Tropical cyclones (also known as typhoons) are a frequent occurrence in East Asia during summer and fall. Typhoons significantly affect not only meteorological parameters but also air quality. That is because air pollution is strongly related to atmospheric conditions, and typhoon circulation typically alters atmospheric stability and air pollutant diffusion in specific locations. For example, researchers revealed that ozone episodes in Hong Kong and southeastern China are strongly related to the passage of typhoons as they approach the area (Lee et al., 2002; Ding et al., 2004; Huang et al., 2005, 2006; Yang et al., 2012; Zhang et al., 2013, 2014; Wei et al., 2016; Yan et al., 2016; Luo et al., 2018; Deng et al., 2019; Huang et al., 2021). The stagnant meteorological conditions associated with strong subsidence and stable stratification in the boundary layer result in pollutant accumulation before typhoons make landfall. Huang et al. (2005) reported that approximately 30 % of total ozone in Hong Kong was due to local chemical production in the lower atmospheric boundary layer, and approximately 70 % was contributed by long-range transport from southern China (i.e., the Pearl River Delta). According to the dynamic process perspective, Chow et al. (2018) reported frequent high-O3 days when typhoons were located between Hong Kong and Taiwan (Fig. 1a) due to the influence of the typhoon position and associated atmospheric circulations on air quality.

Taiwan also experiences air quality deterioration as typhoons approach (Fang et al., 2009; Chang et al., 2011, Cheng et al., 2014; Hsu and Cheng, 2019). However, not all typhoons are associated with poor air quality in Taiwan. The effect of typhoons on air quality is highly related to the location of the typhoon and its circulation's interaction with Taiwan's Central Mountain Range (CMR; Fig. 1b). Thus, the mechanism of the formation of poor air quality may differ between Taiwan and Hong Kong. Air quality deterioration frequently occurs over the western plains of Taiwan when typhoons pass between Taiwan and Luzon island in the Philippines; the distance of the typhoons from Taiwan is typically several hundred kilometers but may even be greater than 1000 km. Under such conditions, the weather is typically stable, with clear skies, strong solar intensity, and weak wind speeds over Taiwan's western plains because of the interactions of the typhoon's easterly circulations with the CMR. Furthermore, such typhoons are usually associated with a Pacific high-pressure system during summer; thus, the air temperature may be high. For example, researchers have noted that typhoon's secondary circulation may enhance subsidence and result in a heat wave, clear skies, and weak wind speed over Taiwan or southern China (e.g., Ding et al., 2004; Huang et al., 2005; Jiang et al., 2015; Shu et al., 2016; Lam, et al., 2018), thus adversely affecting air quality as well. In Taiwan, this phenomenon is particularly attributed to the blocking effect of the CMR. The CMR occupies approximately two-thirds of Taiwan's landmass (300 km × 100 km) and lies NNE–SSW (Fig. 1b), with an average terrain height of approximately 2000 m (Lin and Chen, 2002; Lin et al., 2011) and some peaks of nearly 4000 m. The CMR has a major effect on local circulation and frequently interferes with the easterly or northeasterly prevailing winds such as long-range transport dust storms and air pollution events (Lin et al., 2004, 2005, 2012a, b) during the winter monsoon. When a typhoon is located between Taiwan and Luzon, the low-level easterly airflow easily splits northern and southern Taiwan and moves around the island, forming a vortex at the lee side of the mountain (Hunt and Synder, 1980; Smolarkiewicz and Rotunno, 1989; Y. L. Lin, 1993; Lin et al., 2007). On the lee side of the CMR, wind speeds are weak (Lin et al., 2007) and the atmospheric conditions are more stable than on the windward side of eastern Taiwan. Under these favorable conditions, air pollutants readily accumulate and result in high ozone and aerosol concentrations over western Taiwan. Actually, the interactions between ambient flow and topography result in stable weather conditions, and air pollutant accumulation in the low boundary is common all over the world. According to the obstacle's scale, it could occur on plateaus (Ning et al., 2019), mountains (Lai and Lin et al., 2020), and even among buildings (W. Theurer, 1999) as the airflow interacts with them. For example, Wallace et al. (2010) investigated the spatial and topographic effects of temperature inversions on air quality in the industrial city of Hamilton, located at the western tip of Lake Ontario, Canada. Topographically constrained wind flows and frequent temperature inversions occurred in Los Angeles, California (Lu and Turco, 1995), the Highveld Plateau industrial region in South Africa (Jury and Tosen, 2004), and Perth, Australia (Pitts and Lyons, 1988). Valverde et al. (2016) studied air pollution in Europe and found that the dispersion and transfer of air pollutants are affected by topographic features and weather patterns. Ning et al. (2019) presented synergistic effects of synoptic weather patterns of a low trough, low vortex, and topography on air quality over the Sichuan Basin of China.

