AbstrAct

This paper presents seasonal variation of PM10 over five urban sites in Sabah, Malaysia for the period of January through December 2012. The variability of PM10 along with the diurnal and weekly cycles of CO, NO2, SO2, and O3 at Kota Kinabalu site were also dis­cussed to investigate the possible sources for incre­ased PM10 concentration at the site. This work is cru­cial to understand the behaviour and possible sourc­es of PM10 in the urban atmosphere of Sabah region. In Malaysia, many air pollution studies in the past focused in west Peninsular, but very few local studies were dedicated for Sabah region. This work aims to fill the gap by presenting the descriptive statistics on the variability of PM10 concentration in the urban atmosphere of Sabah. To further examine its diurnal and weekly cycle pattern, its responses towards the variations of CO, NO2, SO2, and ozone were also in­vestigated. The highest mean value of PM10 for the whole study period is seen from Tawau (35.7±17.8

μg m-3), while the lowest is from Keningau (31.9± 18.6 μg m-3). The concentrations of PM10 in all cities exhibited seasonal variations with the peak values occurred during the south­west monsoons. The PM10 data consistently exhibited strong correlations with traffic related gaseous pollutants (NO2, and CO), except for SO2 and O3. The analysis of diurnal cycles of PM10 levels indicated that two peaks were associ­ated during the morning and evening rush hours. The bimodal distribution of PM10, CO, and NO2 in the front and at the back of ozone peak is a representa­tion of urban air pollution pattern. In the weekly cycle, higher PM10, CO, and NO2 concentrations were observed during the weekday when compared to weekend. The characteristics of NO2 concentration

rationed to CO and SO2 suggests that mobile sourc­es is the dominant factor for the air pollution in Kota Kinabalu; particularly during weekdays.

Key words: Particulate Matter, Urban atmosphere, Air quality, Malaysia, Diurnal cycle, Weekly cycle

1. INTRODUCTION Particulate matter (PM10) consists of very small solid

and liquid particles suspended in air. It is of greatest concern to public health because these particles are small enough to be inhaled into the deepest parts of lung and settled there to cause adverse health effects. Particles less than 10 microns in diameter are known as PM10. It is a mixture of substances that include smoke, soot, dust, salt, acids, and metals. Particulate matter also forms when gases emitted from motor vehicles and industry undergo chemical reactions in the atmosphere. It is a major component of air pollution that threatens both our health and our environment. Sources of PM10 in both urban and rural areas are not limited to motor vehicles, but also dust from construction, landfills, and agriculture, wildfires and brush/waste burning, industri­al sources, windblown dust from open lands. Epidemi­ological studies indicated that the fine particle fractions e.g. PM2.5, and PM1.0 have considerable impacts on human health even at concentrations below the Malay­sia Ambient Air Quality Standards PM2.5 of 75 μg/m3 in daily 24­hour average (Gomiscek et al., 2004). These particles when inhaled by human can evade the respira­tory system’s natural defences and lodge deep in the lungs. Health problems begin as the body reacts to these foreign particles. PM10 can increase the number and

Ozone Concentration in the Morning in Inland Kanto Region 109Asian Journal of Atmospheric EnvironmentVol. 12, No. 2, pp. 109-126, June 2018doi: https://doi.org/10.5572/ajae.2018.12.2.109ISSN (Online) 2287-1160, ISSN (Print) 1976-6912

Variability of the PM10 concentration in the urban Atmosphere of sabah and its responses to diurnal and weekly changes of cO, NO2, sO2 and OzoneJackson CHANG Hian Wui1),*, CHEE Fuei Pien2), Steven KONG Soon Kai3) and Justin SENTIAN4)

1)Preparatory Center for Science and Technology, Universiti Malaysia Sabah, Jalan UMS, 88400 Kota Kinabalu, Sabah, Malaysia 2)Energy, Vibration and Sound Research Group (e­VIBS), Faculty Science and Natural Resources, Universiti Malaysia Sabah, Jalan UMS, 88400 Kota Kinabalu, Sabah, Malaysia 3)Department of Atmospheric Sciences, National Central University, Taoyuan 32001, Taiwan 4)Climate Change Research Group (CCRG), Faculty Science and Natural Resources, Universiti Malaysia Sabah, Jalan UMS, 88400 Kota Kinabalu, Sabah, Malaysia

*Corresponding author. Tel: +60­088­320 000, E­mail: jacksonchw@ums.edu.my

110 Asian Journal of Atmospheric Environment, Vol. 12(2), 109-126, 2018

severity of asthma attacks, cause or aggravate bronchi­tis and other lung diseases, and reduce the body’s abil­ity to fight infections (Lelieveld et al., 2015; Nirmalkar and Deb, 2015). Certain people who include children, the elderly, exercising adults, and those suffering from asthma or bronchitis are especially vulnerable to PM10’s adverse health effects (Thurston et al., 2016).

The issue of PM10 pollution in the urban atmosphere has received much attentions in recent years as an in­creasing share of the world’s population lives in urban areas. Previous studies reported that traffic­related emis­sion accounted at least 50% of the total PM emission in the urban centres (Wrobel et al., 2000). Vehicles pro­duce exhaust and non­exhaust emissions. Aerosol ex­haust emissions can be emitted directly as particles such as soot and carbonaceous aggregates that formed during the fuel combustion in the engine or formed during the emission, dilution, mixing and cooling of the vehicle exhaust gases in ambient air (Adame et al., 2014; Pérez et al., 2010). Non­combustion traffic­related PM sour­ces are the result of the resuspension of road dust orig­inating from the mechanical wear and degradation of tires, brakes, and pavement abrasion (Bathmanabhan et al., 2010). Therefore, the contributions to the PM10 con­centration in ambient air are segregated into two groups: (1) primary particles emitted by vehicle exhausts, and (2) nucleation particles in ambient air (Rodríguez and Cuevas, 2007). The primary particles emitted by vehi­cles exhausts tend to exhibit a size distribution with a nucleation mode (<30 nm) and a carbonaceous mode

(50 to >100 nm). The gaseous exhaust emissions may also contribute to new particle formation in ambient air by in situ nucleation occurring sometime after emission

(hours to days). This process frequently occurs in tran­sit from the road to the urban background especially when the exhaust gas is fully diluted within the ambi­ent air and is photo­oxidised by reactive species.

In Malaysia, the government routinely publicized the air pollution index (API) readings throughout the mainstream media to alert the public of any looming smoke haze episode. Severe haze episode often occurs during the Southeast Asian haze resulted by smoke from forest fires in Kalimantan and Sumatra (Aouizerats et al., 2015; Harrison et al., 2009). The location and back­ground of a station, as well as wind speed, seasonal

(monsoon) and weekdays­weekend variations also play important role in influencing PM10 anomalies (Shaadan et al., 2015). Diurnal patterns, rationed between major air pollutants and sensitivity analysis, indicate that the influence of local traffic emissions on urban air quality is critical (Talib et al., 2014). A case study at three sel­ected stations in Malaysia e.g. Petaling Jaya, Melaka, and Kuching, also revealed that the major source of air pollution over urban air is mostly due to the combustion

of fossil fuel in motor vehicles and industrial activities

(Norsukhairin et al., 2013). A prediction study by PCA had further identified that CH4, NmHC, THC, O3, and PM10 are amongst the most significant factors deterio­rating air quality in Malaysia (Azid et al., 2014). A cri­tical understanding of the behaviour of major pollutants in urban atmosphere of big city is important on health concerns. It is also an essential contingency for policy implementation especially when the elevated PM con­centration has caused enormous impacts on economical values (Othman et al., 2014).

