Investigating a Potential Map of PM_2.5 Air Pollution and Risk for Tourist Attractions in Hsinchu County, Taiwan

Abstract

In the past few years, human health risks caused by fine particulate matters (PM_2.5) and other air pollutants have gradually received attention. According to the Disaster Prevention and Protection Act of Taiwan’s Government enforced in 2017, “suspended particulate matter” has officially been acknowledged as a disaster-causing hazard. The long-term exposure to high concentrations of air pollutants negatively affects the health of citizens. Therefore, the precise determination of the spatial long-term distribution of hazardous high-level air pollutants can help protect the health and safety of residents. The analysis of spatial information of disaster potentials is an important measure for assessing the risks of possible hazards. However, the spatial disaster-potential characteristics of air pollution have not been comprehensively studied. In addition, the development of air pollution potential maps of various regions would provide valuable information. In this study, Hsinchu County was chosen as an example. In the spatial data analysis, historical PM_2.5 concentration data from the Taiwan Environmental Protection Administration (TWEPA) were used to analyze and estimate spatially the air pollution risk potential of PM_2.5 in Hsinchu based on a geographic information system (GIS)-based radial basis function (RBF) spatial interpolation method. The probability that PM_2.5 concentrations exceed a standard value was analyzed with the exceedance probability method; in addition, the air pollution risk levels of tourist attractions in Hsinchu County were determined. The results show that the air pollution risk levels of the different seasons are quite different. The most severe air pollution levels usually occur in spring and winter, whereas summer exhibits the best air quality. Xinfeng and Hukou Townships have the highest potential for air pollution episodes in Hsinchu County (approximately 18%). Hukou Old Street, which is one of the most important tourist attractions, has a relatively high air pollution risk. The analysis results of this study can be directly applied to other countries worldwide to provide references for tourists, tourism resource management, and air quality management; in addition, the results provide important information on the long-term health risks for local residents in the study area.

1. Introduction

Air pollution is a topic of concern worldwide; it affects the atmospheric and ecological environment and poses a serious threat to the health of humans. Owing to the rapid development of the modern industrial society, global climate change, and increasing environmental awareness of people, air pollution has received increasing attention. According to the Disaster Prevention and Protection Act of Taiwan’s Government enforced on 22 November 2017, “suspended particulate matter” has officially been acknowledged as a disaster-causing hazard.

Haze is caused by extremely small dry particles in the air, which impair visibility. Suspended particulate matter can be classified according to the particle diameter. Particles with sizes of less than 10 μm are PM₁₀, and those with sizes of less than 2.5 μm are PM_2.5. The different particle sizes have different effects on the human body; PM_2.5 is smaller than PM₁₀ and can therefore penetrate the human cilia and mucus, reach the bronchi and alveoli and then the walls of the bronchioles, and finally interfere with the gas exchange in the lungs. In addition, PM_2.5 is more easily suspend in air, does not settle easily, and interacts with other air pollutants [1,2]. Once inhaled by a human, PM_2.5 can reach the depth of the lungs and even penetrate the alveoli and enter the cardiovascular system. As blood circulates throughout the entire body, the harm to human health and ecology is more severe than that from other suspended particulate matter [3,4,5,6]. Many researchers have further pointed out that airborne fine particulate matter can directly or indirectly lead to chronic respiratory diseases, cardiovascular diseases, cancer, neurotoxicity, and even dementia diseases [7,8,9,10,11]. In addition, the long-term exposure to high-concentrations of air pollutants is even more harmful [12,13,14]. Therefore, analyzing the long-term spatial distributions of air pollution hazards (particularly PM_2.5) will provide valuable information for protecting the health and safety of residents.

Over the past few years, China has repeatedly experienced extremely hazardous PM_2.5 concentrations [15,16]. For example, on 19 October 2016, 11 provinces in China were severely affected by air pollutants. Moreover, many cities in western Taiwan are affected by both transboundary and local pollutants, and their air quality is very poor. As Taiwan’s geographical location is close to the southeast of China, it is also the main route for the cold high pressure traveling from China in winter; the transboundary pollutants from China may affect Taiwan’s air quality and the atmospheric circulation. In addition, local or regional sources of pollutants, such as transportation vehicles and factories, produce airborne particulate matter [17,18,19].