Summer and fall are regarded as the “typhoon season” over Taiwan and throughout East Asia. Statistically, more than 20 typhoons form in the western Pacific Ocean per year, and approximately 3–4 typhoons directly strike Taiwan (Lin et al., 2011; Tu and Chen 2019). Records from Taiwan's Central Weather Bureau (CWB) indicate that 18 % of typhoons (Type 5; https://www.cwb.gov.tw/V8/C/K/Encyclopedia/typhoon/typhoon.pdf, last access: 18 March 2021) between 1911 and 2019 did not make landfall but passed between Taiwan and Luzon. The wind circulations of this type of typhoon were easterly or southeasterly depending on the location of the typhoons. Thus, it is not uncommon for more than 10 typhoons per year to pass near Taiwan and affect the island's air quality. The impact of the interaction between the typhoon's circulation and the CMR on the air quality on the lee side of the mountain is more serious than in other areas.

To date, air pollution episodes with a formation mechanism associated with the interactions between typhoon circulation and the CMR have not been thoroughly documented in Taiwan. In this study, we investigated a major air quality event that occurred on 17 July 2018, with a maximum O3 concentration of 134 ppb and a daily maximum aerosol concentration for PM10 (PM2.5) reaching 152 µg m-3 (70 µg m-3) in inland rural areas of central Taiwan. We used the Weather Research Forecasting with Chemistry model (WRF-Chem version 3.9; Grell et al., 2005) to study the processes and mechanisms of formation of the air pollution episode. The remainder of this paper is organized as follows: Sect. 2 describes the data sources and sampling measurement during the study period; Sect. 3 presents the model and settings used in this study; Sect. 4 presents the air quality characteristics and measurements recorded over the western plains of Taiwan; Sect. 5 describes and discusses the simulation results of air quality associated with the typhoon event using WRF-Chem; and, finally, Sect. 6 provides the conclusions.

We collected measurements of hourly PM10, PM2.5, and other pollutants (O3, NOx, CO, and SO2) as well as meteorological parameters (air temperature, wind field, and rainfall) from Taiwan Environmental Protection Administration (TEPA) air quality monitoring stations. To elucidate the spatial distribution of air pollutants, we classified the observed stations over central Taiwan into “coast,” “urban,” and “mountain.” Each of these categories represents the mean concentration of the numbers derived from stations of the same type. The coast category included two stations: Shalu (SL) and Xianxi (XX; Fig. 1c). The urban category included five stations: Fengyuan (FY), Xitun (XT), Zhongming (ZM), Changhua (CH), and Dali (DL; Fig. 1c). The mountain category included three stations: Nantou (NT), Zhushan (ZS), and Puli (PL), which were located nearby or in basins surrounded by high mountains (Fig. 1c). Two stations on small islands were also considered in the analysis. One was in Kinmen (KM), which is located close to Xiamen city in southeastern China, and the other was Magong (MG) station located in the Taiwan Strait (Fig. 1a).