The paper aims to present the descriptive statistics on the PM10 concentration in the urban atmosphere of Sabah. Understanding the variation of PM10 concentra­tions is crucial for proper health risk assessment and risk management. PM10 was chosen because it is one of the major air pollutants listed in the Recommended Malaysian Air Quality Guideline (RMAQG) and it is also one of the five air pollutants included in the calcu­lation of Malaysian Air Quality Index (MAPI) together with ozone (O3), carbon monoxide (CO), sulfur dioxide

(SO2) and nitrogen dioxide (NO2). This paper also pre­sents the results of the variability of PM10 concentra­tion owing to the diurnal evolution and weekly cycles of CO, NO2, SO2, and ozone, particularly in Kota Kina­balu.

2. MEASUREMENT AND SITEThe PM10 monitoring sites are located at five differ­

ent stations in Sabah. Fig. 1 shows the regional site map of the five monitoring stations. The location details of each station is depicted in Table 1. The town of Kota Kinabalu is the busiest urban center in Sabah, located in the eastern part of Malaysia with a population of approximately 460,000 and area 351 km2. Its industry is not very active and the precursor emissions are main­ly due to traffic exhaust (Dominick et al., 2012; San­suddin et al., 2011). Kota Kinabalu is at 5°58ʹ17ʺN 116°05ʹ43ʺE and with an altitude of 5 m. The area is topographically homogenous and the city lies on a nar­row flatland between the Crocker Range to the east and the South China Sea to the west. The average tempera­ture is 27°C with April and May are the hottest months while January is the coolest. The average annual rain­fall is around 2,400 mm and varies remarkably through­out the year. February and March are typically the dri­est months while rainfall peaks in the inter­monsoon period in October. The wind speed ranges from 5.5 to 7.9 m/s during the Northeast Monsoon (NEM) but is significantly lower to 0.3 to 3.3 m/s during Southwest Monsoon (SWM).

Tawau is on the south­east coast of Sabah surround

Variability of the PM10 Concentration in Urban Atmosphere of Sabah 111

by the Sulu Sea in the east and located at 540 kilometres south­east of Kota Kinabalu. It has tropical rainforest climate under the Köppen climate classification. The climate is relatively hot and wet with average shade temperature about 26°C, with 29°C at noon and falling to around 23°C at night. The town sees precipitation throughout the year, with a tendency for November, December and January to be the wettest months, while February and March are the driest months. Tawau’s mean rainfall varies from 1,800 mm to 2,500 mm.

Labuan’s area comprises the main island (Labuan Island ­ 91.64 square kilometres) and six other smaller islands namely Burung, Daat, Kuraman, Big Rusukan, Small Rusukan and Papan island with a total area of 91.64 square kilometres. It has a tropical rainforest cli­mate with no dry season. Over the course of a year, the temperature typically varies from 25 to 32°C and is rarely below 24°C or above 33°C. Thunderstorms are the most severe precipitation observed in Labuan during 60% of those days with precipitation. They are most

likely around October, when they occur very frequently. Meanwhile, the relative humidity for Labuan typically ranges from 63% (mildly humid) to 96% (very humid) over the course of the year.

Keningau is the capital of the Keningau District in the Interior Division of Sabah, Malaysia. This city has a tropical climate. The temperatures are highest on ave­rage in May, at around 26.4°C. The lowest average tem­peratures in the year occur in January, at around 25.3°C The average annual temperature is 25.8°C in Keningau. The least amount of rainfall occurs in August and the greatest amount of precipitation occurs in May, with an average of 203 mm. The average annual rainfall is 1,825 mm.

Sandakan is located on the eastern coast of Sabah facing the Sulu Sea, with the town is known as one of the port towns in Malaysia. The town is located approx­imately 1,900 kilometres from 319 kilometres from Kota Kinabalu, the capital of Sabah. It has a tropical rainforest climate under the Köppen climate classifica­

6°0.000000000ʹN

4°0.000000000ʹN

6°0.000000000ʹN

4°0.000000000ʹN

114°0.000000000ʹE 116°0.000000000ʹE 118°0.000000000ʹE

114°0.000000000ʹE 116°0.000000000ʹE 118°0.000000000ʹE

50 0 50 100 150 200 km

Fig. 1. Regional site map of the CAQM monitoring stations in Sabah. The five stations are labelled as CA0030: Kota Kinabalu; CA0039: Tawau; CA0042: Labuan; CA0049: Keningau; CA0050: Sandakan.

Table 1. Location details of monitoring site in Sabah.

Site Site ID Location Latitude Longitude Type

Kota Kinabalu CA0030 SM Putatan, Kota Kinabalu, Sabah N05°53.623 E116°02.596 ResidentialTawau CA0039 Pejabat JKR, Tawau, Sabah N04°15.016 E117°56.166 ResidentialLabuan CA0042 Taman Perumahan MPL, Labuan, Sabah N05°19.980 E115°14.315 ResidentialKeningau CA0049 SMK Gunsanad, Keningau, Sabah N05°20.313 E116°09.769 ResidentialSandakan CA0050 Pejabat JKR Sandakan, Sandakan, Sabah N05°51.865 E118°05.479 Residential

112 Asian Journal of Atmospheric Environment, Vol. 12(2), 109-126, 2018

tion. The climate is relatively hot and wet with average shade temperature about 32°C, with around 32°C at noon falling to around 27°C at night. The town sees precipitation throughout the year, with a tendency for October to February to be the wettest months, while April is the driest month. Its mean rainfall varies from 2,184 mm to 3,988 mm.

All selected air quality monitoring stations are oper­ated by Alam Sekitar Sdn. Bhd. (ASMA) under the administration of Department of Meteorology (DOE). The Department of Environment (DOE) monitors the country’s ambient air quality through a network of Con­tinuous Air Quality Monitoring (CAQM). Automatic monitoring is designed to collect and measure data con­tinuously 24 hours a day during the monitoring period. Automatic Continuous Air Monitoring Stations typical­ly include: (a) measurement instrumentation (for both pollutant gases and meteorological parameters); (b) sup­port instrumentation (support gases, calibration equip­ment); (c) instrument shelters (temperature controlled enclosures); and (d) data acquisition system for data collection and storage. The concentration levels of pol­lutants in the ambient air such as sulphur dioxide (SO2), nitrogen dioxide (NO2), ozone (O3) and carbon monox­ide (CO) are measured hourly and analysed in time­weighted average (TWA) in this work. The near­surface atmospheric aerosols are regularly monitored using con­tinuous particulate matter (PM) monitoring instrument with the adoption of Met­One Beta Attenuation Method

(BAM) and manual 24­h monitoring system High Vol­ume Air Samplers (HVAS). Instruments from the Tele­dyne Technologies Inc., USA, such as Teledyne API Model 100A/100E, Teledyne API Model 300/300E, Teledyne API Model 400/400E and Teledyne API Model 200A/200E are used to monitor SO2, CO, O3 and NO2 respectively. The gases SO2, NO2 and O3 are determin­ed using the UV fluorescence method, chemilumines­cence detection method and UV absorption (Beer Lam­bert) method respectively.

