Mohan, M. (2012) Variability of outdoor fine particulate (PM2.5) concentration in the Indian Subcontinent: A remote sensing approach. Remote Sensing of Environment, 127, 153-161.. [https://doi.org/10.1016/j.rse.2012.08.021]
• Draxler, R.R., Rolph, G.D. (2003) HYSPLIT-4 (Hybrid Single Particle Lagrangian Integrated Trajectory) model. NOAA Air Resources Laboratory, Silver Spring, http://www.arl.noaa.gov/ready/hysplit4.html, ..
• Dumka, U.-C., Manchanda, R.-K., Sinha, P.-R., Sreenivasan, S., Moorthy, K.-K., Babu, S.-S. (2013) Temporal variability and radiative impact of black carbon aerosol over tropical urban station Hyderabad. Journal of Atmospheric and Solar-Terrestrial Physics, 105, 81-90.. [https://doi.org/10.1016/j.jastp.2013.08.003]
• Dumka, U.-C., Kaskaoutis, D.-G., Srivastava, M.-K., Devara, P.-C.-S. (2015) Scattering and absorption properties of nearsurface aerosol over Gangetic Himalayan region: the role of boundary-layer dynamics and long-range transport. Atmospheric Chemistry and Physics, 15(3), 1555-1572.. [https://doi.org/10.5194/acp-15-1555-2015]
• Feng, S., Gao, D., Liao, F., Zhou, F., Wang, X. (2016) The health effects of ambient PM2.5 and potential mechanisms. Ecotoxicology and Environmental Safety, 128, 67-74.. [https://doi.org/10.1016/j.ecoenv.2016.01.030]
• Gurung, A., Bell, M.-L. (2012) Exposure to airborne particulate matter in Kathmandu Valley, Nepal. Journal of Exposure Science and Environmental Epidemiology, 22(3), 235.. [https://doi.org/10.1038/jes.2012.14]
• Ibn Azkar, M.-B., Chatani, M.-A.-S., Sudo, K. (2012) Simulation of urban and regional air pollution in Bangladesh. Journal of Geophysical Research, 117(D07303).. [https://doi.org/10.1029/2011JD016509]
• Khanum, F., Chaudhry, M.-N., Kumar, P. (2017) Characterization of five-year observation data of fine particulate matter in the metropolitan area of Lahore. Air Quality, Atmosphere and Health, 10(6), 725-736.. [https://doi.org/10.1007/s11869-017-0464-1]
• Kopacz, M., Mauzerall, D.-L., Wang, J., Leibensperger, E.-M., Henze, D.-K., Singh, K. (2011) Origin and radiative forcing of black carbon transported to the Himalayas and Tibetan Plateau. Atmospheric Chemistry and Physics, 11(6), 2837-2852.. [https://doi.org/10.5194/acp-11-2837-2011]
• Lawrence, M.-G., Lelieveld, J. (2010) Atmospheric pollutant outflow from southern Asia: a review. Atmospheric Chemistry and Physics, 10(22), 11017.. [https://doi.org/10.5194/acp-10-11017-2010]
• Lelieveld, J., Evans, J.-S., Fnais, M., Giannadaki, G., Pozzer, A. (2015) The contribution of outdoor air pollution sources to premature mortality on a global scale. Nature, 525(7569), 367.. [https://doi.org/10.1038/nature15371]
• Ming, J., Cachier, H., Xiao, C., Qin, D., Kang, S., Hou, S., Xu, J. (2008) Black carbon record based on a shallow Himalayan ice core and its climatic implications. Atmospheric Chemistry and Physics, 8(5), 1343-1352.. [https://doi.org/10.5194/acp-8-1343-2008]
• Pant, P., Lal, R.-M., Guttikunda, S.-K., Russell, A.-G., Nagpure, A.-S., Ramaswami, A., Peltier, R.-E. (2018) Monitoring particulate matter in India: recent trends and future outlook. Air Quality, Atmosphere and Health, 12(1), 45-58.. [https://doi.org/10.1007/s11869-018-0629-6]
• Pommier, M., Fagerli, H., Gauss, M., Simpson, D., Sharma, S., Sinha, V., Ghude, S.-D., Landgren, O., Nyiri, A., Wind, P. (2018) Impact of regional climate change and future emission scenarios on surface O3 and PM2.5 over India. Atmospheric Chemistry and Physics, 18(1), 103-127.. [https://doi.org/10.5194/acp-18-103-2018]
• Ramanathan, V., Carmichael, G. (2008) Global and regional climate changes due to black carbon. Nature Geoscience, 1(4), 221.. [https://doi.org/10.1038/ngeo156]
