Keywords

1 Introduction

Air pollutants measured in Malaysia Ambient Air Quality Guideline are Sulphur dioxide (SO2), Nitrogen Dioxide (NO2), Ground Level Ozone (O3), Carbon Monoxide (CO), and Particulate Matter [with an aerodynamic diameter less than 10 μ (PM10) and with an aerodynamic diameter less than 2.5 μ (PM2.5)] [11]. Air Pollutant Index (API) is established to provide easy understandable information about air pollution for the public. Hourly API readings are published to the public to enable the citizen to be informed and prepare themselves if the air quality worsens [12]. Afroz et al. [1] suggest that primary sources of air pollutants in Malaysia are from land vehicles, operating industrial, and biomass burning, whereas PM10 and NO2 are the primary pollutants.

In Malaysia, updated in 2019, there are 68 monitoring stations [10] designated to monitor and record the air quality continuously for 24 h per day. Malaysia has suffered from numerous particulate pollutions since 1994. Effect of these haze episodes are not only limited to human health but as well as towards economy, plants and vegetation [19]. Therefore, it’s important to study the behaviour of each air pollutant.

Air pollution is affected by seasonal monsoons [2, 16] and meteorological [4,5,6,7]. PM10 is mainly released from the combustion of agricultural activities and forest biomasses burning [8, 9]. NO2 is the production of burning fuel in land vehicles and factories waste gaseous [13]. Meanwhile, O3 formed from both natural and anthropogenic sources contribute to the high level of ground-level O3 concentration, and the spatial variation of emission sources has been widely studied [22].

PM10 and O3 are air pollutants that are under most attention from many types of research as well as NO2 that act as a precursor of ground-level O3. Photochemical reaction of ground-level O3 is greatly affected by the presence of UV light and its precursors. However, the effect of the existence of PM10 usually has been ignored because of the inactivity of both air pollutants. This study is giving focus on the screening effect of PM10 towards the formation of O3 as well as towards the concentration of other pollutants. The objectives of this study are to study the behaviour of air pollutants during the non-high particulate events (non-HPEs) and high particulate events (HPEs), to study the diurnal variation of air pollutants and to establish the influence of PM10 towards concentration of O3 and NO2.

2 Methodology

2.1 Site Location

Air pollution data at Taiping monitoring station (4° 53′ 56.5″ N 100° 40′ 44.3″ E) of 2014 has been studied based on daytime/nighttime, seasonal monsoons and HPE conditions. Taiping monitoring station is categorized as an industrial area. The main contributors of O3 precursors are vehicle emission from expressway especially during school hours. Average ambient temperatures of Taiping are around 29 °C, and annual rainfall intensity is 537.3 mm with a relative humidity approximately 85%.

2.2 Statistical Analysis

The research focuses on data collection and analysis of the concentration of O3, NO2, and PM10 to study their behavior based on each criterion and condition. Department of Environment Malaysia has provided the three types of air pollutants data, which are PM10, NO2, and O3 at Taiping monitoring station during 2014 to carry out the objectives mentioned above. HPE incidence in Taiping gives higher value during 2014 and Taiping station was chosen as the wettest are across peninsular Malaysia. The data analysis looks into the effect of seasonal monsoons, high particulate event, daytime and nighttime. Then the hourly concentration trends of air pollutants were compared and contrasted using descriptive analysis and graphical analysis. Correlation coefficient will be used to analyse the relationship between PM10 with O3 and O3 with NO2. Figure 1 shows the flow of this study. The analysis used 8760 data of hourly mean concentration of PM10, O3 and NO2.

Fig. 1
figure 1

Flow chart of this study methodology

Full size image

Seasonal monsoons were categorized according to the months [3] as shown in Table 1. Early November until late March is Northeast monsoon termed as NE. Early April until late May is the first inter-monsoon season termed as AM. Early June until late September is Southwest monsoon which termed as SW. Late September until late October is the second inter-monsoon season which term as SO [16]. Daytime (DT) and nighttime (NT) are categorized by time duration for DT are 7.00 a.m. until 7.00 p.m. while NT from 7.00 p.m. till 7.00 a.m.

Table 1 Monsoonal category
Full size table

The high particulate event (HPE) was declared when API value more than 100 for more than 72 h while second parts used the definition of HPE that stated duration whereby the concentration of PM10 is more than 150 µg/m3 for more than 72 h [17]. This study adopted the New MAAQS for the concentration of PM10, O3, and NO2 as a guideline for data analysis, as shown in Table 2. According to the new Malaysia Ambient Air Quality Standards (MAAQS), the standard value for 24 h basis of PM10 is 100 µg/m3 and 90 ppb for the hourly basis of O3 concentration. For NO2 concentration, the guideline given a limit of 70 µg/m3, which is 36 ppb for 24-h basis. Meanwhile, for O3, the guideline given a limit of 100 µg/m3 for 8-h basis [11].

