Ozone, Nitrogen Dioxide, and PM2.5 Measurement at Three Urban Parks in Manila, Philippines using Portable Sensors

Article history: Received: 29 July, 2020 Accepted: 18 November, 2020 Online: 14 December, 2020 The presence of urban parks and green spaces in the city can play a role in maintaining urban residents’ quality of life. Parks in Manila are located near main thoroughfares. Since people usually go to parks for recreation and relaxation it is important to have an idea of the concentration of criteria pollutants at these parks as they have great effects on people’s health. Using portable sensors, one-hour average ambient concentration of O3, NO2, and PM2.5 was measured once a month between 8 am to 11 am local time from May 2018 to April 2019 at three popular urban parks in Manila, namely, Arroceros Forest Park, Rizal or Luneta Park, and Manila Zoo. Mean concentrations of O3 and NO2 are highest in Luneta Park at 0.071 ppm and 0.032 ppm, respectively. In the case of PM2.5, the mean concentrations at Luneta and Arroceros Park are the same at 0.070 mg/m. Manila Zoo had the lowest mean concentrations of the three criteria pollutants among the three urban parks. In terms of proximity to major thoroughfares, Luneta is closer and surrounded by four busy thoroughfares compared to Manila Zoo, and Arroceros Forest Park. It was also observed that measured concentrations of the three criteria pollutants were relatively higher during the cool dry months from October to February, with January 2019 recording the highest concentrations. Compared to a background site in Halang, Batangas, results of the measurement showed that the mean concentrations of O3, NO2, and PM2.5 at the three urban parks are higher demonstrating the effect of proximity of the parks to vehicular traffic on the air quality inside the parks.


Introduction
According to the State of Global Air Report 2019, air pollution is the fifth leading risk factor for mortality worldwide with the Philippines ranking number 10 in terms of the highest mortality burden attributable to air pollution [1]. Exposure to short-term and long-term air pollution is associated with respiratory and cardiovascular diseases [2]. In highly urbanized places like the National Capital Region (NCR) in the Philippines, popularly known as Metro Manila, air pollution aside from urban heating is one of the region's major challenges [3]. In the latest National Emissions Inventory by source conducted in 2015, 88% of air pollution in Metro Manila comes from mobile sources, 10% from stationary sources, and a mere 2% from area sources. The increase in the number of vehicles is attributable to the increase in population to cater to the transport needs of the people [4]. In Metro Manila alone, there was an average increase of 7.77% in vehicular registration the year 2016-2018 (NCA, 2018). One of the 16 highly urbanized cities (HUCs) in Metro Manila is the City of Manila, the capital of the Philippines. It was the most densely populated among the HUCs with 71,263 persons per square kilometer in 2015 [5]. Criteria pollutants such as Nitrogen Dioxide   Halang  13  17  15  19  23  28  30  16  27  25  17  14  Arroceros  17  27  19  24  27  22  21  14  31  27  20  29  Luneta  18  25  30  29  26  24  23  13  30  28  21  11  Mla Zoo  25  16  31  30  25  23  22  12  24  26  27  30 (NO2), Ozone (O3), and fine particulate matter (PM2.5) mostly come from mobile sources [4].
The increasing population and vehicles make Manila an example of unsustainable urbanization [6]. Hence, the presence of urban parks and green spaces in the city can play a role in maintaining urban residents' quality of life. Urban green space, a category of land cover that includes public parks and other (public or private) vegetated areas in densely populated places plays a great role in improving not only human health and but also the air quality [7][8][9][10][11]. To date, there were no available data on the level of air pollution in urban parks in Manila. People come to public parks to exercise, sometimes have family picnics, and just for relaxation. Some urban poor and street-dwellers make parks as their temporary home. Measurement of air pollution level is important especially if the parks are located or surrounded by main thoroughfares where traffic congestions are always present. These can increase the potential of park users to be exposed to pollutants associated with vehicular emissions. Lam et al made a study in Hongkong urban parks, and they have shown that the air quality in urban parks is better compared to the roadside but not significantly different from that in the ambient conditions [12]. Three of the popular parks in Metro Manila are located within the City of Manila. These are Manila Zoo, Luneta or Rizal Park, and Arroceros Forest Park. These three parks are located near the main thoroughfares, hence the proximity to vehicular traffic, which is the main source of air pollution in the city. There have been no air quality measurements in these parks as most static reference standard instruments are usually positioned near the roadside. The use of low-cost and portable air quality monitoring systems has become very popular as there is a need to provide a more accurate assessment of human air pollution exposure assessments [13]. In this study, the ambient concentrations of criteria pollutants such as nitrogen dioxide (NO2), fine particulate (PM2.5), and ozone (O3) at these three public parks were measured once a month for one year using portable sensors. The measured ambient concentrations at the three parks are then compared to a background site in Halang, Batangas which is far from any vehicular traffic. The comparison between the urban parks and the background site will demonstrate the effect of vehicular traffic on the ambient concentrations of PM2.5, NO2, and O3. Furthermore, this study will explore the usefulness of portable sensors to provide real-time determination and assessment of air quality in public areas so that people will know the quality of the air they are breathing and, once the air quality drops to unhealthy levels, people can be requested to vacate the area. This will also provide information to authorities if there is a need to place static reference standard air quality monitoring system in such public areas.