3. RESULTS AND DISCUSSION

3. 1 Variability of PM10 in the Urban Atmosphere of Sabah

Fig. 2 shows the histogram of PM10 concentration measured at five distinct study areas in Sabah (a) Kota Kinablu, (b) Tawau, (c) Labuan, (d) Keningau, (e) San­dakan from Jan to Dec 2012. Within the measurement period, the average hourly PM10 measured at Tawau was 35.7±17.8 μg/m3, Kota Kinabalu (35.0±18.4 μg/m3), Labuan (34.7±18.1 μg/m3), Sandakan (32.8±11.5

μg/m3), and Keningau (31.9±18.6 μg/m3). Dispersity of the data was the least at Sandakan with the nominal

min and max at 8.0 μg/m3 and 104 μg/m3, respectively. The variability of PM concentrations observed at differ­ent sites are quite different where out of the five moni­toring stations, Kota Kinabalu measured the highest range (KK>KEN>LBN>TWU>SDK), the highest maxima (KK>KEN>LBN>TWU>SDK), and the second highest average (TWU>KK>LBN>SDK> KEN). This could be explained by their geographical environmental conditions and population density. Un­doubtedly, Kota Kinabalu has the highest PM10 concen­tration because it is the state capital of Sabah with pop­ulation of 462,963 for total area 352 meter­sq and pop­ulation density 1,315/km2 (Department of Statistics Malaysia, 2010). Of all monitoring stations studied in this work, Kota Kinabalu is the most urbanized area with high concentration of cities, industrial and eco­nomic activities. Population density is a fundamental amplifier of air pollution problems, and urban popula­tion growth has been an important factor in worsening air quality (Yara et al., 2017). The EPA data also indi­cates a strong association both between higher popula­tion densities and higher traffic densities, higher popu­lation densities and higher road vehicle nitrogen oxides

(NOx) emission intensities. The second highest popu­lation density is remarked in Labuan for 948.4/km2, putting it the mid of all monitoring stations. Besides, Labuan is also a federal territory of Malaysia that best known as being an offshore support hub for deepwater oil and gas activities in the region. On the other hand, Sandakan and Tawau have relatively lower population density at 179.8/km2 and 67.06/km2, respectively. The lowest population density is remarked in Keningau

(50.12/km2) which explains its lowest hourly­averaged PM10 concentration among all other stations. However, it has the second highest range and maxima due to its geographical location near to Kota Kinabalu. The local source from Kota Kinabalu and the advection due to wind could be one of the reasons responsible for trans­boundary aerosol between these two regions.

The highest maximum ~317 μg/m3 was measured at Kota Kinabalu on Jan 17th during the morning hours 0700h­0800h, while the PM10 concentration of other study areas remained low <80 μg/m3 on the same day and time. The exact reason for the unusual high PM10 occurred only on Jan, 17th from 0700h­0800h was un­known but we ruled out the regional or transboundary effects as the wind speed and wind direction was quite low at 1.7 km/h (153°N) and showed no unusual pattern throughout the past three days (see Fig. 3). Not only that, CO and NO2 also showed no irregular pattern and fluctuated within the normal range between 0.10­1.60

ppm and 0.005 to 0.020 ppm, respectively. Besides, the month of January is also not the open biomass burning period in Kalimantan and Sumatera, which is usually

Variability of the PM10 Concentration in Urban Atmosphere of Sabah 113

the main cause for low visibility and haze episode in Sabah. Therefore, the transboundary effect in this case is unlikely to cause this phenomenon. In fact, it could be due to other causes but not limited to local emissions related to traffic and open burning. Unfortunately, the local VOC data measured at the same site during the episode day was unavailable and this unusual incident was also not widely archived due to the TWA­24 h of the day was still within the EPA standard of less than

150 μg/m3. Fig. 4 shows the seasonal variation of PM10 concen­

tration measured at the five distinct study areas in Sabah from Jan 2012 to Dec 2012. The temporal distribution indicated that the PM10 levels remain highly concentrat­ed during the south­west monsoon (SWM) and post south­west monsoon (POST­SWM) which is the hot and dry season in Sabah. From the statistics summary, the highest PM10 concentration measured was always

Fig. 2. Histogram of PM10 concentration for (a) CA0030, (b) CA0039, (c) CA0042, (d) CA0049, and (e) CA0050 from Jan 2012 to Dec 2012.

12%

10%

8%

6%

4%

2%

0%

(a) (b)

(C) (d)

(e)

12%

10%

8%

6%

4%

2%

0%

18%

16%

14%

12%

10%

8%

6%

4%

2%

0%

8%

7%

6%

5%

4%

3%

2%

1%

0%

8%

7%

6%

5%

4%

3%

2%

1%

0%

6.58

12.8

819

.18

25.4

831

.79

38.0

944

.39

50.7

057

.00

63.3

069

.61

75.9

182

.21

88.5

294

.82

101.

1210

7.42

113.

73

6.52

12.6

118

.69

24.7

830

.86

36.9

543

.03

49.1

255

.20

61.2

967

.37

73.4

679

.54

85.6

391

.71

97.8

010

3.88

109.

9711

6.05

122.

14

6.54

15.8

025

.07

34.3

343

.59

52.8

562

.11

71.3

780

.63

89.8

999

.15

108.

4111

7.67

126.

9313

6.20

145.

4615

4.72

163.

9817

3.24

182.

50

6.59

16.1

225

.66

35.1

944

.73

54.2

663

.79

73.3

382

.86

92.4

010

1.93

111.

4712

1.00

130.

5314

0.07

149.

6015

9.14

168.

6717

8.21

8.98

14.8

620

.73

26.6

132

.49

38.3

744

.24

50.1

256

.00

61.8

867

.76

73.6

379

.51

85.3

991

.27

97.1

410

3.02

FrequencyNormal

FrequencyNormal

FrequencyNormal

FrequencyNormal

FrequencyNormal

114 Asian Journal of Atmospheric Environment, Vol. 12(2), 109-126, 2018

during the POST­SWM for all five study areas. Besides, no significant difference was found in the highest aver­age PM10 among the sites, ranging from 37 to 40 μg/m3. After the SWM, the PM10 levels started to decline dur­ing the north­east monsoon (NEM) and post north­east monsoon (POST­NEM) for all areas. The lowest PM10 concentration was measured at Keningau (27±13 μg/m3) during the POST­NEM. This finding is similar to another local study in Klang Valley reported that PM10 level was persistently high during the south­west mon­soon and low during the north­east monsoon (Rahman et al., 2015; Sansuddin et al., 2011). During the north­east monsoon, the precipitation of rain will carry the pollutants to the earth; hence, reducing the level of pol­lutants in the atmosphere. However, during the south­west monsoon, the warmer air near the surface area rises to higher latitude, which results in turbulence and causes the pollutants to become unstable; thus resulting

in a high level of pollutants in the atmosphere (Barm­padimos et al., 2011). During the wet seasons, precipi­tation reduces considerably PM10 concentration by wet deposition. In addition, if it is frontal precipitation, some polluted boundary layer air is replaced by clean air from aloft. This feature may not be possible to depict by a linear relationship, but the pattern is significant and obvious for case­by­case comparisons. For example, as shown in Fig. 4d, the PM10 concentration remained positively correlated with the dry seasons during SWM which peaked in the POST­SWM at an average value 37 μg/m3, and negatively correlated with the wet sea­son during NEM which based in the POST­NEM at an average value 27 μg/m3. Nevertheless, the seasonal vari­ation of PM10 concentration in Sabah has no drastic change but limit to an extent of less than <10 μg/m3 for all study areas. This attribute could be due to the small seasonal variation of planetary boundary layer

Table 2. Summary statistics of PM10 concentration measured in μg/m3 at five distinct study areas in Sabah from Jan 2012 to Dec 2012.