• Rana, M.-M., Sulaiman, N., Sivertsen, B., Khan, M.-F., Nasreen, S. (2016a) Trends in atmospheric particulate matter in Dhaka, Bangladesh, and the vicinity. Environmental Science and Pollution Research, 23(17), 17393-17403.. [https://doi.org/10.1007/s11356-016-6950-4]
• Rana, M.-M., Mahmud, M., Khan, M.-H., Sivertsen, B., Sulaiman, N. (2016b) Investigating incursion of transboundary pollution into the atmosphere of Dhaka, Bangladesh. Advance in Meteorology (2016).. [https://doi.org/10.1155/2016/8318453]
• Schnell, J.-L., Naik, V., Horowitz, L.-W., Paulot, F., Mao, J., Ginoux, P., Zhao, M., Ram, K. (2018) Exploring the relationship between surface PM2.5 and meteorology in Northern India. Atmospheric Chemistry and Physics, 18(14), 10157-10175.. [https://doi.org/10.5194/acp-18-10157-2018]
• Seinfeld, J.H., Pandis, S.N. (1998) Atmospheric Chemistry and Physics-From air pollution to climate change. John Wiley and Sons Inc., New York.. [https://doi.org/10.1063/1.882420]
• Seibert, P., Kromp-Kolb, H., Baltensperger, U., Jost, D.-T., Schwikowski, M., Kasper, A., Puxbaum, H. (1994) Trajectory analysis of aerosol measurements at high alpine sites. Transport and Transformation of Pollutants in the Troposphere, 15(9), 689-693..
• Shahid, S. (2010) Rainfall variability and the trends of wet and dry periods in Bangladesh. International Journal of Climatology, 30(15), 2299-2313.. [https://doi.org/10.1002/joc.2053]
• Sharma, A.-R., Kharol, S.-K., Badarinath, K.-V.-S., Singh, D. (2010) Impact of agriculture crop residue burning on atmospheric aerosol loading-a study over Punjab State, India. Annales Geophysicae (09927689), 28(2).. [https://doi.org/10.5194/angeo-28-367-2010]
• Sharma, S.-K., Mandal, T.-K., Jain, S., Sharma, A., Saxena, M. (2016) Source apportionment of PM2.5 in Delhi, India using PMF model. Bulletin of Environmental Contamination Toxicology, 97(2), 286-293.. [https://doi.org/10.1007/s00128-016-1836-1]
• Shindell, D.T., Kuylenstierna, J.C.I., Raes, F., Ramanathan, V., Rosenthal, E., Terry, S. (2011) Integrated assessment of black carbon and tropospheric ozone: summary for decision makers. Nairobi: United Nations Environment Programme and World Meteorological Organization..
• State of Global Air (2019) A special report on global exposure to air pollution and its disease burden. Health Effect Institute, Boston, USA..
• Stull, R.B. (1988) An Introduction to Boundary Layer Meteorology. Springer, Dordrecht.. [https://doi.org/10.1007/978-94-009-3027-8]
• Tiwari, S., Bisht, D.-S., Srivastava, A.-K., Pipal, A.-S., Taneja, A., Srivastava, M.-K., Attri, S.-D. (2014) Variability in atmospheric particulates and meteorological effects on their mass concentrations over Delhi, India. Atmospheric Research, 145, 45-56.. [https://doi.org/10.1016/j.atmosres.2014.03.027]
• Uria-Tellaetxe, I., Carslaw, D.-C. (2014) Conditional bivariate probability function for source identification. Environmetal Modelling Software, 59, 1-9.. [https://doi.org/10.1016/j.envsoft.2014.05.002]
• Venkataraman, C., Habib, G., Eiguren-Fernandez, A., Miguel, A.-H., Friedlander, S.-K. (2005) Residential biofuels in South Asia: carbonaceous aerosol emissions and climate impacts. Science, 307(5714), 1454-1456.. [https://doi.org/10.1126/science.1104359]
• WHO (2018) World Health Organization: Ambient (outdoor) air quality database Summary results, update 2018. Geneva 27, Switzerland..
• Yasunari, T.-J., Bonasoni, P., Laj, P., Fujita, K., Vuillermoz, E., Marinoni, A., Cristofanelli, P., Duchi, R., Tartari, G., Lau, K.-M. (2010) Estimated impact of black carbon deposition during pre-monsoon season from Nepal Climate Observatory-Pyramid data and snow albedo changes over Himalayan glaciers. Atmospheric Chemistry and Physics, 10(14), 6603-6615.. [https://doi.org/10.5194/acp-10-6603-2010]