The spatial analysis of disaster potentials is a very important part of risk assessments. As assessing the risk of hazards is crucial for a timely evacuation, the analysis of disaster potentials has become very common. The Water Resources Agency of the Ministry of Economic Affairs of Taiwan and researchers have published and applied several generations of flood potential maps for many years [20,21]. In addition, in the Taiwan Central Geological Survey, researchers developed and applied soil liquefaction potential maps [22,23,24]. However, potential disasters caused by suspended particulate matter and the spatial characteristics of air pollution in the past have not been comprehensively investigated; i.e., no potential map of PM_2.5 has been drawn before. In addition, a pollution potential map of various regions would provide valuable information.

Tourist attractions are important gathering places for people, particularly on holidays. Most visitors wish to relax and expect high air quality. Many researchers have studied the relationship between areas of interest and air quality; in particular, they have investigated the integration of low-cost air quality monitoring Internet of Things systems and air quality big data models [25,26,27,28,29,30]. Over the past few years, some Chinese researchers have analyzed the air pollution characteristics of certain specific tourist attractions [31,32]. However, the relationship between the overall tourist attractions and air quality has not been studied.

Hsinchu County in Taiwan has diversified industry, with equal emphasis on agriculture, industry, technology, businesses, and leisure tourism. In addition, Hsinchu County is adjacent to Hsinchu City and Hsinchu Science Park. The population and industry are developing rapidly, and large numbers of people enter Hsinchu County’s major tourism and recreation areas every holiday season. Therefore, high air quality around tourist attractions is very important. A previous study of the characteristics of air pollutants in Hsinchu has shown that the PM_2.5, total PAHs (Polycyclic Aromatic Hydrocarbons), and BaPeq (benzo(a)pyrene equivalent) mass concentrations during the seasons had the following order: winter > autumn > spring > summer with significant seasonal variations [33]. Some early studies focused on the impacts of the large and dense high-tech industries in the Hsinchu Science Park on health and the environment [34,35,36]; in addition, the researchers considered the emissions of toxic compounds such as VOCs (Volatile Organic Compounds) and arsenical emissions; however, there have been few relevant studies in the past decade.

Therefore, the objective of this study was to investigate the exposure risks of tourist attractions based on the potential map of PM_2.5 calculated by the exceedance probability and spatial estimation methods. In this study, the Hsinchu County area was taken as an example. Historical data of PM_2.5 concentrations from the Taiwan Environmental Protection Administration (TWEPA) were used to analyze spatially the air pollution hazard potential of PM_2.5 concentrations in Hsinchu County based on geographic information system (GIS) statistics. The potential threat of PM_2.5 concentrations exceeding a certain standard was spatially investigated with the exceedance probability method; furthermore, the air pollution risk levels of areas with tourist attractions in Hsinchu County were determined. The analysis results of this study can be directly applied to other countries worldwide; they provide references for tourists, tourism resource management, and air quality management, and important information on the long-term health risks for local residents in the study area.

2. Materials and Methods

2.1. Study Area

The terrain of Hsinchu County is mainly composed of flat land, hills, and mountains. There are 13 administrative districts (towns and cities), and its development industries are diverse; they can be mainly classified into agriculture, industry (including science and high technology), commerce, and leisure tourism (Figure 1). Zhubei City is an important town in terms of commerce, economics, and politics; its industries develop high-tech electronics (such as in the Taiyuan Science and Technology Park). The inflow of industrial capital into Zhudong town comprises real estate capital and high-tech manufacturing capital. Like those of Zhubei City, its industries comprise mainly commerce and industry (such as the Industrial Technology Research Institute). The Hukou and Baoshan Townships have the most developed industries and are the production bases for the technology and manufacturing industries, such as the Hsinchu Industrial Park of the Industrial Development Bureau, the Ministry of Economic Affairs in Hukou Township, and high-tech companies such as Taiwan Semiconductor Manufacturing Company and other high-tech factories in Baoshan Township. Baoshan Reservoir and Baoshan Second Reservoir are important water resources for the Hsinchu Science Park. Moreover, the Emei and Wufeng Townships focus on agriculture (tea, oranges, peaches, and sweet persimmons). The Xinfeng, Xinpu, Qionglin, and Beipu Townships exhibit agricultural activities and the establishment of regional industrial zones for the industrial development. Guanxi town, Jianshi Township, and Hengshan Township focus mainly on agriculture and the development of tourism and leisure industries (such as Guanxi Grass, Neiwan Old Street, orchard sightseeing, and visits to the Taiwanese aboriginal people).