Statistics CA0030 CA0039 CA0042 CA0049 CA0050

Average 35.0 35.7 34.7 31.9 32.8Standard Deviation 18.4 17.8 18.1 18.6 11.5Skew 1.90 1.10 2.40 2.70 0.90Excess Kurtosis 11.4 1.8 11.5 13.3 1.8

Median 32.0 33.0 31.0 27.0 31.0Minimum 5.0 5.0 5.0 5.0 8.0Maximum 317.0 148.0 218.0 237.0 104.01st Quartile 23.0 23.0 23.0 21.0 24.03rd Quartile 43.0 45.0 41.0 37.0 39.0

45

40

35

30

25

20

15

10

5

0

TWA­24 h PM1016­Jan: 56 μg/m3

17­Jan: 71 μg/m3

18­Jan: 52 μg/m3

Wind speed (km/h)CO ( × 10 ppm)NO2

( × 1,000 ppm)PM10

(μg/m3)

16­Jan 17­Jan 18­Jan

0100

0300

0500

0700

0900

1100

1300

1500

1700

1900

2100

2300

0100

0300

0500

0700

0900

1100

1300

1500

1700

1900

2100

2300

0100

0300

0500

0700

0900

1100

1300

1500

1700

1900

2100

2300

350

300

250

200

150

100

50

0

WS

| CO

| N

O2

PM10

Fig. 3. Hourly variation in WS, PM10, CO, NO2 concentration measured in Kota Kinabalu from Jan, 16th to 18th, 2012.

Variability of the PM10 Concentration in Urban Atmosphere of Sabah 115

(PBL) in tropics. Previous study revealed that the influ­ence of planetary boundary layer height stands the high­est absolute correlation on the PM10 level in Poznan, Poland (Czernecki et al., 2017). Notably, increasing values of net irradiance are associated with strong de­crease of PM10 concentration due to large sensible heat flux which in turn drives the formation of deeper atmos­pheric boundary layers (Barmpadimos et al., 2011). In contrast, the relationship between the PM10 concentra­tion and net irradiance is reversed for decreasing solar irradiance. All five study areas are in tropic where recei­

ve fairly constant amount of solar irradiance throughout the year. Hence, it is possible that through this mecha­nism, relatively small difference of <10 μg/m3 between seasonal PM10 concentration was remarked for all study areas in Sabah.

3. 2 Diurnal Evolution of PM10 in Kota Kinabalu

Fig. 5 presents the diurnal variation of (a) PM10, (b) CO, (c) Ozone, (d) NO2, and (e) SO2 concentration measured for the total period of 2012 over the study

350

300

250

200

150

100

50

0

350

300

250

200

150

100

50

0

350

300

250

200

150

100

50

0

350

300

250

200

150

100

50

0

350

300

250

200

150

100

50

0

(a) (b)

(c) (d)

(e)

NEM POST­NEM SWM POST­SWM

NEM POST­NEM SWM POST­SWM

NEM POST­NEM SWM POST­SWM

NEM POST­NEM SWM POST­SWM

NEM POST­NEM SWM POST­SWM

Fig. 4. Seasonal variation of PM10 concentration measured at (a) CA0030, (b) CA0039, (c) CA0042, (d) CA0049, and (e) CA0050 from Jan 2012 to Dec 2012.

116 Asian Journal of Atmospheric Environment, Vol. 12(2), 109-126, 2018

area, Kota Kinabalu in the form of box plot. Each box is determined by the 25th and 75th percentiles, and the whiskers are marked at 5th and 95th percentiles. The median value at 50th percentile are marked with solid lines inside the boxes, while the maximum and mini­mum values are marked with solid lines outside the boxes. The box plot gives a good insight into the dyna­mics of the PM10 concentration for each hour. The aver­aged diurnal variation of PM10 showed pronounced vari­ations with two significant peaks during the morning

hours at 7 h and the evening hours at 19 h (see Fig. 5a). Fig. 5b shows that the hourly averaged CO concentra­tion rose abruptly at 7 h and gradually decreased to the normal background concentration <0.50 ppm at 10 h and then started to hike up again at 18 h and persisted until 23 h. This bimodal distribution was also found in the hourly averaged NO2 concentration which peaks at 7 h and 19 h (see Fig. 5d). Nevertheless, the hourly max­ima for both parameters CO and NO2 was still far be­yond the hourly Recommended Malaysia Air Quality

350.0300.0250.0200.0150.0100.050.00.0

(a)

(b)

(c)

(d)

(e)

3.002.502.001.501.000.500.00

0.050

0.040

0.030

0.020

0.010

0.000

0.0300.0250.0200.0150.0100.0050.000

0.0080.0070.0060.0050.0040.0030.0020.0010.000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Time (h)

PM10

(μg

m-

3 )C

O (p

pm)

O3 (p

pm)

NO

2 (ppm

)SO

2 (ppm

)

Fig. 5. Hourly averaged boxplot for (a) PM10, (b) CO, (c) O3, (d) NO2, and (e) SO2 concentration measured in Kota Kinabalu for the total period in 2012.

Variability of the PM10 Concentration in Urban Atmosphere of Sabah 117

Guidelines (RMAQG) of less than 30 ppm and 0.17

ppm, respectively. In contrast, the diurnal evolution of ozone and SO2 has no direct effect on the PM10 trend. The ozone profile was in accordance with the photo­chemical effects where it peaks at the noon hours for high intensity of UV solar light. Fig. 5c clearly exem­plifies this pattern on its unimodal distribution which has a bell­curve shape peaks at 13 h. Among all para­meters, SO2 has the least variability throughout the day where the fluctuation shows no significant anomalies and far behind the hourly RMAQG of <0.13 ppm.

The two rushing hours at 7 h and 19 h are associated with the heavy­traffic hours where the number of vehi­cles on road is the direct reflection of the related emis­sion of CO, NO2, and PM10. The morning peak hour is when people are going to work and sending their child­ren to school, thus causing traffic congestion and incre­asing the pollutant levels. Similar findings were also reported in a study at Klang Valley where the daily concentration of PM10, CO, NO2 and SO2 was clearly recorded as being the highest during traffic congestion, particularly during the morning peak between 7:00 am­ 9:00 am, led to the higher amount of CO in the atmo­sphere (Azmi et al., 2010). In the latter part of the after­noon, PM10 concentration peaked from 7:00 pm in line with the evening rush hour when people started return­ing home from work. The late evening peak can also be attributed to meteorological conditions, particularly atmospheric stability and wind speed (Rahman et al., 2015). The PM levels were lower between 0800h and 1400h because of low traffic volumes and hence low emission rates. Another possible reason could be the

increase in the mixing height that favors dispersion con­ditions (Bathmanabhan et al., 2010). However, during the night time between 2000h and 2300h, the PM con­centrations were slightly higher than the daytime (see Fig. 5). The probable reason for this may be the built­up of particles under inversion conditions. Gomiscek et al. (2004) have also reported a similar trend for three urban sites in Austria.

More specifically, we present two scenarios: (1) the morning effect and (2) the evening effect in the urban atmosphere at Kota Kinabalu on Fig. 6. The left panel Fig. 6a shows the spikes on PM levels due to the eve­ning effect and the right panel Fig. 6b shows the spikes of the morning effect. On the evening effect, the rise of PM level started at 1800h and fall abruptly after 2000h. On the morning effect, the PM level started to rise at 0700h reaching maxima 190 μg/m3 and then fall abrupt­ly to reaching 190 μg/m3 at 0800h. These two effects are highly related to the traffic exhaust gaseous emis­sion such as CO and NO2. Nitrogen dioxides have a substantial impact on PM10 through their atmospheric oxidation to aerosol nitrate and the CO formed from oxidation of VOCs (Wang et al., 2015). The concentra­tions and compositions of nitrate aerosols are determin­ed primarily by their precursor emissions, which are mainly of anthropogenic origin. Among the major pol­lutants, CO, nitrogen oxides (NOx = NO + NO2), PM10, and some types of VOCs (e.g., BTEX: benzene, toluene, ethylbenzene, and ortho­, meta­, and para­xylenes) are primarily traffic­induced, while O3 and NO2 are second­ary trace gases formed from precursors in photochemi­cal reactions (Yoo et al., 2015). The increase of CO was

200

180

160

140

120

100

80

60

40

20

0

200

180

160

140

120

100

80

60

40

20

0

1­Jan 2­Jan 15­Jan 16­Jan

2000

2200

2400

0200

0400

0600

0800

1000

1200

1400

1600

1800

2000

2200

2400

0100

0400

0700

1000

1300

1600

1900

2200

0100

0400

0700

1000

1300

1600

1900

2200

PM10

con

cent

ratio

n (μg

m-

3 )

PM10

con

cent

ratio

n (μg

m-

3 )

Fig. 6. The hourly mean PM10 concentration at Kota Kinabalu from 1­Jan to 2­Jan, 2012 (left) and 15­Jan to 16­Jan, 2012

(right). Left panel shows spikes on the evening effect and right panel shows the morning effect.