Fig. 1.

Locations of air monitoring sites with population sizes of the respective town/city.

Fig. 2.

Wind-roses at different stations in wet and dry season.

Fig. 3.

Time series graph of 24-h average PM10 and PM2.5 Concentrations at different cities of Bangladesh; horizontal lines are national limit values of PM10 (upper) and PM2.5 concentrations (lower).

Fig. 4.

CBPFs of PM2.5 concentrations in two percentile ranges (70-98 and 5-30) shown for different seasons in the cities. GZ=Gazipur, CTG=Chattogram, BAR=Barisal and SYL=Sylhet. The color bar shows the CPF probability of PM2.5 for the concentration range given in the parentheses below the figure.

Fig. 5.

Diurnal trends in PM10 and PM2.5 concentrations in different seasons in the cities.

Fig. 6.

Relationships among normalized level of diurnal variations of PM2.5 and meteorology parameters; ws=wind speed, T=temperature, RH=relative humidity, SR=solar radiation. gz=Gazipur, ctg=Chattogram, syl=Sylhet, bar=Barisal.

Fig. 7.

Gridded back trajectory concentrations showing mean PM2.5 concentrations on each grid cell of 1°×1° area, using CWT method.

Table 1.

Descriptions of the cities, sampling sites, and possible sources of air pollution.

City, area and population,
topography
Sampling site Possible sources of air pollution in the city
Dhaka
Capital and megacity; area
~316 km2, population ~10.0
million, Flat

Coordinates 23.78N; 90.36E, sample intake
~9.0 m above the ground. Main city road
~100 meter away.

Thousands of brick kilns around the outskirts of the
city, congested high emitting vehicle fleet, lots of small-
scale factories, road dusts, open burning, etc.

Gazipur
~25 km to the north from
Dhaka city; 47.23 km2, 0.21
million, Flat

23.99N; 90.42E, sample intake ~9.0 m above the
ground. 15 m away from a local road carrying low
traffic. The site is relatively unaffected by nearby
air pollution sources.

Lots of brick kilns are around, small-scale factories
(Battery, garments, dye, etc.), road dusts,
open burning, etc.

Narayanganj
~23 km to southeast from
Dhaka; 47.23 km2, 0.29
million, Flat

23.63N; 90.51E, sample intake at ~ 9.0 m above
the ground. A road intersection ~1.0 km west
from the site; diesel run water vessels run on the
Shitalakkha river only 200 m south from the site

Industrial town having jute processing plants, paper and
cement factories, and lots of small to medium industries
like textile and dye factories, steel mills. Congested
vehicle fleet and dusts and brick kilns.

Chattogram
~215 km to the southeast from
Dhaka; 155 km2, 2.6 million,
hilly

22.32N; 91.81E, sample intake at ~ 9.0 m above
the ground. Residential site not much influenced
by local sources.

Port city with huge congested traffic, diesel vehicles and
significant number of two-stroke baby taxis; brick kilns,
heavy industries, steel mills, and natural sea salt.

Sylhet
~200 km to the northeast from
Dhaka; 42.0 km2, 0.53 million,
hilly

24.89N; 91.87E, sample intake at ~14.0 m above
the ground. The river Surma is 20 m away to the
south. Two roads pass the site ~12 m to the east
and south.

Vehicles and brick kilns.

Barisal
~120 km to the south from
Dhaka; 69.0 km2, 0.34 million,
Flat
22.71N; 90.36E, sample intake at ~8.0 m above
the ground. A road with low traffic passes the site.
Barisal is a big river port, large number of diesel vessel
operates in the port; small number of vehicles, brick
kilns.

Table 2.

Year-wise rate (in percent) of valid data generated at the stations.

Station Component 2013 2014 2015 2016 2017 2018 (up to April)
Dhaka PM10 93.0 89.3 73.1 75.0 90.6 90.6
PM2.5 91.1 91.3 93.0 68.7 93.0 92.2
Chattogram PM10 85.8 61.1 67.3 77.8 74.3 61.5
PM2.5 83.8 74.1 70.3 72.9 38.5 no data
Gazipur PM10 77.3 91.7 83.58 70.0 69.0 75.6
PM2.5 76.7 85.9 69.5 75.0 64.7 73.7
Narayanganj PM10 91.2 82.7 86.3 79.3 78.7 82.5
PM2.5 69.2 68.2 56.8 69.6 62.7 74.8
Sylhet PM10 81.8 87.3 64.6 66.4 84.5 83.0
PM2.5 82.1 86.2 66.2 63.3 82.2 79.7
Barisal PM10 89.3 59.6 80.0 70.2 22.2 82.3
PM2.5 85.1 85.6 80.7 77.0 61.6 no data

Table 3.

Overview of the daily PM concentrations in different seasons in the cities of Bangladesh. Daily concentrations were calculated with minimum 80% data availability.