Figure 1. Main industry types of 13 townships and cities in Hsinchu County.

To minimize the impacts of disasters, the disaster characteristics of local key industries are studied based on disaster potential data. The results should be sent to the local governmental agencies and key industries (such as the industrial and agricultural management units) in Hsinchu County as an important reference for disaster prevention. More importantly, improvements in areas with higher risks should be prioritized. The main disaster types faced in Hsinchu County can be roughly distinguished according to the topography. The administrative areas on flat land, such as the Zhudong, Hukou, and Xinpu Townships, may experience floods and droughts, and the mountainous administrative areas, such as the Jianshi and Wufeng Townships, can predominantly suffer from landslides or mudflows; in addition, the area close to the sea may face tsunamis. Owing to the development of industrial areas, the Hukou, Baoshan, and Qionglin Townships may suffer man-made disasters caused by toxic chemicals and air pollutants. The industrial characteristics and major and minor risks in the 13 towns and cities in Hsinchu County are summarized in Table 1.

Table 1. Main industrial characteristics and potential disaster risks of 13 townships and cities in Hsinchu County.

District	Industry Type	Major Risk	Minor Risk	Other Risks
Zhubei City	Commerce and industry	Floods and droughts	Toxic chemicals	Tsunamis
Zhudong Township	Commerce and industry	Floods and droughts	Landslides/mudflows
Hukou Township	Industry	Floods and droughts	Toxic chemicals	Air pollution
Baoshan Township	Industry	Landslides/mudflows	Toxic chemicals	Floods and droughts
Emei Township	Agriculture	Landslides/mudflows	Floods and droughts
Wufeng Township	Agriculture	Landslides/mudflows	Floods and droughts
Xinfeng Township	Agriculture and industry	Floods and droughts	Toxic chemicals	Air pollution or tsunamis
Qionglin Township	Agriculture and industry	Floods and droughts	Landslides/mudflows	Toxic chemicals
Beipu Township	Agriculture and industry	Landslides/mudflows	Floods and droughts
Xinpu Township	Agriculture and industry	Floods and droughts	Landslides/mudflows	Toxic chemicals
Guanxi Township	Agriculture and leisure tourism	Floods and droughts	Landslides/mudflows
Jianshi Township	Agriculture and leisure tourism	Landslides/mudflows	Floods and droughts
Hengshan Township	Agriculture and leisure tourism	Landslides/mudflows	Floods and droughts

2.2. Framework of Risk Analysis

The potential refers to the frequency or probability of the occurrence of disasters in an area; the determined potential can be used as a reference for future risk assessments. In this study, the hourly data of PM_2.5 concentrations measured by the TWEPA in Taiwan in 2017 were used, and spatial interpolation was applied to estimate the hourly PM_2.5 concentration of each grid point in the county. Subsequently, the probability that the PM_2.5 concentration of each grid point exceeds the standard value statistically was calculated. The air quality index that corresponds to the unhealthy PM_2.5 concentration for sensitive groups (35.4 μg/m³) was used as the concentration standard. This probability can be represented based on the exceedance probability of older data, which represents the spatial distribution of the potential of PM_2.5. The analysis process is shown in Figure 2.

Figure 2. Flow chart of the construction of an air pollution potential map and risk analysis.

After determining the spatial distribution of the air pollution potentials, the air pollution risk levels in various tourist areas in Hsinchu County were examined. As shown in Figure 2, the PM_2.5 concentrations are based on data from the TWEPA’s Taiwan-wide air quality-monitoring stations from 2017; in addition, radial basis function (RBF) spatial interpolation was used to estimate the grid-like PM_2.5 concentrations in the Hsinchu County area, and the exceedance probability method was applied to calculate the probability that the PM_2.5 concentration of each grid point exceeds the standard. Finally, the potential air pollution risks in the areas of the major tourist attractions in Hsinchu County were examined. The PM_2.5 concentration standard used in this study is based on the air quality index, which considers six levels: good, normal, unhealthy for sensitive groups, unhealthy for all people, very unhealthy, and hazardous. When the “unhealthy for sensitive groups” degree has been reached, it is generally recommended that residents reduce outdoor activities and prolonged vigorous exercise. Therefore, the PM_2.5 concentration standard (35.4 μg/m³) corresponding to the “unhealthy air quality for sensitive groups” degree was used as the threshold. In this study, the analysis results of the probability that the PM_2.5 concentration exceeds the standard were classified into eight levels. In addition, most areas of the Jianshi and Wufeng Townships are too far from the TWEPA’s air quality monitoring station (15 km from the monitoring station) and mainly in high mountainous terrain; thus, they were not included in the calculations.