Sampling Sites
The study was conducted at the three urban parks located in the city of Manila. These are Arroceros Forest Park, Luneta or Rizal Park, and Manila Zoo. For comparison, a background sampling site was included, and this was located in Halang, Batangas which is about 70 km south of Manila. The location and addresses of the sampling sites are given in Table 1 and Figure1 shows the relative location of each sampling site. Yellow and white lines indicate roads for vehicular traffic. Permits were secured with the park administrators for us to conduct the measurements, which should be done only during office hours. Sampling Site 1 is the Arroceros Forest Park (AFP), also called as the "Last Lung of Manila". It is a 2.2-hectare (5.4-acre) manmade urban forest along Pasig River, at the foot of Quezon Bridge. The second sampling site is Rizal Park also known as Luneta Park or simply Luneta is a historical urban park in the Philippines and one of the major tourist attractions of Manila. It is approximately 16.24 hectares (40.01) acre and located along Roxas Boulevard, Manila, and also adjacent to Taft Avenue, Manila. The last urban park is the Manila Zoological and Botanical Garden or Manila Zoo. It is a 5.5-hectare (14-acre) zoo located in Malate, Manila, Philippines.

Measurement
The criteria pollutants measured in this study are PM2.5, NO2, and O3. A DustTrak™DRX aerosol monitor (Model 8533, TSI incorporated) that used a light scattering technique to infer the mass concentration of particles was used for PM2.5 measurement [14]. It was set on a 1-s time resolution at 3L/minutes. Before the start of each measurement, zero calibration was performed using a TSI 800663 zero filter. NO2 and O3 concentrations were measured by a real-time portable battery-operated gas sensor monitor AEROQUAL Series 500 with NO2 and O3 sensor heads. The sensor heads used gas-sensitive electrochemical (GSE) technology where it measures the concentration of a target gas by using oxidation or reduction reactions to generate a positive or negative current flow through an external circuit and the magnitude of this current is proportional to the gas concentration. The operating parameters of the sensor heads are as follows; O3 (detection range 0-10 ppm; resolution and minimum detection limits 0.01 ppm; response time 60 secs; temperature 0 -40 o C; relative humidity 15 -90%), NO2 (detection range 0-1 ppm; resolution 0.001 ppm; minimum detection limits 0.005 ppm; response time 30 secs; temperature 0 -40 o C; relative humidity 15 -90%). A 1-min time resolution was used during the measurement. Although the DustTrak and Aeroqual are not regulatory-grade monitors they are widely used in prior air quality research studies [13,[15][16][17][18][19][20][21][22][23][24]. Measurements done for this paper were performed during the warranty period of the instruments. Being under warranty, instrument calibration and performance was within factory specifications. During the measurements, researchers observed the performance of the instruments, and temperature and humidity on the site were within their specified operating conditions. Simultaneous measurement on all sampling sites was not possible because only one set of instruments was available. Background measurement in Halang, Batangas was done on a Sunday, and the measurements at the three parks in Manila were conducted on the following weekdays if weather permits. The sampling dates are summarized in Table 2. To observe the same atmospheric and vehicular traffic conditions in the site, measurements were done on successive days in the morning between 8 am to 11 am local time for one hour, once a month for each sampling site for one year. Measurement was not conducted during bad weather conditions and sampling time was chosen to also avoid the effect of sunlight on O3 concentration.

Results
One Way Analysis of Variance (ANOVA) was the statistical method used for identifying if there are any significant differences in the measured concentration of a particular criteria pollutant at the different sampling sites. If there's a significant difference, a Tukey honestly significant difference (HSD) test is then carried out to identify what sampling locations are significantly different, and then based on the mean concentration, the locations are ranked from highest to lowest concentration. RealStat (http://www.realstatistics.com/) Microsoft Excel Add-In was used to carry out the statistical calculation. An alpha value of 0.01 was used. To illustrate the variation of air pollution concentrations at each sampling site, Box and Whisker plots were used in terms of the lower quartile, upper quartile, median, mean, minimum, and maximum in each of the four study locations.