118 Asian Journal of Atmospheric Environment, Vol. 12(2), 109-126, 2018

substantially driven by the road traffic emissions of the day. Similar findings are also reported in Masiol et al. (2014) where species mainly linked to road vehicle exhaust emissions such as CO, and NO2 have a bimod­al structure due to the peaks in traffic at 7­9 am and 6­8 pm.

Fig. 7a and 7c shows the diurnal changes of PM10 with comparison to CO and NO2, respectively. The mor­ning peak on PM10 level at 0700h was observed con­currently with the rise on CO and NO2, and the abrupt fall on PM10 between 0900h and 1500h was also con­current with the decrease on both parameters. Similar pattern was observed for the evening peak on PM10 level at 1800h where the peak was corresponding to the rise on CO and NO2 and the abrupt fall of PM10 after 2000h was consistent to the decrease of CO and NO2

(see Fig. 8a and 8c). This pattern was further confirm­ed by the correlation analysis on the scatter plot PM10 against CO and NO2 on Fig. 7b and 7d, respectively. Positive correlation R2 of 0.46 and 0.23 was remarked on PM10 against CO and NO2, respectively. Same goes to the correlation analysis for the evening effect where positive correlation R2 of 0.45 and 0.42 was remarked for CO and NO2

(see Fig. 8b and 8d). The higher corre­lative strength of CO further implies that PM10 level in Kota Kinabalu is more dominantly dependent on the diurnal evolution of CO concentration. This finding was in good agreement with other studies reporting that the daily patterns of each pollutant for NO, NO2, CO and PM10 peak during the morning rush hours, and fall abruptly in the noon owing to the decrease in the inten­sity of emission and the rapid growth of the convective boundary layers (Adame et al., 2014).

The profile of SO2 showed no pronounced changes for both the morning and evening effect. The trend is rather irregular and has no significant pattern against the changes of PM10 level (see Figs. 7e and 8e). Both figures show no concurrent rise of PM10 and SO2, nei­ther the significant drop of both parameters occurred at the same time. Therefore, the effects of SO2 on the evolution of PM10 were somewhat less reactive com­pared to other gaseous e.g. NO2 and CO. A very weak correlative strength of 0.09 and 0.01 was found in the scatter plot between PM10 and SO2 for morning effect and evening effect respectively (see Figs. 7f and 8f), indi cating the changes of both parameters are not entirely related and significant. Another possible rea­son owing to the low variability of SO2 is due to SO2 itself has a very short lifetime and can easily transform into sulphate SO4 through oxidation and interaction with parti cles and water vapour (Kai et al., 2007).

Ozone is a reactive oxidant gas that plays an essential role in the photochemical air pollution and atmospheric oxidation processes. Although in the upper atmosphere

it acts as a barrier for UV­rays, in the troposphere it is a secondary air pollutant induced through a series of com­plex photochemical reactions involving solar radiation and ozone­precursors gaseous such as NO2

(Masiol et al., 2014). When the concentration of NO2, CO, and PM10 started to increase in the morning hours, surface ozone level decreased due to titration reaction with NO and the rupture of the nocturnal inversion layer (Talib et al., 2014). Figs. 7g and 8g supported this findings by showing the daily pattern of ozone decreased for incre­asing PM10 and vice versa. Ozone concentrations con­tinued to increase during the noon hours due to photo­chemical reaction and or horizontal and vertical trans­port processed and reached to a daily maxima 0.035

ppm at 1330h. The decreased in ozone levels was fol­lowed by a reduction in solar radiation when reaching to evening hours at 1700h. This was also associated with the increase of NO2 and PM10 during the evening hours when emission of traffic started to mount up. Over the Borneo, the air mass evolution during the sunset is fast er compared to the Peninsular. In other words, solar inten­sity was dimmed earlier at 1800h in Borneo regions. This is the reason why the ozone level reached to the daily minima 0.005 ppm at earlier time 1900h where the solar activity induced photochemical reaction was no longer active. Hence, a negative correlative R2 -1.35

(see Fig. 7h) and -2.77 (see Fig. 8h) was observed bet­ween the ozone and PM10 concentration, indicating an increase on PM10 concentration was most likely associ­ated with the decreased on ozone levels and vice versa. In general, CO, NO, and PM10 shows two typical pat­terns for urban atmosphere: two daily peaks correspond to morning mode and evening mode at 0700h and 1800h, respectively. The modes are divided by a mini­mum extend from 1100h to 1700h, which is assumed to the results of three factors: (i) lower emission of traffic exhaust, (ii) large availability of ozone inducing the photolysis of NOx and oxidation of CO, and (iii) enhanc­ed dipersion potential. The last effect is more pronounc­ed in the tropics when ozone levels are always higher during the noon hours and the local atmospheric circu­lation during daytime is dominated by the sea breezes, which have a key role in blowing air masses from the sea (Masiol et al., 2014).

3. 3 Weekday and Weekend Effect of PM10 in Kota Kinabalu

Fig. 9 shows the daily average PM10 concentration measured in the urban atmosphere of Kota Kinabalu in Jan 2012. In this figure, two effects are highlighted: the weekend effect and the weekday effect. The weekend effect and the weekday effect are separated by the sym­bol “M” which denotes Monday. It is emphasized the abrupt fall in PM10 concentration is often observed dur­

Variability of the PM10 Concentration in Urban Atmosphere of Sabah 119

Fig. 7. Hourly mean concentration of (a) CO, (c) NO2, (e) SO2, (g) O3 plotted on secondary axes with comparison to PM10 plot­ted on primary axes (Left panel) on Jan, 15­16 for morning effect. Scatter analysis of (b) CO, (d) NO2, (f) SO2, (h) O3 plotted against PM10

(Right panel).

0100

0400

0700

1000

1300

1600

1900

2200

0100

0400

0700

1000

1300

1600

1900

2200

0100

0400

0700

1000

1300

1600

1900

2200

0100

0400

0700

1000

1300

1600

1900

2200

0100

0400

0700

1000

1300

1600

1900

2200

0100

0400

0700

1000

1300

1600

1900

2200

0100

0400

0700

1000

1300

1600

1900

2200

0100

0400

0700

1000

1300

1600

1900

2200

Jan­15 Jan­16

Jan­15 Jan­16

Jan­15 Jan­16

Jan­15 Jan­16

200

150

100

50

0

200180160140120100806040200

200180160140120100806040200

0 0.5 1 1.5CO (ppm)

0 0.005 0.01 0.015 0.02 0.025NO2

(ppm)

0 0.001 0.002 0.003 0.004 0.005SO2

(ppm)

0 0.01 0.02 0.03 0.04O3

(ppm)

200180160140120100806040200

200180160140120100806040200

200180160140120100806040200

200180160140120100806040200

200180160140120100806040200

1.4

1.2

1

0.8

0.6

0.4

0.2

0

0.025

0.02

0.015

0.01

0.005

0

0.00450.0040.00350.0030.00250.0020.00150.0010.00050

0.04

0.035

0.03

0.025

0.02

0.015

0.01

0.005

0

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

CO

(ppm

)N

O2 (p

pm)

SO2 (p

pm)

O3 (p

pm)

R2 = 0.4629

R2 = 0.2314

R2 = 0.0922

R2 = -1.35

PM10CO

PM10NO2

PM10SO2

PM10O3

(a)

(c)

(e)

(g)

(b)

(d)

(f)

(h)

120 Asian Journal of Atmospheric Environment, Vol. 12(2), 109-126, 2018

Fig. 8. Hourly mean concentration of (a) CO, (c) NO2, (e) SO2, (g) O3 plotted on secondary axes with comparison to PM10 plot­ted on primary axes (Left panel) on Jan, 1­2 for evening effect. Scatter analysis of (b) CO, (d) NO2, (f) SO2, (h) O3 plotted against PM10

(Right panel).