2.3. Data Collection

First, the hourly PM_2.5 concentrations collected by 76 air quality-monitoring stations of the TWEPA in Taiwan in 2017 were collected. The 327 datasets from areas with tourist attractions originate from the official open data website of the Hsinchu County Government, which were collected in 2019 (https://www.hsinchu.gov.tw/OpenDataDetail.aspx?n=902&s=272).

2.4. Spatial Analysis of Data

The PM_2.5 concentrations throughout Taiwan were estimated with the data from the monitoring stations and RBF spatial interpolation method [37,38,39]. The RBF interpolation is one of the most precise interpolation methods. The interpolation function must pass through the observation value of each station and generate a smooth surface. RBF interpolation is a mesh-free method, constructing high-order accurate interpolants of unstructured data. It takes the form of a weighted sum of radial basis functions. In addition, the RBF interpolation method uses a symmetric function centered at each observation point and calculates the change in the distance from the observation point to obtain the weight of each function:

$$\left[\begin{array}{ccc}\phi (\parallel x_0-x_0\parallel )& \cdots & \phi (\parallel x_n-x_0\parallel )\\ ⋮& \ddots & ⋮\\ \phi (\parallel x_0-x_n\parallel )& \cdots & \phi \left(\parallel x_n-x_n\parallel \right)\end{array}\right]\left[\begin{array}{c}{w}_0\\ ⋮\\ w_n\end{array}\right]=\left[\begin{array}{c}f\left(x_0\right)\\ ⋮\\ f\left(x_n\right)\end{array}\right]$$

(1)

where $\phi$ is a centrosymmetric function and $w_n$ the weight of each function; the interpolation function $f\left(x_n\right)$ can be obtained by solving the equations.

The RBF interpolation method has a good effect on flat surfaces (for concentration diffusion, for instance). In this study, Taiwan was divided into approximately 32,000 grid points, the hourly PM_2.5 concentration of each grid point was estimated with the RBF interpolation method, and the probabilities that the grid points exceed the concentration standard were determined; and finally, we cut and selected the study area of Hsinchu County; ESRI ArcGIS was used to calculate and draw the exceedance probability map. In this study, the exceedance probability of the hourly air pollution concentration was defined as the probability that the hourly data (of an entire year) exceed a certain concentration standard. The air quality index that corresponds to the unhealthy PM_2.5 concentration for sensitive groups (35.4 μg/m³) was used as the concentration standard:

$${\mathrm{P}}_E=\frac{N_E}{N_{all}}$$

(2)

where ${\mathrm{P}}_E$ is the exceedance probability, $N_E$ the number of times in which the hourly data exceed a certain concentration standard in one year, and $N_{all}$ the total number of hourly data of one year. The research data were analyzed with Python and ESRI ArcGIS.

3. Results

3.1. Analysis of Air Pollution Potential

The PM_2.5 concentration is greatly affected by meteorological factors; therefore, the data were investigated according to the different seasons (spring: March–May; summer: June–August; autumn: September–November; winter: December–February). The results are shown in Figure 3. The gray area is too far from the air quality station and was therefore excluded. The analysis results show that the pollution potential in spring (Figure 3a) and winter (Figure 3d) is higher; the probability that the standard concentration in all towns and cities is exceeded is 9.5%, particularly in spring when the Xinfeng and Hukou Townships have probabilities of more than 18%; the probability decreases from the northwest plain area to the southeast mountainous area. The pollution potential in summer and autumn is relatively low; the probability that the standard is exceeded in autumn is generally only approximately 5%. The potential in the northern area of Hsinchu County adjacent to Taoyuan City is higher. In summer, the probability does not exceed 1%, and the probability of pollution in the area near Zhudong Station is slightly higher. Figure 4 and Table 2 show the detailed boxplots and basic statistics of the exceedance probabilities of the 13 townships and cities in Hsinchu County, respectively.

Figure 3. Distribution of PM_2.5 potential in the study area in Hsinchu County in different seasons. Overall, 76 air quality-monitoring stations of the Taiwan Environmental Protection Administration (TWEPA) across the whole of Taiwan were used for spatial estimation, and we extracted the region of Hsinchu County for further analysis. (a) PM2.5 potential map in spring. (b) PM2.5 potential map in summer. (c) PM2.5 potential map in fall. (d) PM2.5 potential map in winter.