To explore the air pollution episodes during summer, we recorded data in central Taiwan in July 2017 and 2018. For the summer campaigns, we employed three sampling sites (the squares in Fig. 1c): Shalu (SL, 24.23∘ N, 120.57∘ E; the same location as the TEPA station), Zhushan (ZS, 23.76∘ N, 120.68∘ E; the same location as the TEPA station), and Chung Shan Medical University (CSM) (24.12∘ N, 120.65∘ E; Fig. 1c). ZS is a suburban site located in a complex valley surrounded by hills (300–500 m) and high mountains (CMR; elevation >2000 m) to the east and south, respectively. The remaining two sampling sites, SL and CSM, are located in a coastal suburban and urban area (Fig. 1c), respectively. The sampling period of each sample was 11 h; daytime samples were collected from 08:00 to 19:00 LST, whereas nighttime sampling was conducted from 20:00 to 07:00 LST. We determined mass concentrations of the aerosols using a gravimetric measurement of the samples collected on polytetrafluoroethylene membrane filters (Chou et al., 2008; Lee et al., 2019). The filter samples were analyzed for water-soluble ions (Ca2+, Mg2+, Na+, NH4+, K+, SO42-, NO3-, and Cl-) via ion chromatography (Dionex ICS 1000, Thermo Scientific). Organic carbon (OC) and elemental carbon (EC) were measured by a thermal–optical carbon analyzer (DRI, 2001A, Atmoslytic Inc.), following the IMPROVE thermo-optical reflectance (TOR) protocol (Chow et al., 2001). The instruments for the hourly measurement of PM10 and PM2.5 from TEPA are METONE_BAM1020 (https://airtw.epa.gov.tw/CHT/EnvMonitoring/Central/Tools.aspx, last access: 18 March 2021). Two reactive gases, ozone (O3) and sulfur dioxide (SO2), were measured in parallel with the aerosol measurements. A non-dispersive ultraviolet photometer (ML9810, Ecotech, Australia) and an ultraviolet fluorescence spectrometer (ML9850, Ecotech, Australia) were used to measure O3 and SO2 concentrations, respectively. Sounding data were obtained from the CWB; the site on Penghu island (the World Meteorological Organization – WMO – station number code is 46734) was close to the MG TEPA station (Fig. 1a).

During summer, the land–sea breeze easily combines with the mountain upslope and downslope wind during daytime and nighttime. As the sea breeze develops, airflows are typically transported from coastal areas and pass over the Taichung metropolitan region (Fig. 1c) coupled with mountain slope flow to the inland area. The Taichung metropolis is a large urban environment comprising residential, industrial, and agricultural lands (Cheng et al., 2009). In particular, Taichung Power Plant (TPP, Fig. 1c), which is coal-fired, and the Taichung Harbor industrial area (THI, Fig. 1c) are both located on the coast and are responsible for substantial emissions in central Taiwan. Thus, severe emission sources contribute to and affect the air quality in the Taichung metropolitan area under favorable weather conditions. Meteorological parameters, including wind speed and wind direction, temperature, and relative humidity, were acquired from a meteorological station in the same location where data were collected for this study.

In this study, we used the Weather Research and Forecasting (WRF) model coupled with the WRF-Chem version 3.9 to study air pollutant transport during the episode. We obtained the meteorological initial and boundary conditions for WRF-Chem from the National Center for Environmental Prediction (NCEP) Operational Global Forecast System (GFS) 0.25∘×0.25∘ data sets at 6 h intervals. We selected the Yonsei University (YSU) planetary boundary layer (PBL) scheme for this study. The coarse and fine domains had 259×370 and 301×301 grid nets with resolutions of 9 and 3 km, respectively. The vertical had 41 levels, with the lowest level approximately 40 m above the surface. To ensure that the meteorological fields were well simulated, we employed the four-dimensional data assimilation scheme according to the NCEP GFS data. Transport processes included advection by winds, convection by clouds, and diffusion by turbulent mixing. Removal processes included gravitational settling, surface deposition, and wet deposition (scavenging in convective updrafts and rainout or washout in large-scale precipitation). The kinetic preprocessor (KPP) interface was used in both the chemistry scheme of the Regional Atmospheric Chemistry Mechanism (Stockwell et al., 1990). The secondary organic aerosol formation module, the Modal Aerosol Dynamics Model for Europe (MADE) (Ackermann et al., 1998) volatility basis set (VBS) (Ahmadov et al., 2012), was employed in the WRF-Chem model. The anthropogenic emissions in Taiwan were obtained from the air pollutant monitoring database of the TEPA. Its emission inventory system is called the Taiwan Emission Data System (TEDS). The TEDS version in this study is V9.0 (2013) and contains data on eight primary atmospheric pollutants: CO, NO, NO2, NOx, O3, PM10, PM2.5, and SO2.