Table 2 New Malaysia ambient air quality standards 2015 [11]
Full size table

Data analysis of air pollutants is using time-series plot, descriptive analysis, diurnal plot and Pearson correlation coefficient. Time series plot is a graph of data against time whereby the time is the horizontal axis and data recorded as variables is the vertical axis. Diurnal plots have similar axis unit as time series plot. The difference is the horizontal axis represents as daily 24-h average instead of monthly hours average. Box and whisker plot is part of the descriptive analysis whereby comparison of mean and quartile value between different sets of results can be easily made. For correlation coefficients, Eq. (1) is used to calculate the R-value.

$$ \text{Correlation}\;\text{coefficient,}\;\text{p}_{{\text{xy}}} \text{ = }\frac{{\text{cov}\left( {\text{x,y}} \right)}}{{\upsigma_{\text{x}}\upsigma_{\text{y}} }} $$
(1)

  • where

  • cov (x, y): covariance of two variable x and y

  • \( \sigma_{x} \sigma_{y} \): product of standard deviation of two variables.

Pearson Correlation coefficient takes no units and always have a value between −1.0 to +1.0. In the correlation coefficient, zero value indicates that both variables do not react with each other. The positive and negative signs of the result determine the type of interaction.

3 Results and Discussion

3.1 Variations of PM10, O3, and NO2

Descriptive analysis of PM10, O3, and NO2 of Taiping in 2014 are illustrated using a box and whisker plot in Fig. 2. This analysis has been carried out for three pollutants namely PM10, O3 and NO2 during HPE, difference of seasonal variation, daytime and nighttime variation.

Fig. 2
figure 2

Box and whisker plot of PM10, O3 and NO2 (light blue: NE monsoon; Red: inter-monsoon (AM and SO); Yellow: SW monsoon, Red-Line: HPE)

Full size image

For PM10 concentration, highest mean occurred during HPE (85.07 µg/m3). The lowest mean PM10 concentration happened during AM, which is 29.33 µg/m3 (April) and 24.09 µg/m3 (May). SW and SO have an almost consistent mean value of 52.02 µg/m3 (June), 52.53 µg/m3 (August), 52.53 µg/m3 (September), and 53.20 µg/m3 (October). The highest mean O3 concentration occurred during August (24.13 ppb), then followed by SO (22.33 ppb). The lowest mean concentration shares the same trait as PM10 whereby April (5.36 ppb) and May (8.83 ppb), AM have the lowest concentration. Results from NO2 concentration showed that the highest mean concentration is also during HPE (10.62 ppb). However, the lowest mean concentration is during September with only 5.01 ppb. Inter-monsoon AM having 6.61 ppb (April) and 7.93 ppb (May) while SO have 8.29 ppb. NE and SW monsoon has an average concentration ranging from 8.41 ppb (November) to 10.24 ppb (June).

Table 3 shows that from the results, PM10, O3 and NO2 recorded the highest mean concentration in July, because the effects of HPE occurrences happen during that month. The overall view, O3 has most fluctuated concentration from DT to NT similar as found by Tong et al. [20]. The percentage difference is ranging from 46.06% (October) to the highest of 70.37% (February). Then, NO2 also have high percentage difference of 24.78% (April) to 58.93% (July) from DT to NT. PM10 percentage difference has a value of less than 10% (March, June, July, August and December), more than 10% but less than 30% (May, October and November) and rest of the months are more than 30% difference.

Table 3 Mean average concentrations of PM10, O3 and NO2 for DT and NT variation
Full size table

3.2 Seasonal Variations of PM10, O3, and NO2

The diurnal plot revealed the behaviour of the concentration of air pollutants. In the morning where human activities are low, all three pollutants have a slightly smooth curve until daytime. Since 8.00 a.m., the concentration of O3 begins to fluctuate until reaching its first peak at around 11.00 a.m. whereas the concentration of PM10 increase as well. However, the concentration of NO2 remains low during the daytime. This condition can be explained from the photochemical reaction of O3 where the formation of NO can react with VOCs to form the stable NO2. Therefore, the breaking and reforming of the NO2 molecule do not affect much on the concentration. After the O3 achieve its second peak near dusk, the UV intensity started to drop as the loss of sunlight for the transformation of O3 [21]. However, this promotes the increase in the concentration of NO2 to reach its peak at night.

Figure 3 illustrated that the maximum PM10 concentration occurred during SO with 80.20 µg/m3 and second is during NE with 79.75 µg/m3. The lowest PM10 mean value occurs during AM at 46.39 µg/m3. For SE, the PM10 maximum mean value is 63.14 µg/m3. Photochemical reaction of formation of O3 required sunlight [18].