Figure 4. Boxplot of exceedance probabilities of 13 townships and cities in Hsinchu County.

Table 2. Basic statistics of exceedance probabilities of 13 townships and cities in Hsinchu County.

District	Entire Year	Spring	Summer	Fall	Winter
	Mean ± Standard Deviation (%)
Zhubei City	7.6 ± 0.4	12.1 ± 0.8	0.1 ± 0.1	4.9 ± 0.6	13.2 ± 0.6
Zhudong Township	6.8 ± 0.3	11.4 ± 0.7	0.5 ± 0.3	4.4 ± 0.3	11.1 ± 0.6
Hukou Township	8.9 ± 0.6	16.8 ± 1.7	0.3 ± 0.1	5.8 ± 0.5	12.9 ± 0.3
Baoshan Township	6.5 ± 0.2	10.2 ± 0.4	0.2 ± 0.1	4.2 ± 0.3	11.7 ± 0.5
Emei Township	6.4 ± 0.1	9.9 ± 0.2	0.3 ± 0.1	4.5 ± 0.2	10.9 ± 0.4
Wufeng Township	5.9 ± 0.2	9.6 ± 0.4	0.4 ± 0.1	4.2 ± 0.2	9.5 ± 0.3
Xinfeng Township	9.0 ± 0.4	16.6 ± 1.5	0.2 ± 0.1	6.3 ± 0.4	13.1 ± 0.3
Qionglin Township	7.2 ± 0.1	12.1 ± 0.3	0.7 ± 0.2	4.6 ± 0.2	11.6 ± 0.3
Beipu Township	6.5 ± 0.3	10.6 ± 0.6	0.5 ± 0.1	4.5 ± 0.1	10.5 ± 0.4
Xinpu Township	7.7 ± 0.2	13.2 ± 0.7	0.3 ± 0.1	4.8 ± 0.4	12.5 ± 0.2
Guanxi Township	7.3 ± 0.4	11.8 ± 0.8	0.7 ± 0.1	5.0 ± 0.2	11.8 ± 0.5
Jianshi Township	6.2 ± 0.2	9.8 ± 0.4	0.5 ± 0.1	4.4 ± 0.3	10.1 ± 0.3
Hengshan Township	6.8 ± 0.2	11.2 ± 0.5	0.8 ± 0.1	4.6 ± 0.1	10.8 ± 0.3

3.2. Risk Analysis of Areas with Tourist Attractions

The spatial distribution map of the PM_2.5 potential was overlaid on a map of the various tourist areas in Hsinchu County; the most severe spring PM_2.5 potential was chosen, as shown in Figure 5, Table 3 and Table 4. The results show that the probability that the standard is exceeded is greater than 18%; the areas with the most severe air pollution potential level (level 6) have three important tourist attractions: the Caixiang Trail, Xiansheng Temple, and Hukou Armored New Village (Village B). The areas of level 5 (16% to 18% chance of exceeding the standard) and level 4 (14% to 16% chance of exceeding the standard) potential—slightly higher potential—have 11 and 7 tourist attractions, respectively. The 11 tourist attractions with level 5 potential are Rongyuanpu Farm, Laohukou Catholic Church Cultural Center, Renhe Trail, Yao Art Street and Bicycle Taro, Hanqing Trail, Hukou Old Street, Xinfeng Sanyuan Temple, Yongning Temple, Chifu Wangye Temple, Hongmaogang Ecological Recreation Area, and Xinfengpuyuan Temple. Another 114 tourist areas are at level 3 (exceeding rates of 12% to 14%), and 124 tourist areas are at level 2 (exceeding rates of 10% to 12%); these locations still exhibit rates greater than 10% in spring (Table A1). These areas encounter a higher risk of air pollution with excessive PM_2.5 concentrations. The highest air pollution potentials of the tourist attractions with levels 5 and 6 in Hsinchu County are shown in Table 4; they are located in the Hukou and Xinfeng Townships. As many tourist areas in Hsinchu County are located in hilly or mountainous areas, they are less exposed to PM_2.5. Only the scenic spots in the Hukou and Xinfeng Townships experience relatively high PM_2.5 concentrations. The detailed PM_2.5 air pollution potential of each tourist attraction in Hsinchu County is shown in Appendix A.