2000

2200

2400

0200

0400

0600

0800

1000

1200

1400

1600

1800

2000

2200

2400

2000

2200

2400

0200

0400

0600

0800

1000

1200

1400

1600

1800

2000

2200

2400

2000

2200

2400

0200

0400

0600

0800

1000

1200

1400

1600

1800

2000

2200

2400

2000

2200

2400

0200

0400

0600

0800

1000

1200

1400

1600

1800

2000

2200

2400

Jan­01 Jan­02

Jan­01 Jan­02

Jan­01 Jan­02

Jan­01 Jan­02

200180160140120100806040200

(a)

(c)

(e)

(g)

(b)

(d)

(f)

(h)

200180160140120100806040200

200180160140120100806040200

200180160140120100806040200

200180160140120100806040200

200180160140120100806040200

200180160140120100806040200

R2 = 0.4505

R2 = 0.4187

R2 = -1.563

R2 = -2.766

0 0.2 0.4 0.6 0.8 1 1.2CO (ppm)

0 0.005 0.01 0.015 0.02NO2

(ppm)

0 0.005 0.01 0.015 0.02O3

(ppm)

0 0.0005 0.001 0.0015 0.002 0.0025SO2

(ppm)

200180160140120100806040200

1.2

1

0.8

0.6

0.4

0.2

0

0.0180.0160.0140.0120.010.0080.0060.0040.0020

0.0180.0160.0140.0120.010.0080.0060.0040.0020

0.0025

0.002

0.0015

0.001

0.0005

0

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

PM10

(μg/

m3 )

CO

(ppm

)N

O2 (p

pm)

O3 (p

pm)

SO2 (p

pm)

PM10CO

PM10NO2

PM10SO2

PM10O3

Variability of the PM10 Concentration in Urban Atmosphere of Sabah 121

ing the weekend and gradual rise is often observed dur­ing the weekday. Within the measurement period, the PM level never exceeded 100 μg/m3 over the weekend. The PM concentration was significant low <50 μg/m3 on Sunday and slightly higher on Saturday. Meanwhile, high PM10 concentration >150 μg/m3 was observed on two Mondays of the month on Jan, 2nd and 16th. A har­monic pattern was observed on the PM10 levels where the peaks are often during the first or second day of the week and the lows are during the last two days of the week.

Fig. 10 shows the average diurnal changes of PM10, CO, ozone, NO2, and SO2. The weekend­weekday dif­ferences of PM10 varied greatly at the peak hour 0700h. The highest values ~160 μg/m3 was remarked on week­day and ~40 μg/m3 on weekend (see Fig. 10a). Similar pattern is observed for CO where the highest difference is observed in the morning peak hours from 0700h to 0800h (see Fig. 10b). The increase in PM10 concentra­tion during morning hours as a result of direct road traf­fic emission is observed but this increase is more mark­ed in the cases of CO which reflects direct exhaust emis­sions. Daily cycles of traffic­related exhausts NO2 is also marked by road traffic evolution on Fig. 10d where the weekday concentration is consistently higher than that of weekend. This effect has been discussed else­where that the pollutant number concentration increases with road traffic intensity in the morning is due to the

ultrafine particle vehicle emission and formation of new particles during dilution and cooling of the vehi­cle exhaust (Pérez et al., 2010). Besides, the second peak observed during the noon hours was attributed to new formation of aerosols by photochemically induced nucleation by solar radiation intensity, and by dilution of aerosols due to the growth of the mixing layer (Rod­ríguez and Cuevas, 2007). During the night hours, the PM10 concentration started to decline (see Fig. 10a) owing to the decrease of the traffic congestion and also probably due to the thinning of the boundary layer depth that favors condensation and coagulation processes that sink the aerosol particles (Minoura and Akekawa, 2005). As a whole, during the weekday the levels of vehicle exhaust emission such as CO and NO2 follow the traf­fic intensity with maxima during the morning hours, decreasing during the day because of dilution process­es and increasing again the evening. During the night hours, the concentrations remained high owing to the reduction of the boundary layer height and probably lower wind speeds that prevent the dispersion of pollu­tants (Pérez et al., 2010).

The average diurnal ozone volume fraction is higher during the weekend compared to weekdays (Fig. 10c). This finding is similar to that report from Gvozdic et al. (2011). This behaviour is contrary to other air pollu­tants such as PM10, NO2, CO, and SO2 where the week­end concentration of traffic pollutions were lower than

350

300

250

200

150

100

50

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Time (day)

M

M

M

MM

PM10

con

cent

ratio

n (μg

m-

3 )

Fig. 9. Daily distribution of hourly PM10 concentration in Jan 2012 over the urban atmosphere of Kota Kinabalu, Sabah. Alphabet symbol M denotes Monday.

122 Asian Journal of Atmospheric Environment, Vol. 12(2), 109-126, 2018

weekday concentration (see Fig. 11). Ozone level in­creased in the afternoon during both weekdays and weekends, following the solar radiation intensity and its formation is dependent on the photochemistry pro­cess. However, during the weekends, the ozone concen­tration was higher. This is due to the effect of the lower general pollution levels observed during the weekends,

thus reducing the interaction of ozone with other spe­cies such as nitrate or carbonaceous compounds (Pérez et al., 2010). Similar results were also reported in Qin et al. (2004) that low emission of NO during the week­end could be the reason for high ozone concentration at night.

Finally, we present two types of average diurnal vari­

(a) (b)

(c) (d)

(e)

200

150

100

50

0

200

150

100

50

0

0.035

0.030

0.025

0.020

0.015

0.010

0.005

0.000

0.025

0.020

0.015

0.010

0.005

0.000

0.004

0.003

0.003

0.002

0.002

0.001

0.001

0.000

1 3 5 7 9 11 13 15 17 19 21 23Time (h)

1 3 5 7 9 11 13 15 17 19 21 23Time (h)

1 3 5 7 9 11 13 15 17 19 21 23Time (h)

1 3 5 7 9 11 13 15 17 19 21 23Time (h)

1 3 5 7 9 11 13 15 17 19 21 23Time (h)

PM10

(μg

m-

3 )

CO

(×10

-3 pp

m)

Ozo

ne (p

pm)

NO

2 (ppm

)

SO2 (p

pm)

WeekdayWeekend

WeekdayWeekend

WeekdayWeekend

WeekdayWeekend

WeekdayWeekend

Fig. 10. Average diurnal changes of (a) PM10, (b) CO, (c) Ozone, (d) NO2, (e) SO2 by the day of the week. Weekday is repre­sented by blue color line and weekend is represented by orange color line.