Ozone (O3) Measurement
Shown in Figure 2 is the box plot of the O3 concentrations from the four different locations from May 2018 to April 2019. Table  3a shows the result of the statistical analysis for O3. In all the sampling months, there was a significant difference in O3 concentration between each sampling site. The O3 concentrations at the background site always have the lowest concentration on all sampling months.
The DENR EMB National Ambient Air Quality Guideline Values (NAAQGV) for 1-Hour Averaging is 0.07 ppm [4], [25]. As shown in Figure 3, in some months, the measured O3 concentration was 0 ppm at the background site Table 3: One way Anova P-value between each sampling sites per month for (a) Ozone, (b) Nitrogen Dioxide, (c) PM2.5.

Nitrogen Dioxide (NO2) Measurement
Shown in Figure 4 is the box plot of the NO2 concentration from the four different locations from May 2018 to April 2019. Table 3b shows the result of the statistical analysis. In all the sampling months, there was a significant difference in NO2 concentration between each sampling site. The NO2 concentrations at the background site have the lowest concentration on all sampling months. On some months, the NO2 concentration was 0 ppm indicating that the NO2 concentration was below the detection limit of the NO2 sensor. This was true most especially in the case of the background site where the lower quartile value was equal to the minimum value measured by the NO2 sensor. For the whole year, as shown in Figure 5, there is a significant difference in all sampling locations. Luneta having the highest concentration of 0.032 ppm and Arroceros is a close second at 0.025 ppm, followed by Manila Zoo at 0.022 ppm and the background site at 0.017 ppm.. The National guideline value for NO2 is 150 µg/Ncm (0.08 ppm) maximum for a 24-hour exposure [25]. As in the case of O3, the coldest month of January and February 2019 registered higher NO2 concentrations for all the urban parks except the background site.

Particulate Matter (PM2.5) Measurement
For Particulate Matter measurement, TSI DUSTTRAK DRX Aerosol Monitor 8533 was used and this can measure PM10, PM2.5, and PM1.0 but only PM2.5 will be presented. For statistical analysis, only PM2.5 measurement will be used since all the other PM sizes follow the same trend. Furthermore, in terms of Air Quality Guidelines, PM2.5 is the one that is reported. No air quality guidelines yet for PM1.0. Shown in Figure 6 are the box plots of the PM2.5 concentration measured by the DUSTTRAK. The lowest concentration of PM is in Halang, Batangas while Arroceros always gives the highest concentration. The result of One-way ANOVA indicates a significant difference in the PM concentration for all months between each sampling location. Arroceros always ranks the highest while Halang is the lowest. The WHO Air Quality Guideline values for PM2.5 is 10 μg/m 3 annual mean and 0.025 mg/m 3 24-hour mean [4]. NAAQGV are 0.025 mg/m 3 annual mean and 0.050 mg/m 3 24-hour mean [4]. It can be seen from Figure 7 that the PM2.5 concentration measured from TSI DUSTTRAK at the sampling locations in Manila exceeds the annual mean guideline values. However, please note that the instrument is not the reference instrument. Like O3 and NO2, it was also observed that in January 2019, high values of PM2.5 concentration were measured at the urban parks exceeding the 24-hour mean NAAQGV.

Conclusion
With the unsustainable urbanization in Manila, the presence of urban parks can provide residence with quality of life. However, the location of these parks affects the air quality within the park. The measured values of O3, NO2, and PM2.5 at the three well known urban parks in Manila are higher compared with a background site in Halang, Batangas. Using portable sensors, the Sampling Location ambient concentration of O3, NO2, and PM2.5 was measured once a month between 8 am to 11 am local time from May 2018 to April 2019 at three known urban parks in Manila, namely, Arroceros Forest Park, Rizal or Luneta Park, and Manila Zoo. Mean concentrations of O3 and NO2 are highest in Luneta Park at 0.071 ppm and 0.032 ppm, respectively. Among the three parks, Luneta is the one surrounded by major roads. For mean concentration of PM2.5, Luneta, and Arroceros Park are the same at 0.070 mg/m 3 . One of the main reasons is the proximity of the three parks to vehicular traffic. It was also observed that measured concentrations of the three criteria pollutants were relatively higher during the cool dry months from October to February. Although the instruments used were not the reference instruments, the result provides information on the necessity of monitoring the air quality within the parks. Many residents most especially people who cannot afford, and don't have time to go out of the city need these urban parks for relaxation and recreational activities.