Figure 5. Distribution of PM_2.5 potential in tourist areas in Hsinchu County in spring in 2017.

Table 3. Levels of air pollution potential and numbers of affected tourist attractions.

Level of Air Pollution Potential	Exceedance Probability	Number of Tourist Attractions
0 (mild)	Below 5%	34
1	5% to 10%	34
2	10% to 12%	124
3	12% to 14%	114
4	14% to 16%	7
5	16% to 18%	11
6	18% to 20%	3
7 (severe)	More than 20%	0

Table 4. Highest air pollution potentials of tourist attractions—levels 5 and 6—in Hsinchu County.

Number	Name	District	Longitude	Latitude	Level of Air Pollution Potential
1	Caixiang Trail	Hukou Township	121.02028	24.891221	6
2	Xiansheng Temple	Hukou Township	121.047989	24.902892	6
3	Hukou Armored New Village (Village B)	Hukou Township	121.047808	24.904483	6
4	Rongyuanpu Farm	Hukou Township	121.0442	24.8754	5
5	Laohukou Catholic Church Cultural Center	Hukou Township	121.05516	24.87657	5
6	Renhe Trail	Hukou Township	121.058497	24.877032	5
7	Yao Art Street and Bicycle Taro	Hukou Township	121.0575	24.8773	5
8	Hanqing Trail	Hukou Township	121.05192	24.877399	5
9	Hukou Old Street	Hukou Township	121.052612	24.877742	5
10	Xinfeng Sanyuan Temple	Xinfeng Township	120.9979	24.8999	5
11	Yongning Temple	Xinfeng Township	120.985265	24.90248	5
12	Chifu Wangye Temple	Xinfeng Township	120.9764	24.9102	5
13	Hongmaogang Ecological Recreation Area	Xinfeng Township	120.976365	24.910229	5
14	Xinfengpuyuan Temple	Xinfeng Township	120.977599	24.924916	5

3.3. Analysis of Population Density and Air Pollution Exposure Risk

Moreover, the PM_2.5 potential spatial distribution map was investigated based on the population density of each township in Hsinchu County (Table 5) to analyze the long-term air pollution exposure risks for residents. According to Figure 6, the population density is correlated with the PM_2.5 potential distribution. The Pearson correlation coefficient between the PM_2.5 potential and population density in towns throughout the year is 0.44. If it is explored according to the season, the correlation coefficients between the PM_2.5 potential and population density in spring, summer, autumn, and winter are 0.36, −0.46, 0.34, and 0.64, respectively. Zhubei City (3885.10 persons per square kilometer), Zhudong town (1811.10 persons per square kilometer), Hukou Township (1325.41 persons per square kilometer), and Xinfeng Township (1226.25 persons per square kilometer) have higher population densities than the remaining areas and therefore higher PM_2.5 potentials. A high population density reflects the degree of development and traffic in the city. According to Figure 7, the main industrial areas of Hsinchu County are mostly concentrated in these towns and villages and the main source of pollution. Owing to the prevailing northeast monsoon conditions in winter, these areas have higher pollution risks. Although many tourist attractions are not located in the areas with high air pollution potentials, many residents live in areas with relatively high air pollution potentials for a long time.

Table 5. Population density of each township in Hsinchu County in 2020.

District	Population Density (Persons/km2)
Zhubei City	3885.10
Zhudong Township	1811.10
Hukou Township	1325.41
Baoshan Township	224.58
Emei Township	118.33
Wufeng Township	20.02
Xinfeng Township	1226.25
Qionglin Township	491.86
Beipu Township	185.33
Xinpu Township	462.87
Guanxi Township	230.21
Jianshi Township	18.09
Hengshan Township	196.89

Figure 6. PM_2.5 potential distribution and population density of each township in Hsinchu County.

Figure 7. Map of industrial areas and air quality stations.