Variability of the PM10 Concentration in Urban Atmosphere of Sabah 123

ations during the weekday (see Fig. 11a). Firstly, the diurnal variation with a single maxima for ozone and secondly, diurnal variations with double maxima for CO, NO2, and PM10. The single maxima for ozone occurred during the noon hours from 1300h to 1400h while the double maxima for CO, NO2, and PM10 occur­

red during the morning hours and afternoon hours at 0700h and 1900h, respectively. The peak of CO, NO2, and PM10 concentration are in the front and at the back of the peak for ozone is a representative pattern of urban atmosphere in tropical climatic region (Gvozdic et al., 2011). It is corresponding to the increase of traffic den­

0.025

0.020

0.015

0.010

0.005

0.000

0.025

0.020

0.015

0.010

0.005

0.000

140

120

100

80

60

40

20

0

80

70

60

50

40

30

20

10

0

SO2,

NO

2, O

zone

(ppm

)

SO2,

NO

2, O

zone

(ppm

)

CO

(ppm

), PM

10 (μ

g/m

3 )

CO

(ppm

), PM

10 (μ

g/m

3 )

1 3 5 7 9 11 13 15 17 19 21 23Time (h)

1 3 5 7 9 11 13 15 17 19 21 23Time (h)

(a) (b)OzonePM10NO2COSO2

OzonePM10NO2COSO2

Fig. 11. Average diurnal variations of ozone, NO2, SO2, PM10, and CO during (a) weekday and (b) weekend.

14000

12000

10000

8000

6000

4000

2000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Time (h)

0.4

0.35

0.3

0.25

0.2

0.15

0.1

0.05

0

CO

/NO

2

SO2/N

O2

SO2/NO2 weekdaySO2/NO2 weekendCO/NO2 weekdaysCO/NO2 weekends

Fig. 12. Hourly distribution of SO2/NO2 ratio and CO/NO2 ratio during weekdays and weekend over the urban atmosphere of Kota Kinabalu. Clustered column represents SO2/NO2 ratio and line with marker represents CO/NO2 ratio.

124 Asian Journal of Atmospheric Environment, Vol. 12(2), 109-126, 2018

sity in the morning hours from 0600h to 0800h, and afternoon hours from 1800 to 2000h. However, this pat­tern is somewhat less reflective during the weekend (see Fig. 11b). The bimodal distribution of CO during the weekend is still legibly observable but less significant when compared to weekday. The peak of CO during the weekend was observed at 0700h and 1900h, which is similar to that of weekday. However, the peak of PM10 had no extreme variations during the weekend. Same pattern was observed for NO2 levels. The hourly aver­age PM10 concentration was constantly capped below 50 μg/m3 (see Fig. 11b) which is an indicator of good air quality. Meanwhile, the evolution pattern for SO2 during the weekdays and weekend have no much differ­ence. This indicates that the traffic­related emissions is less influential on the changes of SO2. Therefore, the sources of SO2 are somewhat not directly related to traf­fic exhausts but possibly due to the industrial emissions. This can be explained by the characteristics of NOx con­centration rationed to CO and SO2. High CO/NOx and low SO2/NOx ratios values indicate mobile sources while high SO2/NOx and low CO/NOx ratio values typi­cally indicate point sources, e.g. from industrial activi­ties (Talib et al., 2014). Fig. 12 shows that the air qual­ity level in the study area was dominantly influenced by the CO/NO2 ratio value with values ranging from 80 to 140 for weekday and 40 to 80 for weekend. For SO2/NO2 ratio, the values range from 0.10 to 0.25 for weekdays and 0.15 to 0.35 for weekend. These ratio values indicate that air pollution in the study area of Kota Kinabalu, is more seriously influenced by mobile sources rather especially during the weekday period.

4. CONCLUSIONIn the present study, the seasonal, diurnal and weekly

cycles of PM10 concentration in the urban atmosphere of Sabah was investigated. Five distinct study areas were selected to represent the urban air of Sabah, name­ly Kota Kinabalu, Labuan, Sandakan, Tawau, and Ke­ningau. The major sources of elevated PM10 over the region was high likely related to the emissions of traf­fic related gaseous such as NOx and CO. On the tem­poral distribution, PM10 levels remain highly concen­trated during the south­west monsoon (SWM) and post south­west monsoon (POST­SWM) which is the hot and dry season in Sabah. On the contrary, the PM10 lev­els were slightly lower during the north­east monsoon

(NEM) which is the cold and wet season. However, the variance of change was not extremely large but rela­tively small difference of <10 μg/m3 between season­al PM10 concentration was remarked for all study areas in Sabah.

To further scrutinize the variability of PM10 concen­tration due to the changes of CO, NO2, SO2, and ozone, Kota Kinabalu site was selected as it is the most urba­nized central city in Sabah and has the most variability among all sites. The analysis of PM10 concentration measured between Jan 2012 to Dec 2012 in Kota Kina­balu showed clear diurnal and weekly cycles. In the diurnal cycles, two peaks were observed at the morning and evening hours which is corresponding to the heavy traffic emissions. Similar pattern was also observable on the diurnal cycles of CO and NO2, which indicate that the major precursors of elevated PM10 were the traf­fic and mobile sources. The profile of SO2, however has no much significant variability throughout the day. Ozone is photochemical reactive gaseous that peaks during the noon hours but starts to decline after the absence of intense solar radiance. Besides, the correla­tive analysis further highlights the positive proportion­ality between PM10 and CO, PM10 and NO2, while neg­ative relationship between PM10 and ozone.

In the weekly cycles, higher PM10, CO, and NO2 con­centrations were observed during the weekday when compared to weekend. The ozone concentration was somehow higher during the weekend especially during afternoon hours. It is owing to the lower general pol­lution levels observed during the weekends reducing the interaction of ozone with other species such as nit­rate or carbonaceous compounds. Besides, two types of average diurnal variations were highlighted during the week day: (a) diurnal variation with a single maxi­ma for ozone and (b) diurnal variations with double maxima for CO, NO2, and PM10. The peak of CO, NO2, and PM10 concentration are in the front and at the back of the peak for ozone is a representative pattern of urban atmosphere in this climatic region. This pattern is less reflective during the weekend except for the unimodal distribution of ozone during noon hours. The PM10 and NO2 levels have no extreme variations during the week­end except for the bimodal distribution CO was still legibly observable but less significant. The evolution pattern for SO2 during weekdays and weekend have no much difference. Finally, the characteristics of NO2 concentration rationed to CO and SO2 suggests that air pollution in Kota Kinabalu is more seriously influenc­ed by mobile sources rather especially during weekday period.

ACKNOWLEDGEMENTThis research was supported by the Malaysian Min­

istry of Education under the research grant number SBK0352­2017 and was greatly acknowledged.

Variability of the PM10 Concentration in Urban Atmosphere of Sabah 125

REFERENCESAdame, J.A., Hernández­ceballos, M.Á., Sorribas, M.,

Lozano, A., Arturo, B., Morena, D. (2014) Weekend­weekday effect assessment for O3, NOx, CO and PM10 in Andalusia, Spain (2003­2008). Aerosol and Air Qual­ity Research 14, 1862­1874.

Aouizerats, B., Werf, G.R. Van Der, Balasubramanian, R., Betha, R. (2015) Importance of transboundary trans­port of biomass burning emissions to regional air qual­ity in Southeast Asia during a high fire event. Atmo­spheric Chemistry and Physics 15, 363­373.

Azid, A., Juahir, H., Toriman, M.E., Kamarudin, M.K.A., Saudi, A.S.M., Hasnam, C.N.C., Aziz, N.A.A., Aza­man, F., Latif, M.T., Zainuddin, S.F.M., Osman, M.R., Yamin, M. (2014) Prediction of the level of air pollu­tion using principal component analysis and artificial neural network techniques: a case study in Malaysia. Water, Air, & Soil Pollution 225(8), 2063. Available at: http://link.springer.com/10.1007/s11270­014­2063­1.