The wind direction over Taiwan during summer is mostly southerly to southwesterly (Table 1). However, the wind direction during the episode was westerly to northwesterly (Table 2). The wind direction changed because of the critical interaction between typhoon circulations and the CMR. Moreover, the concentration of PM2.5 and its composition during the episode also differed significantly from the monthly mean, as revealed in Table 2. A substantial increase in daily mean PM2.5 was observed at all sites, especially at the CSM site (urban), where the concentration increased from 16.9 to 40.5 µg m-3 (Table 2). Furthermore, SO42- became the dominant species in PM2.5. Its concentration increased from the coast (9.4 µg m-3; 30.5 % of PM2.5) to the mountain station (11.6 µg m-3; 29.7 % of PM2.5), and the urban site had the highest concentration (12.1 µgm-3; 29.9 % of PM2.5) in PM2.5 among these stations. The SO42- concentration during the episode (Table 2) was more than twice that of the monthly mean (Table 1) in the Taichung area. This variation was due to the wind direction changing from southwesterly to northwesterly, resulting in a contribution increase from the upstream TPP and THI (Fig. 1c), which are the major sources in central Taiwan.

On 17 July 2018, Taichung city not only experienced high air pollutant concentrations but also a maximum air temperature as high as 35.4 ∘C. That is, a heat wave (Lin et al., 2017; Kueh et al., 2017) occurred on 17 July because of the subsidence of the typhoon circulation on the lee side of the mountain. The daily mean temperatures for the sampling sites between 15 and 17 July for SL, CSM, and ZS were 29.9, 30, and 29.4 ∘C, respectively. However, the monthly mean temperatures (July in 2017 and 2018) during the sampling period for SL, CSM, and ZS were 28.9, 28.8, and 26.5 ∘C, respectively. Thus, the daily mean temperature during the episode was 1–2 ∘C higher than is typical for days in July. In general, the mean wind speed on the episode days at these three sites was weaker (<1 m s-1) than the monthly mean (Tables 1 and 2). Such stable weather conditions, weak wind speed, and high air temperature were conducive to the generation and formation of a secondary aerosol. This is exemplified by the concentrations of other species, such as OC, NO3-, and NH4+, being considerably higher during the episode days (Table 2) compared with the monthly mean in Table 1. Notably, EC increased to a lesser extent than the other species. These results suggest that secondary aerosol plays a critical role under such stable weather conditions and wind direction. Because ambient wind changes during typhoon formation between Taiwan and Luzon island in the Philippines are not uncommon, the air quality impacts in such weather conditions merit further research. A detailed discussion of variations in aerosol chemical composition transformation will be presented in a separate paper.

Summer was the season with the lowest air pollution levels in Taiwan during 2004–2019. The monthly mean concentrations of PM10, PM2.5, and daytime ozone (08:00–17:00 LST) in summer (June–August) over central Taiwan are 35–40 µg m-3, 18–22 µg m-3, and 30–42 ppb, respectively. The contribution of OC to PM2.5 can exceed 30 % in urban and inland mountain sites. However, episodes of poor air quality frequently occur over the western plains when an easterly typhoon circulation interacts with the complex topographic structure in Taiwan. Under such a weather condition, concentrations of PM2.5 and ozone could be 2 times higher than the monthly mean. During the episode in this study, SO42- became the major contributor to PM2.5, and its concentrations and contribution proportions (%) in PM2.5 at coastal, urban, and mountain sites were 9.4 µg m-3 (30.5 %), 12.1 µg m-3 (29.9 %), and 11.6 µgm-3 (29.7 %), respectively.

To explore the mechanism of air pollution formation, we conducted a detailed data analysis and WRF-Chem model simulation of an episode of poor air quality associated with a typhoon event between 15 and 17 July 2018. Numerical modeling indicated that not only wind direction changes due to lee vortex but also land–sea breeze and boundary layer development were the key mechanisms in the transport of air pollutants. We summarize the key mechanisms and processes of the interaction between typhoon circulation, lee vortex, land–sea breeze, boundary layer development, and topography as well as their effects on air quality in Fig. 12.

First, typhoon circulations provided a strong easterly ambient flow. This easterly flow interacted with the CMR, resulting in a lee vortex formation over western Taiwan (Fig. 12, left panel).