4. Discussion

The change in and accumulation, diffusion, and transmission of PM_2.5 concentrations are greatly affected by the meteorological conditions or weather patterns [40,41,42]. The analysis results of the air pollution potentials in Figure 3 are consistent with the general air pollution season in Taiwan (winter and spring). The main reason is that the main prevailing wind in Taiwan in winter and spring is the northeast monsoon; thus, the western half is not affected because of the mountains. The leeward places are likely to experience accumulations of pollutants, particularly central and southwestern Taiwan [41,43,44]. Furthermore, the northeast monsoon tends to bring foreign pollutants from west China into this area [45]. Therefore, the Xinfeng and Hukou areas in Hsinchu County have the highest pollution potentials in winter and spring. In addition, Hsinchu Industrial Park lies in the Xinfeng and Hukou area, and the northern region is close to major stationary pollution sources, such as Taoyuan Youth Industrial Park, Pingjhen Industrial Park, and Yongan Industrial Park (Figure 7). Zhubei City and Hsinchu Science Park in the south are densely populated areas with long-term traffic congestion and are the main sources of mobile pollution in Hsinchu County and Hsinchu City [34,35,36,46]. Both spring and winter are high-pollution seasons, but spring exhibits more evident pollution sources (Figure 3).

In order to further compare the PM_2.5 potential distribution in different years, in addition to Figure 5 showing 2017, Figure 8 shows the dynamic distributions of PM_2.5 potential in tourist areas in Hsinchu County in spring in 2018 and 2019. They show spatial distributions similar to 2017, and Xinfeng and Hukou also have the highest potential. However, it is obvious that the overall probability of PM_2.5 exceeding the standard has been declining in the entire region in recent years. In addition to the influences of meteorological conditions in different years, it may be due to the implementation of government policies and the increase in people’s awareness of environmental protection.

Figure 8. Dynamic distribution of PM_2.5 potential in tourist areas in Hsinchu County in spring in (a) 2018 (b) 2019.

Moreover, Xinpu, Guanxi, Qionglin, Baoshan, Emei, and Beipu are dominated by hilly land; this less densely populated area exhibits agricultural, industrial, and touristic activities; thus, the air quality is evidently better than in other areas in all seasons. The Hengshan, Jianshi, and Wufeng Townships have mostly mountainous terrain, and the populations are sparser; consequently, they have the best air quality. In addition, because the west side of Hsinchu is adjacent to the sea and the east side exhibits mostly hilly terrain, the topographical effect is affected by the prevailing wind and major sources of emissions in the air pollution season [47]. Therefore, air pollutants in Hsinchu accumulate easily in the relatively flat plains, such as in Xinfeng and Hukou, which is consistent with the results of this study. Some researchers have investigated the impacts of terrain effects on air pollution [48], particularly the basin effects [49,50]; some researchers have used geostatistical models to estimate the PM_2.5 concentrations [51]. Fortunately, most of the tourist areas in Hsinchu County are located in areas with lower PM_2.5 air pollution potentials, and the areas with higher air pollution potentials are mostly those with industrial and technological activities. Nevertheless, the areas with high pollution potentials have higher population densities. A high population density leads to more emission sources. Some researchers have used the spatial econometric model to investigate the relationship between the population density and air pollution in Chinese cities; they have discovered a significant positive correlation between the population density and PM_2.5 concentration [52,53], which is consistent with the results of this study.

5. Conclusions

In this study, an air pollution potential map was constructed. The results show that the potentials of different seasons are quite different. The most severe air pollution seasons are spring and winter, whereas summer exhibits the best air quality. Xinfeng and Hukou Townships in Hsinchu County have the highest potential (approximately 18%). Hukou Old Street, which is the most famous tourist attraction, has a relatively high pollution risk. The population density is positively correlated with the PM_2.5 potential distribution in most seasons, except for summer. In this study, the hazard potential levels of PM_2.5 concentrations exceeding a certain standard were investigated; the exceedance probability and the air pollution potential levels of various tourist areas in Hsinchu County were examined. However, the information on tourist attractions considered in this research study is limited and based on only few important attractions. The air pollution potential map can be combined with more detailed tourist attraction maps in the future. In addition, the map can be applied to investigate the impacts of pollution on schools, elderly people, hospitals, and nurseries to determine their potential long-term exposure risks. Although the study area in Hsinchu County has only three important tourist attractions with the most severe air pollution potential levels (level 6), there are still many schools and residents in these areas.

In the future, a map for the entire country will be constructed; the proposed framework can be directly applied to other countries worldwide. In addition, the spatial and temporal changes in the air pollution potential during different years can be analyzed, and the air pollution data of one year can be expanded to more than five or ten years. In addition to reducing the possibility of being more extreme in certain years, understanding the temporal changes in the spatial distribution of the pollution potentials is more effective for assessing dynamic risks. In addition to providing a reference for tourists, the results provide information on the long-term health risks for local residents in the study area.