Azmi, S.Z., Latif, M.T., Ismail, A.S., Juneng, L., Jemain, A.A. (2010) Trend and status of air quality at three dif­ferent monitoring stations in the Klang Valley, Malay­sia. Air Quality, Atmosphere and Health 3(1), 53­64.

Barmpadimos, I., Hueglin, C., Keller, J., Henne, S., Pre­vot, A.S.H. (2011) Influence of meteorology on PM10 trends and variability in Switzerland from 1991 to 2008. Atmospheric Chemistry and Physics 11, 1813­1835.

Bathmanabhan, S., Nagendra, S., Madanayak, S. (2010) Analysis and interpretation of particulate matter ­ PM10, PM2.5 and PM1 emissions from the heterogeneous traf­fic near an urban roadway. Atmospheric Pollution Re­search 1, 184­194. Available at: http://dx.doi.org/ 10.5094/APR.2010.024.

Czernecki, B., Półrolniczak, M., Kolendowicz, L., Marosz, M., Kendzierski, S., Pilguj, N. (2017) Influence of the atmospheric conditions on PM10 concentrations in Poz­nań, Poland. Journal of Atmospheric Chemistry 74, 115­139.

Dominick, D., Talib, M., Zain, S.M., Zaharin, A. (2012) Spatial assessment of air quality patterns in Malaysia using multivariate analysis. Atmospheric Environment 60, 172­181.

Gomiscek, B., Hauck, H., Stopper, S., Preining, O. (2004) Spatial and temporal variations of PM1, PM2.5, PM10 and particle number concentration during the AUPHEP ­

project. Atmospheric Environment 38, 3917­3934.Gvozdic, V., Kovac­Andric, E., Brana, J. (2011) Influence

of meteorological factors NO2, SO2, CO and PM10 on the concentration of O3 in the urban atmosphere of east­ern Croatia. Environmental Modeling and Assessment 16(5), 491­501.

Harrison, M.E., Page, S.E., Limin, S.H. (2009) The global impact of Indonesian forest fires. Biologist 56(3), 156­163.

Kai, Z., Yuesi, W., Tianxue, W., Yousef, M., Frank, M. (2007) Properties of nitrate, sulfate and ammonium in

typical polluted atmospheric aerosols (PM10) in Bei­jing. Atmospheric Research 84, 67­77.

Lelieveld, J., Evans, J.S., Fnais, M., Giannadaki, D., Pozzer, A. (2015) The contribution of outdoor air pol­lution sources to premature mortality on a global scale. Nature 525(7569), 367­371.

Masiol, M., Agostinelli, C., Formenton, G., Tarabotti, E., Pavoni, B. (2014) Thirteen years of air pollution hourly monitoring in a large city: Potential sources, trends, cycles and effects of car­free days. Science of the Total Environment 494­495, 84­96. Available at: http://dx. doi.org/10.1016/j.scitotenv.2014.06.122.

Minoura, H., Akekawa, H. (2005) Observation of number concentrations of atmospheric aerosols and analysis of nanoparticle behavior at an urban background area in Japan. Atmospheric Environment 39, 5806­5816.

Nirmalkar, J., Deb, M.K. (2015). Impact of intense field burning episode on aerosol mass loading and its possi­ble health implications in rural area of eastern central India. Air Quality, Atmosphere and Health. Available at: http://doi.org/10.1007/s11869­015­0330­y.

Norsukhairin, S., Abdul, S., Azid, A., Sharif, S.M., Latif, T., Aris, Z., Zain, S.M., Dominick, D. (2013) Spatial and temporal air quality pattern recognition using envi­ronmetric techniques: a case study in Malaysia. Envi­ronmental Science Processes & Impacts 15, 1717­1728.

Othman, J., Sahani, M., Mahmud, M., Sheikh, K. (2014) Transboundary smoke haze pollution in Malaysia: Inpatient health impacts and economic valuation. Envi­ronmental Pollution 189, 194­201. Available at: http://dx.doi.org/10.1016/j.envpol.2014.03.010.

Pérez, N., Pey, J., Cusack, M., Reche, C., Querol, X., Alastuey, A., Viana, M. (2010) Variability of particle number, black carbon, and PM10, PM2.5, and PM1 lev­els and speciation: Influence of road traffic emissions on urban air quality. Aerosol Science and Technology 44, 487­499.

Qin, Y., Tonnesen, G.S., Wang, Z. (2004) Weekend/week­day differences of ozone, NOx, Co, VOCs, PM10 and the light scatter during ozone season in southern Cali­fornia. Atmospheric Environment 38, 3069­3087.

Rahman, S.R.A., Ismail, S.N.S., Ramli, M.F., Latif, M.T., Abidin, E.Z., Praveena, S.M. (2015) The assessment of ambient air pollution trend in Klang Valley, Malay­sia. World Environment 5(1), 1­11.

Rodríguez, S., Cuevas, E. (2007) The contributions of “minimum primary emissions” and “new particle for­mation enhancements” to the particle number concentra­tion in urban air. Journal of Aerosol Science 38, 1207­1219.

Sansuddin, N., Azam, N., Ahmad, R., Yahaya, S., Faizah, N., Yusof, F., Adyani, N. (2011) Statistical analysis of PM10 concentrations at different locations in Malaysia. Environmental Monitoring and Assessment 180, 573­588.

Shaadan, N., Jemain, A.A., Latif, M.T., Deni, S.M. (2015) Anomaly detection and assessment of PM10 functional data at several locations in the Klang Valley, Malaysia.

126 Asian Journal of Atmospheric Environment, Vol. 12(2), 109-126, 2018

Atmospheric Pollution Rese arch 6(2), 365­375. Avail­able at: http://dx.doi.org/10.5094/APR.2015.040.

Talib, M., Dominick, D., Ahamad, F., Khan, F., Juneng, L., Hamzah, F.M., Nadzir, M.S.M. (2014) Long term assessment of air quality from a background station on the Malaysian Peninsu la. Science of the Total Environ­ment 482­483(2), 336­348. Available at: http://dx.doi.org/10.1016/j.scitotenv.2014.02.132.

Thurston, G.D., Burnett, R.T., Turner, M.C., Shi, Y., Krewski, D., Lall, R., Ito, K., Jerrett, M., Gapstur, S.M., Diver, W.R. Iii, C.A.P. (2016) Ischemic heart disease mortality and long­term exposure to source­related components of U.S. fine particle air pollution. Environmental Health Prespectives 124(6), 785­795.

Wang, Y., Zhang, Q.Q., He, K., Zhang, Q., Chai, L., Wang, C.Y. (2015) Sulfate­nitrate­ammonium aerosols over China: response to 2000­2015 emission changes of sulfur dioxide, nitrogen oxides. Atmospheric Chem­

istry and Physics 13, 2635­2652.Wrobel, A., Rokita, E., Maenhaut, W. (2000) Transport of

traffic­related aerosols in urban areas. The Science of the Total Environment 257, 199­211.

Yara, C., Viegas, M., Lemos, P., Cavalcanti, B., Pais, R., Oliveira, E. De, Gil, R., Neto, D.A. (2017) Atmospher­ic distribution of organic compounds from urban areas near Olympic games sites in Rio de Janeiro, Brazil. Microchemical Journal 133, 638­644. Available at: http://dx.doi.org/10.1016/j.microc.2017.04.027.

Yoo, J., Jeong, M., Kim, D., Stockwell, W.R., Yang, J., Shin, H., Lee, M.I., Song, C.K., Lee, S. (2015) Spatio­temporal variations of air pollutants (O3, NO2, SO2, CO, PM10, and VOCs) with land­use types. Atmo­spheric Chemistry and Phy sics 15, 10857­10885.

(Received 30 November 2017, revised 10 January 2018, accepted 4 February 2018)