City Season PM10 conc. (μg m-3)
Percentiles
PM2.5 conc. (μg m-3)
Percentiles
25 50 75 Mean±sd 25 50 75 Mean±sd
*Annual mean was calculated with the data from 2013 to 2017.
Dhaka Winter 222.7 278.7 342.7 288.0±151.5 153.4 184.0 226.0 190.0±93.0
Summer 128.6 192.0 273.9 208.6±152.0 65.7 97.8 147.6 110.0±85.0
Wet 48.5 62.6 88.4 72.5±51.6 23.3 32.3 45.9 37.3±27.3
Annual* 66.3 114.8 229.3 155.8±140.2 33.1 55.0 129.0 86.1±85.0
Gazipur Winter 210.6 256.0 310.6 262.0±145.4 145.0 173.0 221.4 180.0±96.0
Summer 165.7 231.0 282.0 224.0±132.2 83.5 121.1 157.6 122.3±76.1
Wet 33.4 48.7 77.9 60.3±48.5 18.2 27.5 43.2 32.3±26.1
Annual 59.3 153.0 244.0 161.3±136.0 34.8 88.8 155.4 100.7±86.8
Narayanganj Winter 317.0 377.0 432.0 377.0±157.5 188.4 225.4 265.7 227.0±102
Summer 156.7 250.6 349.4 261.0±156.4 68.5 125.0 187.8 132.3±93.2
Wet 55.7 83.7 120.0 99.2±84.4 17.2 25.0 38.0 34.4±34.0
Annual 86.7 160.2 310.8 203.0±163.2 26.0 57.4 173.0 108.0±103
Chattogram Winter 182.0 216.0 260.0 224.0±113.0 116.6 138.5 167.8 143.3±76.0
Summer 108.0 166.4 237.6 173.8±100.0 52.4 85.9 125.3 91.2±63.4
Wet 34.4 45.7 61.6 53.7±40.8 17.6 22.3 31.3 28.4±21.3
Annual 51.8 113.7 200.2 132.2±106.0 26.8 64.4 120.2 78.7±106.0
Sylhet Winter 135.7 170.3 199.4 172.7±114.0 78.7 103.8 128.2 105.0±76.0
Summer 101.7 160 192.5 151.0±92.0 56.8 84.0 110.5 83.4±55.0
Wet 34.4 47.5 64.1 52.0±38.3 13.5 19.5 28.7 22.7±21.1
Annual 47.5 83.1 149.1 102.5±91.2 19.8 38.8 83.6 54.6±57.3
Barisal Winter 184.0 210.4 245.6 218.4±112.0 134.6 154.0 187.0 163.7±82.3
Summer 87.0 129.6 190.6 145.0±97.0 52.0 83.5 124.5 92.7±64.0
Wet 34.0 44.6 63.0 56.0±41.6 25.0 32.3 41.9 37.6±24.3
Annual 50.9 112.7 190.6 126.7±102.7 34.0 66.7 130.7 85.6±71.5

Table 4.

Average PM concentrations (μg m-3) at daytime and nighttime in different seasons in the cities.

Winter Summer Wet
Day (avg±sd) Night (avg±sd) Day (avg±sd) Night (avg±sd) Day (avg±sd) Night (avg±sd)
DHK=Dhaka, NG=Narayanganj, GZ=Gazipur, CTG=Chittagong, SYL=Sylhet, BAR=Barisal
DHK PM10 243.4±122.0 325.6±162.2 186.3±131.3 229.6±164.2 64.7±38.7 78.5±60.0
PM2.5 166.0±84.0 209.8±95.4 100.8±80.4 117.6±87.5 34.1± 23.1 39.5±30.2
NG PM10 331.3±143.8 418.2±155.5 224.5±127.5 275.8±173.3 93.0±66.2 107.0±96.5
PM2.5 203.9±105.0 248.6±96.0 110.0±85.2 136.1±97.7 32.1±29.0 37.0±39.2
GZ PM10 195.8±81.0 320.7±159.2 167.1±82.1 245.4±153.2 49.4±36.2 66.2±55.6
PM2.5 139.0±66.7 214.9±101.2 94.4±61.5 131.0±82.7 27.1±22.3 35.5±28.1
CTG PM10 178.8±71.7 260.3±127.0 162.0±82.2 184.0±111.5 46.0±26.0 59.6±49.3
PM2.5 114.8±59.4 167.0±81.6 80.8±57.7 98.2±67.0 23.8±15.2 30.8±25.2
SYL PM10 118.3±51.5 219.4±128.6 116.5±54.0 177.4±106.0 40.2±20.0 60.8±46.7
PM2.5 68.0±41.7 135.1±83.0 61.8±35.8 97.4±62.3 15.5±11.2 28.5±25.2
BAR PM10 180.2±58.6 249.0±133.4 118.0±57.0 151.8±118.0 52.1±30.7 55.6±48.7
PM2.5 135.0±49.8 182.5±96.4 76.3±44.5 94.8±74.8 35.5±20.0 36.1±27.1