Fig. 3
figure 3

Diurnal plot of PM10, O3, and NO2 under the influence of seasonal monsoon and HPE

Full size image

In Malaysia, sunrise usually occurred after 7.00 a.m. This trait can be comparable to the daily plot of O3 concentration as the increase of O3 concentration started only after 8.00 a.m. O3 reached its maximum concentration at around 2.00 p.m. to 3.00 p.m. for all seasons except SO, which is at 1.00 p.m. At near dusk (5.00 p.m. to 7.00 p.m.), the sunlight got weaker eventually. Therefore, O3 concentration started to fall. From an overview of the diurnal plot of O3, SO has the highest trend, followed by NE and SW. AM has the lowest amongst all the seasons. NO2 have a remarkable crest and trough pattern. Two crests occurred from 8.00 a.m. to 10.00 a.m. and 8.00 p.m. to 12.00 a.m. Two troughs occurred from 4.00 a.m. to 7.00 a.m. and 12.00 p.m. to 5.00 p.m. However, the trend was almost uniform among the monsoon seasons. This was because no season shows domination of the highest value at all in 24 h. In the morning, SE has the highest NO2 concentration. In the afternoon, the diurnal plot shows that all seasons are having almost similar NO2 level of 4.00–8.00 ppb (12.00 p.m. to 5.00 p.m.).

For nighttime, the concentration of NO2 reaches a peak around 9.00 p.m.to 11.00 p.m. (12–16 ppb). The morning peaks were due to the increase in traffic volume in the morning. However, in the presence of sunlight, the concentration of NO2 falls for the transformation of O3 to occur. The condition remained until the night where the intensity of the sun lowered, the formation of NO2 from NO followed by the destruction of O3.

HPE shows a different trend compare to other seasons. For PM10, two peaks were observed during HPE, which is at noon (286 µg/m3) and 7.00 p.m. (242.33 µg/m3). The value is very high compared to the limit set by MAAQS. The situation is similar for O3 whereby average O3 concentration has two peaks at noon (59.0 ppb) and 6.00 p.m. (66.0 ppb). NO2 concentration slightly differs from diurnal plot from seasonal variation. The average concentration of NO2 about 13.0–19.0 ppb (1.00 a.m. to 10 a.m.) and fall after 10 a.m. to slightly more than 5.0 ppb. However, the progressions of NO2 showed a step-growth at 5.00 p.m. and maintain around 14.0 ppb. After 7.00 p.m., the concentrations increase gradually and achieve maximum value at 10.00 p.m. (29.0 ppb).

Besides, an observation of double-peaks can be noticed from the concentration of O3. The trend of double-peaks happened where the first peak occurred during the morning, and the second peak occurred during the afternoon. It also shows that the first peak is lower than the second peak. According to Liu and Liu [14], the first peak occurred due to the vertical or horizontal transport of O3 in the morning as well as due to a lower concentration of NO. The higher second peak is due to the increase in accumulation of ground-level O3 due to the photochemical reactions.

3.3 Correlation Coefficient of PM10, O3, and NO2

Table 4 shows the correlation coefficient (R-value) of PM10 and O3 have ranged from +0.40 to +0.88, which is showing positive result except for February is −0.13 and this finding is similar to that found by Awang et al. [4, 5]. The correlation coefficient of O3 and its precursors NO2 shows a negative trend which varies from −0.19 to −0.75. All the results were significant at p < 0.05.

Table 4 Pearson correlation coefficient of PM10 with O3 and NO2 with O3 (p < 0.05)
Full size table

As NO2 is the O3 precursor for the photochemical transformation of O3, the correlation coefficient of these two variables provide promising negative results that change in one pollutant gives a different influence on the concentration of the other pollutants. Results of the correlation coefficient showed that the R-value decreased from ‘during non-HPE’ to ‘during HPE’. The photochemical formation of O3 results in breaking a molecule of NO2 for the completion in DT. Besides, during NT in the absence of sunlight, NO2 does not form O3. Therefore, a negative correlation coefficient correctly proved the condition whereby the decrease of O3 comes together with an increase in NO2 concentration and vice versa [15]. Thus, O3 concentration during HPE significantly depends on precursor concentrations because an increment in precursors will promote high production in O3 concentration. The unfavourable trends of O3 during HPE can possibly lead to increased health risks that require immediate mitigation and prevention plans.

4 Conclusions

The study looks at the fluctuation in concentrations variabilities of PM10 with O3 and its precursors in 2014 because HPE occurred. Both gaseous maximum values that were observed do not exceed the recommended value in MAAQS. During HPE, the maximum concentration of PM10 achieved a very unhealthy level of 346 µg/m3. Analysis of daytime and nighttime results is significant for O3 and NO2. O3 have very high mean concentration during DT and become very low during NT. The situation is directly opposite of NO2 whereby concentration during DT is lower than during NT. This condition is supported by the Pearson correlation coefficient between NO2 and O3 that shown the negative value which indicates that the changes occurred in one pollutant will result in the opposite result of the other pollutants. Therefore, NO2 has a direct negative effect on the concentration of O3. These observations provide useful guidance that increase of PM10 concentration will increase in O3 concentration. During the HPE period, screening effect of PM10 has positively correlated with O3, but during non-HPE the correlation coefficient has decreased. Therefore, in nutshell, during HPE, a high concentration of PM10 induced high formation of O3 compared to during non-HPE.